[From the U.S. Government Printing Office, www.gpo.gov]
ECOLOGICAL STUDY OF THE AMOCO CADIZ OIL SPILL Report of the NOAA-CNEXO Joint Scientific Commission % 7707 1$: -71 7 @Iw 0', U. S DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration CENTRE NATIONAL POUR I'EXPLOITATION DES OCEANS ECOLOGICAL STUDY OF THE AMOCO CADIZ OIL SPILL Report of the NOAA-CNEXO Joint Scientific Commission October 1982 2qf2qt4qo8q"6qr6qt4qy 0q02q9 8qe4qs4qe 2qs4q08qs8qa0qw q028q228q@ U. S. DEPARTMENT OF COMMERCE A Malcolm Baldrige, Secretary National Oceanic and Atmospheric Administration John V. Byrne, Administrator 24qw CENTRE NATIONAL POUR 2qIq'EXPLOITATION DES OCEANS 88qM88qW US Department of Commerce NOAA Coastal Services center Library 2234 South Hobson Avenue Charleston, SC 29405q-2413 DISCLAIMER Mention of a commercial company or product does not constitute an endorsement by NOAA Environmental Research Laboratories. Use for publicity or advertising purposes of information from this publi- cation concerning proprietary products or the tests o* such products is not authorized. TABLE OF CONTENTS Pa_ge Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . v I. Physical, Chemical, and Microbiological Studies After the AMOCO CADIZ Oil Spill ATLAS, R.M. Microbial Degradation within Sediment Impacted by the AMOCO CADIZ Oil Spill . . . . . . . . . . . . . . 1 BALLERINI, D., DUCREUX, J., and RIVIERE, J. Laboratory Simulation of the Microbiological Degradation of Crude Oil in a Marine Environment 27 BOEHM, P.D. The AMOCO CADIZ Analytical Chemistry Program . . . . 35 DOU, H., GIUST, G., and MILLE, G. Studies of Hydrocarbon Concentrations at the Ile Grande and Baie de Lannion Stations Polluted by the Wreck of the AMOCO CADIZ . . . . . . . . . . . 101 DUCREUX, J. Evolution of the Hydrocarbons Present in the Sediments in the Aber Wrac'h Estuary . . . . . . . . MARCHAND, M., BODENNEC, G., CAPRAIS, J.-C., and PIGNET, P. The AMOCO CADIZ Oil Spill, Distribution and Evolution of Oil Pollution in Marine Sediments . . . 143 WARD, D.M., WINFREY, M.R., BECK, E., and BOEHM, P. AMOCO CADIZ Pollutants in Anaerobic Sediments: Fate and Effects on Anaerobic Processes . . . . . . . 159 II. Biological Studies After the AMOCO CADIZ SPILL GLEMAREC, M. and HUSSENOT, E. Reponses des Peuplements Subtidaux a la Perturbation Creee par VAMOCO CADIZ dans les Abers Benoit et Wrac'h . . . . . . . . . . . . . . . . . . . . . . . 191 CABIOCH, L., DAUVIN, J.-C., RETIERE, C., RIVAIN, V. and ARCHAMBAULT, D. Les Effets des Hydrocarbures de VAMOCO-CADIZ sur les Peuplements Benthiques des Baies de Morlaix et de Lannion d'Avril 1978 a Mars 1981 . . . . . . . . . 205 Page BOUCHER, G., CHAMROUX, S., LE BORGNE, L., and MEVEL, G. Etude Experimentale d'une Pollution par Hydrocarbures dans un Microecosysteme Sedimentaire. 1: Effet de la Contamination du Sediment sur la fleiofaune . . . . . 229 BODIN, P. and BOUCHER, D. Evolution a Moyen-Terme du Meiobenthos et du Microphytobenthos sur Quelques Plages Touchees par la Maree Noire de l'AMOCO-CADIZ . . . . . . . . . . . . 245 NEFF, J.M. and HAENSLY, W.E. Long-Term Impact of the AMOCO CADIZ Crude Oil Spill on Oysters Crassostrea gigas and Plaice Pleuronectes platessa From Aber Benoit and Aber Wrac'h, Brittany, France. I. Oyster Histopathology. II. Petroleum Contamination and Biochemical Indices of Stress in Oysters and Plaice . . . . . . . . 269 LEVASSEUR, J.E. and JORY, M.-L. Retablissement Naturel d'une Vegetation de Marais Maritimes Alteree par les Hydrocarbures de l'AMOCO- CADIZ: Modalites et Tendances . . . . . . . . . . . . 329 SENECA, E.D. and BROOME, S.W. Restoration of Marsh Vegetation Impacted by the AMOCO CADIZ Oil Spill'and Subsequent Cleanup Operations at Ile Grande, France . . . . . . . . . . . 363 LE CAMPION-ALSUMARD, T., PLANTE-CUNY, M.-R., and VACELET, E. Etudes Microbiologiques et Microphytiques dans les Sediments des Marais Maritimes de l'Ile Grande a la Suite de la Pollution par 1'AMOCO CADIZ . . . . . 421 CHASSE, C. and GUENOLE-BOUDER, A. 1964-1982, Comparaison Quantitative des Populations Benthiques des Plages de St Efflam, St Michel-en- Greve Avant, Pendant et Depuis le Naufrage de 1'AMOCO-CADIZ . . . . . . . . . . . . . . . . . . . . . 451 iv PREFACE At approximately 11:30 p.m. on Thursday, March 16, 1978, the super- tanker Amoco Cadiz went aground on a rock outcropping 1.5 km offshore of Portsall on the northwest coast of France. The vessel contained a cargo of 216,000 tons of crude oil and 4,000 tons of bunker fuel. At 6:00 a.m. on Friday, March 17, the vessel broke just forward of the wheelhouse and thus started the largest oil spill in maritime history. During the course of the next 15 days, the bunker fuel and contents of all 13 loaded cargo tanks, which contained two varieties of light mideastern crude oil, were released into the ocean. The oil quickly became a water-in-oil emulsion (mousse) of at least 50% water, and heavily impacted nearly 140km of the Brittany coast from Portsall to Ile de Brehat. At one time or another, oil contamination was observed along 393 km of coastline and at least 60 km offshore. Impacted areas included recreational beaches, mariculture impoundments, and a substantial marine fishery industry. On March 18, Dr. Wilmot N. Hess, Director of the Environmental Research Laboratories (ERL) of the National Oceanic and Atmospheric Administration (NOAA), contacted Dr. Lucien Laubier, Director of the Centre Oceanologique de Bretagne (COB) of the Centre National pour l'Exploitation des Oceans (CNEXO), the French national oceanographic organization. Dr. Hess and Dr. Laubier arranged for participation by United States scientists in a joint Franco-American investigation of physical and chemical manifestations of the spill. On March 24, the agreement was expanded to include cooperative biological investigations through contacts initiated by Dr. Eric Schneider, Director of the Environmental Protection Agency's Environmental Research Laboratory in Narragansett, Rhode Island. NOAA personnel arrived on March 19 to join the investigation initiated on March 17 by several French scientific teams. Initial photographic over-flights and active beach sampling began on Tuesday, March 21, followed by initial chemical sampling by vessel on Friday, March 24. The team was supplemented with EPA biological observers on Sunday, March 26. Sampling has continued by some segments of this original team until the present time. Throughout the period of investigation, active interaction and coordination with the French scientific community have,taken place under the auspices of COB/CNEXO. All sampling has been coordinated with the general ecological impact study designed by the French Ministry of L_ by CNEXO, and operated by several scientific Environment, organize@/ institutions in France- , making possible a more thorough evaluation of the effects of the incident than would otherwise have been possible. Y National Museum of Natural History, National Geographic Institute, French Institute of Petroleum, Scientific and Technical Institute of Marine Fisheries, University of Western Brittany, University P. and .M. Curie, Paris VI, and the National Center for the Exploitation of the Oceans. V About three months after the oil spill the U.S. team prepared a "Preliminary Scientific Report on the Amoco Cadiz Oil Spill" covering data up to May 15, 1978. This document covered only the period of acute effects. A one-day symposium on the Amoco Cadiz spill was held in Brest on June 7, 1978, and published soon @_fter. It was obvious from these initial observations that a period of years would be required to under- stand what had happened to these portions of the coast where the oil had settled in and not been cleansed promptly. During this early period of study of the spill Mr. Russ Mallatt of the Amoco Trading Company had several discussions with Drs. Hess, Laubier and Schneider. Mr. Mallatt was the General Manager for Environmental Conservation and Toxicology of Amoco. Discussion with Mr. Mallatt during the first two months after the spill identified Amoco's interests in carrying out long-term studies of the effects of the oil spill. These early contacts were followed up by substantial discussions between Mr. John Linsner of Amoco and Mr. Eldon Greenberg, General Counsel of NOAA. These discussions culminated with an agreement being signed by Amoco and NOAA to carry out long-term studies of the effects of the spill. The study would cover three years and would be a joint French-U.S. activity. A Joint NOAA/CNEXO Scientific Commission was established through another agreement between the two agencies signed June 2 , 1978. Amoco would transfer money to NOAA and the Joint Commission, chaired by Drs. Hess and Laubier, would determine the research program to be carried out, the investigators to do the research, and the funding levels. The Joint Commission would also-monitor the progress of the studies and be responsible for making the final report. One of its major goals was to make U.S. and French scientific teams work together in a common effort to better understand the consequences of the wreckage. The Joint Commission first met in Brest at the CNEXO Laboratory on July 18, 1978. Taking i'nto account the French program to assess the long-term ecological impact of the oil spill funded by the Ministry of Environment, it determined that the most important areas for research were: 1. Heavily impacted subtidal areas like the Abers and the Bays of Morlaix and Lannion. 2. Heavily impacted intertidal areas such as St. Efflam and the salt marsh at Ile Grande. 3. The detailed chemical evolution of the petroleum hydrocarbons. 4. Biodegradation of petroleum. The second meeting of the Joint Commission, held in Washington, D.C., on October 12, 1978, reviewed the work carried out during the first months of the first year and planned the research program for the second year's study. Vi In November 1979, an international conference was held in Brest sponsored by CNEXO. Investigators sponsored by the Joint NOAA/CNEXO Scientific Commission, as well as a number of other scientists, gave papers at this conference. The proceedings of this conference entitled "Amoco Cadiz: Fates and Effects of the Oil Spill" make a very good summary of the first one and one-half year study after the spill. Following the second meeting of the Joint Commission, Dr. Hess left NOAA and was replaced as co-chairman by Dr. Joseph W. Angelovic from the Office of Ocean Programs in NOAA. The third meeting of the Joint Commission was held in Paris, France, October 28, 1980, in conjunction with the meeting of the U.S.-French Cooperative Program in Oceanography. The previous work was reviewed and the final year of the research program was planned. Now the three-year study is over and attempts are being made to bring together the findings of the investigators. A workshop was held in Charleston, South Carolina, on September 17-18, 1981, to report on the physical and chemical studies. A second workshop was held in Brest, France, on October 28-30, 1981, to report on the biological effects studies. This document is the report of those workshops and forms the body of the final report to Amoco from the Joint NOAA/CNEXO Scientific Commission. Speaking for all who worked on the spill, we would like to thank the Amoco Transport Company for sponsoring this three-year study of the effects of the spill. Without Amoco's help, we would be nowhere near our present state of knowledge of what the effects of the spill were or how the recovery back to normal conditions has proceeded. Other studies have been carried out, sponsored by the French Government and other sources, but an important part of the work has been sponsored by Amoco. .Mr. Ru ss Mallatt, Dr. James Marum, Mr. John Lamping, Ms. Carol Cummings and others from Amoco attended meetings of the Joint Commission and the scientific sessions. They were always helpful and supportive of the Commission's work and never intruded on the design or conduct of the programc Vie have, through this cooperative effort, obtained more detailed and more useful knowledge of the effects of this oil spill than of any other large oil spill in history. A major reason for this is that the . biological communities present before the spill had been studied in great detail by French scientists. Today many of the areas impacted by the spill appear to the casual observer to be recovered from the effects of the oil. However, investi- gations have shown that differences still exist between some of the current ecosystems and those present prior to the spill. Hopefully other studies will continue to watch and document the recovery processes. These studies have added substantially to man's knowledge about oil spills. We can only hope that others will follow and build on the understanding of oil spill effects accumulated through these studies. Lucien Laubier Wilmot Hess Joseph Angelovic Viii CNEXO-NOAA Joint Scientific Commission MEMBERSHIP L. Laubier, Cochairman Centre National pour l'Exploitation des Oceans Paris, FRANCE Wilmot N. Hess, Cochairman National Oceanic & Atmospheric Administration Boulder, Colorado Joseph W. Angelovic, Cochairman NOAA Office of Ocean Programs Rockville, Maryland Jack Anderson Battelle Pacific Northwest Laboratory Sequim, Washington J. Bergerard Station Biologique Roscoff, FRANCE Edward S. Gilfillan Bowdoin College Bowdoin, Maine 1. R. Kaplan University of California, Los Angeles Los Angeles, California R. Letaconnoux Institut des Peches Maritimes Nantes, FRANCE J. M. Peres Station Marine and Endoume Marseille, FRANCE Philippe Renault Institut Francais du Petrole Rueil Malmaison, FRANCE Douglas A. Wolfe NOAA Office of Marine Pollution Assessment Boulder, Colorado ix PART I Physical, Chemical, and Microbiological Studies After the AMOCO CADIZ Oil Spill Edited by E. R. Gundlach Research Planning Institute, Inc. Columbia, South Carolina, U.S.A. 29201 MICROBIAL HYDROCARBON DEGRADATION WITHIN SEDIMENT IMPACTED BY THE AMOCO CADIZ OIL SPILL by Ronald M. Atlas Department of Biology University of Louisville Louisville, Kentucky 40292 INTRODUCTION The wreck of the AMOCO CADIZ in March 1978 released over 210,000 tons of oil into the marine environment. As much as one third of the spilt oil may have been washed into the intertidal zone. The spill occurred during storm surges, thereby spreading the oil throughout the intertidal zone. Two years after the AMOCO spill, the wreck of the tanker TANIO resulted in another oil spill that contaminated much of the same Brittany shoreline impacted by the AMOCO CADIZ. This study was undertaken to determine the fate of petroleum hydrocarbons within surface sediments along the Brittany coast with reference to the role of microorganisms in the oil weathering process. METHODS Sampling Regime Duplicate samples. were collected at intertidal sites along the Brittany coast which had received varying degrees of oiling from the AMOCO CADIZ spillage (Fig. 1). The sampling sites included the salt marsh at Ile Grande, a beach near Portsall in the vicinity of the wreck site, a mudflat in Aber Wrac'h, a beach at St-Michel-en-Gr@ve near where. a large bivalve kill had been reported, a relatively lightly oiled reference site at Trez Hir and a site at Tregastel which was not oiled by the AMOCO CADIZ spill, but was later oiled by the spill f rom the tanker TANIO (Table 1) . Surface sediment samples (upper 5 cm) were collected with a 3 cm diameter soil corer. Samples were placed in metal cans for hydrocarbon analyses and in Whirlpak bags for microbial analyses. Samples were collected during December, 1978; March, 1979; August, 1979, November, 1979, March 1980, July, 1980 and June, 1981; 9, 12, 17, 20, 24, 28 and 39 months after the spillage, respectively. During November, 1979 sediment samples were also collected at four offshore sites in the Bay of Morlaix. Samples for microbiological analyses were processed within four hours of collection. For hydrocarbon analyses, samples were frozen and shipped to Energy Resources Company (ERCO) for extraction and analysis FRANCE ILE GRAND BAY OF OD MORLAIX OC :ST MICHEL-EN-GREVE VABER WRACH (4) (5,6) OORTS (7,8) TREZOV R FIGURE 1. Location of intertidal and subtidal sampling sites. TABLE I - Description of sampling sites. Site Description I Ile Grande - sandy - low energy - NE of bridge - relatively unoiled. 2 Ile Grande - sandy - low energy - SW of bridge - near end of excavation area. 3 Ile Grande - soil - heavily oiled - amid Juncus - above excavation area. 4 St-Micbel-en-Gr6ve - sandy - high energy - near low tide mark. 5 Aber Wrac'h - mud - 100m offshore at Perros.1 6 Aber Wrac'h - mud - 200m offshore at Perros. 7 Portsall - sandy - high energy - near wreck site - below high tide line. 8 Portsall - sandy - high energy - near wreck site - near rocks - 100m below high tide line. 9 Trez Hir - sandy - moderate energy - reference site - below high tide. 10 Trez Hir - sandy - moderate energy - reference site - 20m below high tide line. 11 Tregastel - sandy - low energy - Tanio spill site - 20m below high tide line. 12 Tregastel - sandy - low energy - Tanio spill. site - 50m below DRANCI high tide line. 2 by silica gel column chromatography, weight determination, glass capillary gas chromatography and mass spectrometry. Enumeration of Microbial Populations Total numbers of microorganisms per gram dry weight of sediment were determined by direct count procedures. Portions of collected sediment samples were preserved with formalin. Microorgansims in the preserved samples were collected on a 0.2 mm pore size Nuclepore filter which had been stained with irgalan black. The microorganisms were stained with acridine orange and viewed using an Olympus epifluorescence microscope. Cells staining orange or green were counted in 20 randomly selected fields and the mean concentration determined. Hydrocarbon utilizing microorganisms were enumerated using a three tube Most Probable Number (MPN) procedure. Serial dilutions of sediment samples, prepared using Rila marine salts solutions, were inoculated into sealed serum vials containing 10 ml Bushnf@ .1 Haas broth (Difco) and 50 ml of Arabian crude oil spiked with C Wadecane (sp. act. I mCi/ml). After 14 days incubation at 15'C, the C02 (if any) in the head space was collected by flushing and trapping in oxifluor CO 24 and quantitated by liquid scintillation counting. Vials showing Co2 production (counts significantly above background) were scored as positive and the Most Probable Number of hydrocarbon utilizers per gram dry weight calculated from standard MPN tables. Biodegradation Potentials Portions of sediment samples were placed into serum vials containing 10 ml [email protected] Haas broth end 50 ml lig@@ Arabian crude 1 51 spiked with either 14 C hexadecane, C pristane, C naphthalene, C benzanthracene or C 9-methylanthracene. After 14 days incubation, microbial hy@rocarbon degrading activities were stopped by addition of KOH. ' The 1 C02, produced from mineralization of the radiolabelled hydrocarbon was determiqV by acidifying the solution, flushing j4e headspace, tra ping the CO in oxifluor CO and quantitating the C p 2 1 by liquid scintillation coun?ing. The residua undegraded hydrocarbons and ?kodegradation products were recovered by extraction with hexane. The C in each solvent extract was determined and fractionated, using silica gel column chromatography, into undegraded hydrocarbon fractions (hexane + benzene eluates) and degradation product fractions (methanol eluate + residual non-eluted counts). A 0.75 cm diameter X 10 cm column packed with 70-230 mesh silca gel 60 was used. Radiolabelled material in each fraction was quantitated by liquid scintillation counting. Sterile controls were used to correct for efficiency of recovery and fractionation. Triplicate determinations were made for each sample and radiolabelled hydrocarbon substrate combinaf@on. The percent hydrocarbon minerw zation was calculated as C02 produced (above sterile control)/ C hydrocarbor4 added. The aercent hydrocarbon bioftgradation was calculated as CO produced + C methanol fraction + C residual (all above sterile 2 control) /14 C hydrocarbon added. Carbon balances generally accounted for approximately 90% of the radiolabelled carbon added to the sediment (except for naphthalene where volatility losses prevented efficient recovery). 3 Chemical Hydrocarbon Analyses Performed at ERCO For hydrocarbon analyses the samples were thawed, dried with methanol and extracted by high energy shaking with a mixture of methylene chloride-methanol (9:1). The extract was fractionated into an aliphatic (f 1) fraction and an aromatic (f 2) fraction using silica gel/alumina column chromatography. A 1 cm diameter X 25 cm column G cm alumina on top of 15 cm silica gel) was used. The f I f raction was eluted with 18 ml hexane; the f 2 fraction subsequently was eluted with 21 ml of a 1:1 mixture of hexane-methylene chloride. After reducing the volume of solvent by evaporation, the gross amount (weight) of hydrocarbon in each fraction was determined gravimetrically from an evaporated and dried aliquot of the extract. The extracts were subjected to quantitative glass capillary-gas chromatographic (GC) analysis. Selected aromatic fractions also were analysed by combined glass capillary gas chromatographic /mass spectrometric (GC/MS) analysis for qualitative identification of individual compounds and quantification of minor components. Participation in an intercalibration exercise under the direction of the National Analytical Laboratory indicated that these analyses were at the state Iof the art with repeatable t 20% detection of hydrocarbons in the ng g dry weight sediment range. The details of GC and GC/MS analysis employed are as follows: GC: Hewlett Packard 5840A reporting GC with glass; splitless injection inlet system; 30 m glass capillary column coated with SE-30 (=- 100,000 theorVcal plates); FID detector; temperfture programmed at 60-275*C min ; helium carrier gas I ml min transmission of integrated peak areas and retention time through HP 18846A digital communications interface to a PDP-10 computer for storage, retention index and concentration calculations. Deuterated anthracene (f ) and androstane (f I) were used as internal standards and response iactors were determined with known concentrations of the reported compounds. CC analysis was used to quantitate components of the f 1 fraction. GC/MS: Hewlett Packard 5985 quadrapole system (GC/MS Computer); mass spectrometer conditions: ionization voltage=70 eV, electron mHltiplier voltage=2200 V, scan conditions 40 amu to 500 amu at 225 amu S . Quantification of components of the f 2 fraction was accomplished by mass fragm@ntography wherein the stored GC MS data is scanned for parent ions (m ). The tabulated total ion currents for each parent ion is compared with deuterated anthracene (internal standard) and an instrumental response factor applied. Authentic polynuclear aromatic hydrocarbon standards were used to determine relative response factors (when no standard was available a response factor was assigned by extrapolation). In vitro Biodegradation Sediment was collected at sites 6 and 7 in November, 1979 for in vitro biodegradation experiments. Replicate one hundred gram portions of sediment were placed into 250 ml flasks to which 50 ml of a sterile solution containing 0.5% KNO + 0.5% KH 2PO 4. and 0.5 ml of light Arabian crude oil were added. The @lasks were agitated on a rotary shaker at 4 100 RPM. After two, four, and six weeks of incubation at 150C, the oil remaining in replicate flasks (two at each sampling time) was extracted and analysed as described below. Additionally, replicate 100 g portions of sediment were placed into I liter stainless steel buckets. The containers were continuously flushed with a solution of Rila marine salts supplemented with 10 ppm KNO3 + 10 ppm KH 2PO4. The height of the water level was adjusted to be 3 cm above the surface of the sediment layer. The f low rate was adjusted to 10 ml/h. After two, four, and six weeks of incubation at 150C the oil remaining in replicate sediment portions was extracted and analysed as described below. Analyses of in vitro Experiments Residual oil was recovered from samples by extraction with sequential portions of diethyl ether and methylene chloride. The sediment was shaken at 200 RPM with repetitive portions of solvent. The extracts were subjected to column chromatography to split the extracts into aliphatic (f L) and aromatic (f ) fractions. Columns were prepared by suspending silica gel 100 (E. W Reagents, Darmstadt, W. Germ.) in CH 2C12 and transferring the suspension into 25 ml burets with teflon stopcocks to attain a 15 ml silica gel bed. The CH 2C1 was washed from the columns with three volumes of pentane. Portions o? the extracts in pentane were applied to the columns, drained into the column bed, and allowed to stand for three to five minutes. The aliphatic fraction (f 1) was eluted from the column with 25 ml pentane. After 25 ml pentane had been added to the column, 5 ml of 20% (v/v) CH 2Cl2 in pentane was added and allowed to drain into the column bed. Fraction f I was 30 ml. The aromatic fraction (f 2) was eluted from the column with 45 ml of 40% (v/v) CH2Cl 2 in pentane. The fractions of each extract were then concentrated to about 5 ml at 35*C and transferred quantitatively to clean glass vials. Fractions f and f 2 were prepared for analysis by gas chromatography or gas c@romatography mass spectrometry. An internal standard, hexamethyl benzene (Aldrich Chem. Co., Milwaukee, WI.), was added to each sample. In fraction f , hexamethyl benzene (HMB) was present at 12.6 ng/ml; in fraction f 29 @&B was present at 25.2 ng/ml. Fraction f was analyzed by GC on a Hewlett-Packard 5840 reporting GC with FID dAector. The column was a 30 m, SE54 grade AA glass capillary (Supelco, Bellefonte, PA.). Conditions for chromatography were injector, 240OC; oven 70% for 2 min. to 270'C at VC/min. and hold for 28 min.; FID, 300OC; and carrier, He at 25 cm/sec. A valley-valley intergration function was used for quantitative data acquisition. Response factors were calculated using a-alkanes, (C 10-C 28 ), pristane and phytane standards. Fraction f was analyzed with a Hewlett-Packard 5992A CC-MS. Conditions for ciromatography were injector, 240'C; oven 70*C for 2 min. to 270'C at VC/min. and hold for 18 min. Data was acquired using a selected ion monitor program. Thirteen ions were selected for representative aromatic compounds. The ions monitored were 128, 142, 5 147 , 156, 170, 178 , 184, 192, 198, 206, 212, '220, and 226. The representative compounds were naphthalene, methyl naphthalene-, HMB as an interanal standard, dimethyl naphthalene, trimethyl naphthalene, phenanthrene, dibenzothiophene, methyl phenanthrene, methyl dibenzothiophene, dimethyl phenanthrene, dimethyl dibenzothiophene, trimethyl phenanthrene, and trimethyl dibenzothi 'ophene, respectively. The dwell time per ion was 10 msec. Instrument 'response factors were calculated by injecting known quantities of unsubstituted and C I and C 2 substituted authentic aromatic hydrocarbons and determining the integrated response for each compound. These values were used to extrapolate for quantitation of isomers and C 3 substituted compounds. For analysis of the polar fraction including microbial degradation products, three samples were selected for analysis by the University of New Orleans Center for Bio-organic Studies. The samples were: 1) flow through, 6 week incubation from site 6; 2) flow through, 6 week incubation from site 7; 3) agitated flask, 6 week incubation from site 7. Frozen samples were sent for analysis. At the Center for Bio-organic Studies the samples were extracted with successive portions of CH OH CH OH/CH Cl and CH Cl . The extracts were fractionated using 3 3 2 2 . 2 2 silica gel and the f fraction was collected, methylated and analysed by high resolution GC-ML RESULTS AND DISCUSSION The enumeration of hydrocarbon utilizing microorganisms indicated that numbers of hydrocarbon utilizers in the intertidal sediments increased significantly in response to hydrocarbon inputs (Table 2). Site 3, which is covered with seawater only at times of extreme high tide, showed very high populations of hydrocarbon utilizing microorganisms even three years after the AMOCO CADIZ spill-age. Sites 5 and 6 (located within Aber Wrac'h) and Sites 7 and 8 (located near Portsall) showed variable, but apparently elevated, numbers of hydrocarbon utilizers for up to two years following the spill. it appears that hydrocarbons contained within the mud sediments of Aber Wrac'h continued to exert a selective pressure on the microbial community that favored elevated populations of hydrocarbon utilizers for a longer period of time than sites on high-energy sand beaches. Site 2 showed evidence that the TANIO spill impacted the Ile Grande region. This site did not show elevated numbers of hydrocarbon utilizers in December 1978 or at later sampling times as a result of the AMOCO CADIZ spill, but in July of 1980, several months after the wreck of the TANIO, numbers of hydrocarbon utilizers were greatly elevated. A year later, however, the numbers of hydrocarbon utilizers had returned to background levels at this site. The unoiled control sites 9 and 10 and sites I and 4, which were impacted by the AMOCO CADIZ spill, did not show any evidence of elevated hydrocarbon-utilizing populations during the sampling period. Similarly, the offshore sites A-D in the Bay of Morlaix did not appear to be elevated at the time of sampling in November 1979. Sites 11 and 12 were added following the wreck of the TANIO and showed obviously elevated populations of hydrocarbon utilizers that persisted for over a year. 6 TABLE 2. MPN-Hydrocarbon Utilizers. (# X 103 /g dry wt.) Site 2-78 3-79 8-79 11-79 3-80 7-80 6-81 1 0.2 0.5 1 0.7 5 16 1 2 5 7 1 14 30 45000 1 3 2200 14000 41000 13000 160000 48000 24000 4 2 0.4 2 7 1 4 5 5 8 18 8 450 19 10 15 6 9 390 20 27 190 11 17 7 40 1900 1 2 12 2 2 8 57 350 150 8 3 1 10 9 0.7 0.4 3 1 4 1 4 10 0.1 0.2 4 1 2 2 3 11 - - - - 19000 140000 24000 12 - - - - 920000 140000 24000 A - - - 66 - - - B - - - 32 - - - C - - - 13 - - - D - - - 13 - - - The elevation in hydrocarbon utilizing populations, when detected, represented a shift within the microbial community. There generally was no evidence that total microbial biomass increased as a result of oiling although there generally was a tenfold variation in the microbial biomass between different sampling times (Table 3). TABLE 3. Direct Count. 0 x 10 8/g dry wt.) Site 12-78 3-79 8-79 11-79 3-80 7-80 6-81 1 3, 1 4 3 1 1 3 2 4 2 16 7 3 40 2 3 10 6 220 18 24 40 38 4 2 1 3 0.4 0.5 0.6 2 5 3 2 19 12 17 2 15 6 6 7 150 20 27 24 26 7 3 1 8 1 1 1 2 8 1 4 t 1 2 2 4 9 1 0.5 2 1 0.3 1 4 10 0.5 0.4 13 1 0.4 2 7 11 5 1 4 12 - 39 36 40 A 15 - - - B 16 C 3 D 10 7 The microbial hydrocarbon biodegradation p6tential measurements showed that following the AMOCO CADIZ oil spillage, indigenous microbial populations in the sediment at all sampling sites were capable of degrading both aliphatic and aromatic components of crude oil (Tables 4-8). The variability in the results is not indicated in these tables, but the standard error was less than 4% for the percentage degraded and less than 10% for the percentage mineralized in all cases. The biodegradation potentials indicated that n-alkanes were preferentially degraded and that pristane was degraded at approximately half the rate of n-hexadecane. For aliphatic hydrocarbons approximately 30% of the amount of hydrocarbon biodegraded was converted to CO 2 (mineralized). Methodological difficulties in handling naphthalene made it difficult to assess the true extent of biodegradation for this compound. It is apparent, though, that the indigenous microbial populations were capable of degrading light aromatic hydrocarbons. The rates of degradation of the 3- and 4-ringed polynuclear aromatic hydrocarbons were lower than for branched and straight chained aliphatic hydrocarbons. In the case of the polynuclear aromatic hydrocarbons, a very low proportion of the amount of hydrocarbon degraded was converted to CO 2* TABLE 4. Hexadecane biodegradation showing % degraded and (% mineralized). Site 12-78 3-79 8-79 11-79 3-80 7-80 6-81 1 40 41 21 17 10 25 17 (8) (11) (1) (15) (2) (12) (10) 2 43 38 26 22 25 19 6 (11) (13) (8) (13) (17) (12) (7) 3 45 46 29 23 51 19 33 (15) (15) (8) (18) (39) (14) (26) 4 36 48 21 25 8 26 17 (14) (13) (7) (14) (6) (19) (10) 5 42 46 25 35 32 24 23 (14) (14) (13) (14) (20) (18) (15) 6 34 47 29 26 36 18 20 (11) (12) (11) (20) (18) (11) (13) 7 31 45 13 31 2 20 28 (10) (13) (3) (21) (1) (14) (19) 8 40 43 21 35 3 17 34 (15) (11) (5) (19) (2) (12) (20) 9 28 32 22 35 7 27 34 (12) (3) (3) (14) (3) (15) (24) 10 37 30 21 45 8 22 25 (10) (3) (10) (32) (3) (14) (17) 8 TABLE 5. Pristane biodegradation showing % degraded and (% mineralized). Site 12-78 3-79 8-79 11-79 3-80 7-80 6-81 1 18 22. 12 17 18 17 27 (3) (3) (1) (3) (2) (3) (3) 2 23 22, 16 15 17 24 24 (3) (4) (3) (5) (3) (3) (1) 3 19 21 16 14 18 20 24 (2) (4) (3) (5) (5) (3) (6) 4 26 23 16 18 17 19 .20 (3) (4) (2) (4) (1) (3) (2) 5 21 28 21 17 16 22 25 (3) (6) (4) (5) (4) (3) (6) 6 21 30 19 19 17 20 21 (3) (6) (3) (5) (4) (4) (3) 7 25 24 16 25 12 23 23 (3) (4) (1) (4) (1) (2) (4) 8 31 23 21 18 20 21 23 (3) (4) (1) (5) (1) (2) (4) 9 27 20 22 19 17 22 24 (3) (2) (1) (5) (2) (2) (7) 10 29 - 21 20 18 21 20 (2) (3) (5) (2) (2) (8) TABLE 6. Biodegradation of naphthalene showing % degraded and (% mineralized). Site 3-79 8-79 11-79 3-80 1 3(2) 2(1) 3(31) 1(1) 2 9(7) 5(3) 2(2) 2(2) 3 12(10) 5(3) 2(2) 6(6) 4 7(6) 1(1) 5(5) 1(1) 5 8(6) 1(1) 7(7) 7(7) 6 11(10) 1(1) 1(1) 2(21) 7 10(9) 1(1) 6(6) IM 8 9(7) 1(1) 7(7) 10) 9 1(1) 2(1) 1(1) 1(1) 10 2(1) 1(1) 1(1) 9 TABLE 7. Biodegradation of 9-methylanthracene showing % degradation and (% mineralization). Site 3-79 8-79 11-79 3-80 7-80 6-81 1 10 - 1 5 6 9 (0) (0) (0) (0) (0) (0) 2 19 8 6 8 10 9 (0) (0) (0) (0) (0) (0) 3 18 18 6 7 7 10 (0) (0) (0) (0) (0) (0) 4 23 2 2 1 10 6 (0) (0) (0) (0) (0) (0) 5 17 4 2 7 21 6 (0) (0) (0) (0) (0) (0) 6 19 1 3 4 11 5 (0) (0) (0) (0) (0) (0) 7 15 7 3 2 9 5 (0) (0) (0) (0) (0) (0) 8 21 2 4 1 6 5 (0) (0) (0) (0) (0) (0) 9 15 11 1 4 11 4 (0) (0) (0) (0) (0) (0) 10 - 6 3 13 5 4 (0) (0) (0) (0) (0) TABLE 8. Biodegradation of benzanthracene showing % degradation and (% mineralization). Site 12-78 3-79 8-79 11-79 3-80 7-80 6-81 1 5 21 18 2 5 10 13 (0) (0) (0) (0) (0) (0) (0) 2 8 17 10 - 4 2 6 (0) (0) (0) (0) (0) (0) (0) 3 11 15 7 7 8 2 12 (0) (0) (0) (0) (0) (0) 4 4 5 6 2 6 3 5 (0) (0) (0) (0) (0) (0) (0) 5 8 13 14 2 10 11 8 (0) (0) (0) (0) (0) (0) (0) 6 4 8 11 2 4 1 6 (0) (0) (0) (0) (0) (0) (0) 7 11 3 6 7 4 3 6 (0) (0) (0) (0) (0) (0) (0) 8 6 5 5 4 1 7 4 (0) (0) (0) (0) (0) (0) (0) 9 2 8 5 7 2 13 5 (0) (0) (0) (0) (0) (0) (0) 10 1 - 14 8 11 12 4 (0) (0) (0) (0) (0) (0) 10 Based on the changes in the composition of the microbial community, as evidenced by elevations in numbers of hydrocarbon utilizing microorganisms and based on the microbial biodegradation potentials, it can be stated that biodegradation appears to have been a very important process that had the potential for significantly altering the composition of the hydrocarbon mixture that impacted the sediments of the Brittany Coast following the AMOCO CADIZ spill. With time the residual hydrocarbon mixture should contain increasingly high proportions of complexed branched and condensed-ring hydrocarbon compounds that are degraded relatively slowly by the indigenous microorganisms. The weight of the extractable hydrocarbons confirmed the occurrence of contaminating hydrocarbons at site 2 in July 1980, presumably as a result of the TANIO spillage (Table 9). Similarly, high concentrations of hydrocarbons were found in at Sites 11 and 12 which were closer to the TANIO wreck. The levels of hydrocarbons at Site 3 remained high throughout the sampling program. Sites 1, 4, 9, and 10 showed a general lack of significant hydrocarbon concentration that would be indicative of petroleum pollution.. Sites 5, 6, 7, and 8, in contrast, showed somewhat elevated hydrocarbon concentrations. TABLE 9. Weights of extractable hydrocarbons (ug/g). Site 12-78 3-79 8-79 11-79 3-80 7-80 6-81 I fq1 27 9 5 17 8 20 4 f 57 5 16 58 37 25 22 2 f1 52 21 42 147 50 2370 9 f2 53 q1q1 47 136 50 2160 29 3 f 272 2232 121092 108000 33000 16600 3680 f 338 2537 70329 95833 22500 32400 8080 4 fq1 21 22 9 20 35 L2 8 f 57 17 q1q1 19 29 17 49 5 fqi 122 146 56 213 65 68 21 F2103 82 99 233 73 156 42 6 fqL178 458 177 536 874 109 56 f2226 416 209 5q56 830 281 75 7 f 91 72 qL52 80 58 23 15 f2 75 59 123 63 39 26 29 8 f1q179 382 164 243 98 34 31 f 148 298 135 194 83 24 59 9 fq1 7 32 77 20 21 43 13 f2 3 34 78 32 21 36 25 q10 f 8 29 13 36 23 23 21 f2 q1q1 20 29 96 31 48 74 q10q1 f0q1 q- q- q- q- 515000 60800 320 f2 q- q- q- q- 512000 36300 440 12 f q- q- q- q- q- 73200 67 q-1 qt2 q- q- q- q- q- 14300 109 A f0ql q- q- q- 102 q- q- q- f q- q- q- 88 q- q- q- B f0ql q- q- 210 q- q- q- f q- q- q- 210 q- q- q- qC fI q- q- q- 13 q- q- q- f2 q- q- q- 21 q- q- q- D fI q- q- q- 21 q- q- q- fq2 q- q- q- 10 q- q- q- 8q12q1 The detailed gas-chromatographic and mass-spectral analyses of the samples collected at each site indicated a lack of significant petroleum hydrocarbons throughout the study at Sites 1, 4, 9, and 10 (Tables 10, 13, 18, 19). Site 2 showed some evidence of weathered hydrocarbons in 1978 and. a significant input of fresh petroleum hydrocarbons in July 1980 (Table 11). Site 3 had significant concentrations of weathered petroleum origin throughout the study (Table 12). Sites 5 and 6 showed an alteration of the hydrocarbon mixture with time that indicated the occurrence of biodegradation (Tables 14, 15). Samples at Sites 7 and 8 continued to show the presence of a relatively unweathered hydrocarbon mixture up to two years following the AMOCO CADIZ spill (Tables 16, 17). It appears that undegraded hydrocarbons were seeping into the surface sediments at Site 8 and it is postulated that either shifts in the sediment were repeatedly exposing hydrocarbons that had been protected from microbial degradation and/or that some oil continued to be washed ashore from the sunken AMOCO CADIZ vessel. Site 11 showed clear evidence of heavy oiling from the TANIO spill which persisted for a year following the spill (Table 20). The offshore sites sampled in November 1979 in the Bay of Morlaix failed to show the presence of AMOCO CADIZ oil. TABLE 10. Hydrocarbon concentration ng/g. SITE I C_# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 0 0 1 2 0 - 2 15 20 0 125 250 23 1 123 16 2 0 14 12 3 3 9 17 65 15 438 86 156 5 91 pristane 62 22 27 76 43 30 102 18 3 2 3 .5 8 4 10 phytane 11 2 2 5 6 3 2 19 0 2 2 3 7 3 5 20 5 2 2 4 7 3 4 21 6 1 3 4 8 2 7 22 6 2 3 4 8 3 9 23 9 2 4 6 9 5 18 24 9 2 3 5 8 2 24 25 17 2 5 14 9 13 34 26 8 2 2 2 6 3 28 27 10 1 4 10 7 6 35 28 7 1 1 2 5 2 21 29 18 5 6 10 10 18 38 30 8 1 7 6 3 4 35 alkanes: 1.2 0.9 20.0 2.3 5.6 6.0 2.3 isoprenoids pristane: 5.6 7.8 13.5 - 7.3 1.6 50.0 phytane 12 TABLE 11. Hydrocarbon concentration ng/g. SlTE 2 C-# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 0 3 6 0 0 11700 3 15 44 115 331 154 39 18200 ill 16 0 11 18 0 4 19800 11 17 65 40 169 68 18 22000 93 pristane 27 38 160 1100 6 19800 46 18 8 9 8 181 5 22600 to phytane 17 126 32 151 15 25900 15 19 30 26 5 35 2 23400 6 20 18 16 9 9 8 18600 9 21 14 40 18 15 5 16700 13 22 16 29 7 10 9 12800 14 23 14 11 8 38 8 10900 23 24 12 11 7 6 13 10000 22 25 45 70 30 64 146 7120 35 26 22 12 6 10 16 6450 30 27 34 15 23 36 24 6320 33 28 53 40 6 29 10 4780 15 29 51 47 9 39 53 6040 36 30 31 95 71 116 55 4380 20 alkanes: - 1.5 2.5 0.3 3.2 1.3 3.5 isoprenoids pristane: 0.4 0.8 5.1 7.3 0.4 0.8 3.0 phytane TABLE 12. Hydrocarbon concentration ng/g. SIT.E 3 C-# 12-78 3-79 8-79 t)-79 3-80 7-80 6-81 14 19 - 3525 1040 1390 - - 15 - - 11025 0 633 - - 16 17 121 9350 0 400 - - 17 32 12 9550 3440 200 - - pristane 213 809 145150 47900 104 - - 18 48 86 20250 3600 106 - - phytane 985 3088 275900 130000 673 - - 19 -155 948 63075 29200 283 825 - 20 149 247 36000 ltqOO 0 - 1270 21 31 241 17975 6210 508 2103 2380 22 25 12 8875 1440 130 - 3050 23 38 56 11100 0 0 - 3280 24 54 48 7925 3510 0 - 4640 25 - - 12600 21300 2340 - 5700 26 - 160 14900 2540 3229 - 5190 27 172 898 45250 0 1160 516 10800 28 81 934 18600 0 109 - 2460 29 41 2025 31250 2090 3330 1692 142120 30 - 400 75775 0 118 13900 3120 alkanes: 0.1 - 0.1 0.1 0.8 - - isoprenoids pristane: 0.5 0.3 0.5 0.4 0.2 - 1.9 phytane 13 TABLE 13. Hydrocarbon concentration ng/g. SITE 4 C-# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 0 0 0 0 - 1 2 15 0 2 0 0 3 13 8 16 0 2 0 0 5 8 7 17 18 4 4 4 12 15 8 pristane 60 6 0 2 3 4 3 18 37 3 0 2 10 6 8 phytane 153 18 0 8 8 6 11 19 68 9 2 5 9 7 6 20 37 10 2 3 8 7 6 21 30 8 4 4 12 5 9 22 21 5 1 3 7 1 8 23 27 6 2 6 8 7 17 24 20 4 0 3 5 3 17 25 23 6 1 3 L4 3 27 26 42 7 0 3 4 6 16 27 37 7 10 7 17 11 48 28 60 9 3 3 3 3 7 29 47 19 9 18 28 22 61 30 43 1 30 1 50 2 12 alkanes: 0.2 0.7 - 0.6 2.8 4.1 2.3 isoprenoids pristane: 0.4 0.4 - - 0.4 0.6 0.3 phytane TABLE 14. Hydrocarbon concentration ng/g. STTE 5 C-# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 19 37 2 16 - 5 3 15 20 65 167 59 11 35 29 16 - - 18 0 - 12 8 17 25 112 122 0 95 18 585 pristane 24 [50 Ll 885 8 175 108 18 16 50 17 0 4 2 4 phytane 158 293 43 158 12 36 20 19 64 140 12 16 4 3 8 20 43 5 L7 19 16 7 9 21 57 50 23 208 8 13 17 22 39 18 19 14 6 18 11 23 25 36 23 30 20 18 20 24 4 30 15 18 9 13 20 25 22 5 39 115 99 ill 100 26 46 5 11 24 21 11 2L 27 5t 93 30 87 48 29 62 28 70 64 6 13 7 13 10 29 96 1136 17 113 13Cj 86 122 30 17 57 80 10 20 0 23 alkanes: 0.3 0.5 5.1 0.3 5.5 2.7 4.4 isoprenoids pristane: 0.2 0.5 0.3 12.0 - 0.7 5.4 phytane 14 TABLE 15. Hydrocarbon concentration nal/g. SITE 6 C-# 12- 7 83-79 8-79 11-79 3-80 7-80 6-81 14 - 82 0 10 - 4 2 15 8 61 28 ill 68 28 30 16 - - 0 0 - 9 9 17 23 117 14 164 200 6 101 pristane 102 125 15 71 70 0 12 18 - 8 0 28 - 6 7 phytane 360 489 135 289 225 23 22 19 101 121 20 20 - 0 15 20 35 12 0 42 - 2 10 21 86 180 59 223 35 31 30 22 - 149 4 48 38 9 6 23 39 58 16 81 80 28 61 24 5 - 5 37 27 5 10 25 109 159 80 128 484 47 85 26 128 151 10 50 53 6 30 27 68 126 88 135 .155 48 124 28 97 56 13 105 28 11 21 29 159 319 55 200 590 122 245 30 36 140 254 342 30 18 55 alkanes: 0.1 0.3 0.2 0.8 0.9 1.5 2.7 isoprenoids pristane: 0.3 0.3 0.1 0.3 - 0.0 0.5 phytane TABLE 16. Hydrocarbon concentration ng/g. SITE 7 C-# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 9 4 9 0 2 14 43 15 7 7 43 11 11 29 77 16 51 10 58 18 32 29 64 17 109 20 96 27 63 38 85 pristane 154 24 102 25 47 35 7 t8 121 33 113 36 39 41 64 phvtnne 256 65 159 43 86 43 39 19' 94 46 139 55 44 47 65 20 t22 36 124 23 56 35 67 21 83 25 106 36 49 29 42 22 108 24 94 30 58 0 45 23 84 23 86 28 59 24 43 24 61 t8 80 46 47 24 54 25 Ill 27 89 62 25 32 26 105 27 62 25 42 22 35 27 54 20 110 48 30 17 35 28 126 42 52 15 28 14 17 29 92 28 62 28 49 26 44 30 112 36 356 152 29 37 29 alkanes: 0.6 0.7 1.0 1.4 1.1 1.6 3.3 isoprenoids pristane: 0.6 0.4 0.6 0.6 0.6 0.8 phytane 15 TABLE 17. Hydrocarbon concentration ng/g. SITE 8 C-# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 160 34 16 24 - 8 12 15 368 77 66 73 7 21 80 16 518 87 47 73 22 23 29 17 640 113 80 L04 44 31 64 pristane 941 427 98 135 41 38 108 18 818 219 70 108 50 34 40 phytane t839 955 116 171 74 54 46 19 1122 352 85 158 57 54 42 20 849 203 81 97 51 35 29 21 530 121 62 66 44 28 27 22 430 135 50 64 38 24 30 23 314 85 47 59 34 32 36 24 272 85 42 68 30 43 21 25 54 190 52 42 56 18 25 26 489 234 33 46 28 19 28 27 328 197 71 31 26 41 22 28 431 292 24 19 11 12 11 29 362 326 15 30 63 27 33 30 315 411 170 10 '118 14 29 alkanes: 0.7 0.3 1.1 1.1 1.0 1.1 1.2 isoprenoids pristane: 0.5 0.4 0.8 0.8 0.6 0.7 2.3 phytane TABLE 18. Hydrocarbon concentration ng/g. SITE 9 C-# 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 0 0 0 0 - 28 - 15 0 0 3 8 - 42 16 0 0 3 3 - 21 - 17 3 11 5 22 8 36 19 pristane 1 6 2 - 0 1 18 3 6 3 5 6 9 9 phytane 2 4 2 3 - 0 5 19 4 3 4 4 8 6 12 20 4 3 6 11 6 12 21 4 6 4 6 11 4 15 22 4 7 3 5 9 6 12 23 4 9 3 10 8 7 19 24 4 F 2 5 4 4 15 25 -7 13 8 15 7 62 29 26 8 23 1 3 1 2 15 27 8 12 3 11 3 14 44 28 11 34 0 3 1 0 12 29 11 22 0 12 15 24 65 30 6 8 13 1 1 3 25 alkanes: 1.8 2.0 1.6 7.1 - - 5.7 isoprenoids pristane: 0.6 0.2 2.6 0.9 - - phytane 16 TABLE 19. Hydrocarbon concentration ng/q. SITE 10 C-fl, 12-78 3-79 8-79 11-79 3-80 7-80 6-81 14 0 1 2 - - 25 7 15 0 5 33 3 - 42 747 16 L 5 29 3 - 19 17 17 3 11 53 14 8 41 228 pristane 1 3 7 3 - 7 4 18 1 7 9 10 4 8 10 phytane 2 0 4 2 - 2 5 19 3 11 5 6 1 7 11 20 4 8 4 6 3 6 8 21 3 4 5 6 4 246 14 22 2 10 5 5 3 6 15 23 2 8 7 7 4 10 31 24 1 4 5 5 2 1 11 25 3 226 13 24 7 55 76 26 2 13 6 7 2 2 33 27 1 11 22 21 5 31 137 28 1 11 8 7 5 7 23 29 0 5 38 33 14 53 194 30 0 3 23 3 - 4 34 alkanes: 1.9 - 9.0 - - 4.0 63.6 isoprenoids pristane: 0.5 - 1.9 - - 3.5 0.8 phytane TABLE 20. Hydrocarbon concentration ng/g. SITE 11 C-It 3-80 7-80 6-81 14 605000 73200 69 15 661000 88200 285 16 625000 89100 83 17 636000 93600 110 pristane 342000 66900 418 18 643000 110700 76 phytane 231000 53400 37" 19 629000 129000 80 20 680000 1,42000 128 21 724000 132000 122 22 719000 141000 141 23 719000 142000 196 24 724000 140000 209 25 685000 133000 202 26 718000 155000 224 17 669000 167000 207 28 62?000 192000 108 29 620000 168000 152 -10 479000 190000 3L8 Alkanes:Isoprenoids 0.2 2.4 0.3 Pristane:Phytane 0.7 1.3 1.1 17 The significant features of the chemical changes that were observed included a marked decrease in the proportion of n-alkanes relative to isoprenoid hydrocarbons, the transient occurrence of an increase in unresolved hydrocarbons within the first year following the AMOCO CADIZ spillage, and the decreased importance of unsubstituted polynuclear aromatic hydrocarbons relative to dibenzothiophenes and the comparable or substituted forms of the polynuclear aromati c hydrocarbons (Figs. 2 and 3). The in vitro hydrocarbon biodegradation experiments confirmed the fact that the indigenous microbial populations were capable of rapid and extensive degradation of Arabian crude oil. Much greater rates of biodegradation were observed in agitated compared to flow through experiments (Figs-* 4-9). Both the in vitro experiments and the analysis of field experiments support the TFY@othesis that mixing energy has a very significant effect on the rates of hydrocarbon biodegradation. Rates of biodegradation appear to be environmentally influenced by the turbulence of mixing which can ensure a continued supply of nutrients and oxygen as well as dispersing the oil so as to establish a favor- able surface area to volume ratio for rapid microbial hydrocarbon biodegradation. The similarity of changes. observed in the composition of the hydrocarbon mixture in vitro compared to the analysis of field samples also suggests that nutrients were not a limiting factor that determined the rates of hydrocarbon biodegradation. The analysis of the polar fractions from the in vitro experiments showed some surprising results (Table 21). There was a lack of oxygenated aromatic compounds. It had been predicted that there would be a greater accumulation of polar products from aromatic biodegradation since less CO was being produced than from aliphatic biodegradation where a signiAcant proportion of the hydrocarbon that was biodegraded was released as CO 2. There were significant accumulations of polar compounds that appear to be biodegradation products of aliphatic hydrocarbons, especially as C,6-C,8 acids. Interestingly, the major polar products included unsaturat e acids. As a rule, the predominant biochemical pathway for the biodegradation of straight chained hydrocarbons does not involve the formation of unsaturated compounds, although a biochemical pathway has recently been elucidated for some bacteria that does introduce a double bond into the hydrocarbon. It appears that the microbial populations indigenous to the sediment of the Brittany Coast possess such a biochemical capability. 18 wm@f .0"I 0 ABER WRACH MA RCH 1979 ca 10 0 -11 GOA--- -, S .... . . .. ... ..... . . . . . . .. . . . . .. . .. . .. . . . . .. . 10.11- P( -All 0 0 eo =0 FIGURE 2. Changes in the relative concentrations of aliphatic hydrocarbons at sites 3, 5, and 7. 19 PORTSALL ABER RAC- ILE GRANDE NOVEMOER 1979 NOVEMBER .9?9 NOVEMBER 1979 ".Pb 1. so Poo 1. n r-Lffl N C C C C P C C C C 0 C C C N C C C C P C C C C 0 C C C N C C C C P C CC C 0 C C C 1 2 3 A 1 2 3 4 ' 1 3 1 2 3 4 1 2 3 4 1 2 3 1 1 3 1 1 1 3 4 1 2 3 NAPHTHA PHENAN DWENZO NAPHTHA PHENAN OISENZO NAPHT A PHEN AN OISE NZ 0. LENE THRIENE T"OP.ENE ILENE THRENE T..OPHENIE LE NE THRENIE TMIOPHENE FIGURE 3. Changes in the relative concentrations of aromatic hydrocarbons at sites 3, 5, and 7. 20 FLOW THROUGH 2 WEEKS 5- FLOW THROUGH SITE 6 2 WEEKS SITE 7 4- 3- 3 Z 2- 0 I'- cc 1- II- z W U z 05- FLOW THROUGH 6 WEEKS SITE 7 LU >4- 4 WEEKS 1-- 4 SITE 6 4 7- -j- W cc 2- 2 11 12 13 14 15 16 17 PR 18PH 19 20 21 22 232425 2627 28 11 12 13 14 IS 16 11 PR 18 PH 19 20 21 22 23 24 2 5 26 2 7 28 F IGURE 4. (Left column) Changes in the relative concentrations of aliphatic hydrocarbons in flow-through experiment with sediment from site 7. FIGURE 5. (Right column) Changes in the relative concentrations of aliphatic hydrocarbons in fLow-through experiment with sediment from site 6. 21 FLASK 4 0 TIME SITE 7 3 FLOW THROUGH 2 3- SITE 6 2 WEEKS z 2- 0 Cr 3 z LLJ 4 WEEKS 1 2 WEEKS 2- z 0 0 El ul 4 WEEKS 6 WEEKS 3- LU Cr 2- 6 WEEKS It C3 11 12 13 14 15 16 1? PR 18 PH 19 20 21 22 23 24 25 26 27 28 N C' C2 C3 P CI C2 C3 0 CI C2 N N N P P P B D D D T B 8 -B T T T FIGURE 6. (Left column) Changes in the relative concentrations of aromatic hydrocarbons in flow-through experiment with sediment from site 6. FIGURE 7. (Right column) Changes in the relative concentrations of aliphatic hydrocarbons in flask experiment with sediment from site 7. 22 5- FLASK 0 TIME SITE 6 12.1 4- 10 FLASK SITE 6 3- 8 7 2- z 0 S 0 Wk. -4 -3 z LU -2 B. 4 Ppm L) 2 WEEKS z -1 0 2 > I IPP 2 Wk. TM I r-@j -j 4 WEEKS S.1 Uj cc 4 Wk 6 WEEKS -2 6 Wk r --- ra 11 12 13 14 15 16 17 PR 18 Pk 19 20 21 22 23 24 25 26 21 28 N C' C2 C3 P C, C, C, D C, C, C 'F C, N N N p p p 8 D 0 D F T B 8 8 T T T FIGURE 8. (Left column) Changes in the relative concentrations of atiphatic hydrocarbons in flask experiment with sediment from site 6. FIGURE 9. (Right column) Changes in the relative concentrations of aromatic hydrocarbons in flask experiment with sediment from site 6. ILL 23 TABLE 21. Concentrations and Identities of Polar Compounds ng/g SITE SITE SITE 7 6 7 FLASK FLOW FLOW THROUGH THROUGH dodecanoic acid 16.5 2.3 4.7 tetradecanoic acid 22.2 13.7 24.5 methyltetradecanoic acid 15.0 19.7 43.3 pentadecanoic acid 10.5 13.5 24.0 hadecenoic acid 47.0 129.0 217.0 hexadecanoic acid 73.9 109.0 157.0 isoheptadecanoic acid 8.3 7.8 11.9 heptadecanoic acid 3.2 8.6 14.3 octadecenoic acid 59.3 76.4 99.2 octadecanoic acid 44.4 34.2 44.6 non.adecanoic acid 10.1 25.4 - eicosanoic acid 9.8 25.4 55.8 tetracosanic acid 14.5 3.0 13.8 hexacosanic acid 7.7 1.4 10.8 octacosanic acid 15.8 13.3 15.6 isocyclopropaneoctanoic acid 15.5 25.5 42.2 methyloctahydrophenanthrene- 21.3 2.0 - carboxylic acid (tent.) 24 CONCLUSIONS Microbial degradation appears to have played a very important role in the weathering of oil spilled from the AMOCO CADIZ. Microbial hydrocarbon degradation potentials are in general agreement with the observed changes in the composition of oil stranded within the littoral zone. The chemical evolution of the hydrocarbon mixture within intertidal sediments led to a relative enrichment in isoprenoid alkanes, a transient complex unresolved mixture, and a relative enrichment of dibenzothiophenes and alkylated phenanthrenes. There was a general, but variable decline in concentrations of hydrocarbons over the three year period following the AMOCO CADIZ spill within Aber Wrac'h. The concentrations of hydrocarbons also declined at sites that were regularly covered by tides. At the one site in Ile Grande, which is not subject to daily tidal washing, the concentrations of hydrocarbons remained high even three years following the spill. At nearby sites within the Ile Grande salt marsh, which were physically cleansed of AMOCO CADIZ oil, there was little chemical or microbial evidence of any impact from 'the AMOCO CADIZ spill at any of the sampling times. The incurrence of oil from the TANIO wreck was apparent even at sites that had been oiled as a result of the AMOCO CADIZ spill. The microbial population levels generally reflected the relative degrees of persistence of petroleum hydrocarbons. The microbial community at ail of.the sites studied had essentially the same potential capability for degrading hydrocarbons and as such the differences in the hydrocarbon concentrations and composition recovered from the field samples probably reflect the initial rates of oiling and environmental influences. The indigenous microbial community retained the capability of responding to a second incursion of oil resulting from the TANIO spill. Both the field experiments and the in vitro studies suggest that mixing energy, related to nutrient a@-d oxygen availability, was extremely important in permitting the high rates of observed oil weathering. The occurrence of both saturated and unsaturated acids in the sediments studied in vitro suggest that several biochemical pathways were active in the bioTei-radation of the aliphatic hydrocarbon fraction. The hydrocarbon biodegradation potential suggested that relatively high concentrations of oxygenated aromatic hydrocarbons should accumulate, but for unexplained reasons the analyses of the polar fraction generally failed to show such accumulations. 25 LABORATORY SIMULATION OF THE MICROBIOLOGICAL DEGRADATION OF CRUDE OIL IN A MARINE ENVIRONMENT by D. Ballerini(l), J. Ducreux(l) and J. Rivi&re(2) (1) Institut Frangais du P6trole - Direction de Recherche "Environnement et Biologie P6troli&rell 1 et 4 avenue de Bois-Pr6au - 92506 RUEIL-MALMAISON - FRANCE (2) Institut National Agronomique, Paris-Grignon 16, rue Claude Bernard - 75231 PARIS 05 - FRANCE This study essentially intends to quantify the biodegradation process of a crude oil in optimum conditions compatible with the marine environment. Experiments were conducted in the Laboratory reactors (batch and continuous cultures), with perfect monitoring of all physicochemical parameters such as pH (pH 8.1), temperature at 200C, mixing rate 600 rpm, and aeration velocity (1 liter of air/liter of medium per hour). The composition of the mineral medium was defined by taking the mean composition of salts in the Atlantic Ocean as a basis, and enriching it with nitrogen (235 mg.1-1), phosphorus (26.7 mg. 1-1) and iron (0.4 mg.1-1). In order to reproduce conditions prevailing at sea as closely as possible, in which the evaporation of light products is not negligible Arabian Light Crude was employed (ALC 240+) from which all fractions distilling below 2400C were removed by low pressure distillation. The analytical methodology employed to observe the crude oil biode- gradation process is shown schematically in the following figure. The gas flow was passed through a trap containing CC14, which retained the evaporated hydrocarbons, and then through a second trap containing a known quantity of 1 N KOH, which retained the carbon dioxide. The hydrocarbons were then determined by infrared spectrometry. The C02 produced was determined by titrimetry. Liquid samples were taken during fermentation. The first sample was centrifuged to separate the hydrocarbon phase from the aqueous phase, which was then filtered (filter pore diameter 0.22 p) to eliminate fine particles in suspension. The following were analyzed in this perfectly clarified aqueous phase: � total organic carbon (Dohrman DC.50 instrument), � dissolved C02 using the Warburg equipment, 27 E R 'l 141, E EXHAUST GAS < ROCARM-,,O,!.q LIQUID 56AWPL.C- LIQUID SINMPLIE -'IJGArioN EXT R h CTI ori (CH 0@,@ T + rlL I KA I ON (0. IZ2 PHASE C -LS ORATiON WAS V I I'l C. O-Rz@rl I c RCSIDUAL C) L 11-1 E-11 OF EXTPP,c rioli ASPSALT SNES ;7, 7N 1 THIN LhYER CHROMA-70<vRATHY OR LIQUil) C@,Re%!'-AAr0G.R%APHY ...... . .. ...... . .. ig C c H Rop,,,,N r(,.) .5f.A PbfY ,)merP SMC.TPOMFTCJ il-A Ic st-acr" y 4,44 t-VIS's I:- c 1 A;@ 0 N T's< FAL kl. r, r@ 1, C@@j Zlij p- @-c 7A 'A@, =br" A" F i r. v iz 28 � residual phosphorus, � residual ammoniacal nitrogen and the intracellular nitrogen concen- tration (Kjeldahl's method); these two analyses served to determine the quantity of biomass formed. The residual hydrocarbons were extracted from a second liquid sample. Asphaltenes were precipitated from the hydrocarbon residue using hot heptane for one hour, dried and weighed. The residue obtained after evaporation of the heptane was processed to separate the three main families of hydrocarbons in crude oil: saturates, aromatics and resins, by thin layer chromatography (50 mg samples) or liquid chromatography (samples weighing about 1 g). The sum of the weights of the three fractions thus recovered, using liquid chromatography, compared with the initial rate of the hydrocar- bons deposited on the column, always accounted for a proportion between 90 and 100 %. The loss percentage increased whenthe test samples were taken at increasingly long culture times, hence with samples that underwent the longest biodegradation times. These losses are likely to be due largely to the retention of polar compounds of the resins on the column, compounds that are formed during oxydation reactions, or pos- sibly by biochemical co-oxydation, and whose concentration increases with biodegradation time. Using the different fractions obtained (saturates, aromatics, resins), we performed more detailed analyses by gas phase chromatography (Varian 3700 chromatograph) equipped with "Splitless" injection and flamme ionization detector), a combination of gas phase chromatography and mass spectrometry (Varian CH5DF spectrometer), proton NMR that yielded the fraction of hydrogen belonging to methyl groups in the sa- turates family, 13C NMR, which yields the percentage of aromatic carbon in comparison with total carbon in the aromatic fraction, and by infra- red spectrometry on the resins. 1. BATCH CULTURES 1.1. Biodegradation of hydrocarbon families and sub-families in ALC 240+. We selected a mixed culture of bacteria from samples of muds and slud- ges collected on places hit by crude oil spills. The experiment was conducted with ALC 240+ in an initial concentration of 2.65 g.1-1, over a period of 48 hours. Of the 2.65 g.1-1 of initial hydrocarbons, 1.08 g.1-1 were consumed, representing 41 Y. degradation. It appears clearly that the saturates fraction is most sensitive to biodegradation, because 67 % of this fraction were consumed, whereas only 27 % of the aromatics fraction were degraded. The quantity of hydrocarbons evaporated was negligible. From the standpoint of reproducibility of results, a previous experi- ment yielded the following results: hydrocarbons consumed 44 %, satu- rates degraded 63.1 %, aromatics disappeared 48.6 29 The saturated hydrocarbons were most rapidly biodegraded. At the end of the culture, the disappearance of aromatic compounds is accompanied by an enrichment of the aqueous phase in organic carbon, the concentration of which may reach 250 mg.1-1. This observation tends to show that a large part of the aromatics are only partly oxi- dized before passing into the aqueous phase. The resins were only slightly attacked if at all, and the asphaltene concentrations at the start and end of the batch culture were absolu- tely comparable, demonstrating total insensitivity of these substances to biochemical processes. The determination of n-alkanes (C14-C35) and detectable isoprenoids (C16-C23) by gas phase chromatography showed that these compounds disappeared almost totally by the end of the culture. The mass spectrometry analysis of the "saturates" fraction showed that the alkanes were mainly biodegraded, as 88.9"% disappeared at the end of the culture. This enables us to postulate that, in addition to the n-alkanes and isoprenoids, which only account for 14.8 % of the "saturates" fraction, the bulk of the iso-alkanes present in the crude oil was consumed by microorganisms. Among the naphtenic compounds, the 1- and 2-cycle naphtenes were mainly consumed, with respective biodegradation rates of 44 and 47 %. Proton NMR analyses giving the CH3/CH2 ratio, conducted on the satu- rates, failed to indicate any significant difference between the start and end of the batch culture. With respect to the "aromatics" fraction, the action of microorganisms mainly affected the mono- and di-aromatic compounds. At the end of the culture, all the mono- and di-aromatics with a number of carbons less than 16 had disappeared. Among the mono-aromatics, the substances most sensitive to microbial action were the alkylbenzenes, of which 67.7 % disappeared at the end of the culture, and the benzocycloparaffins, with a consumption rate of 46.2 %. The differences measured for benzodicycloparaffins were not sufficiently wide to be meaningful. As for di-aromatic compounds, the microorganisms displayed a very clear effect on the residual concentration of naphtalenes, of which 50 % disappeared after 48 hours of culture. Through a second experiment, we investigated the changes in composition of the aromatics fraction, by drawing a distinction between sulfu- compounds and other aromatics. Apart from those with a rough formula CnH2n-10S, the sulfur-containing compounds were not attacked by bacteria. The aromatics/sulfur-compounds ratio of 0.98 before biodegradation decreased to 0.82 after biodegra- dation, showing that it was mainly the non-sulfur-containing aromatics (mono- and di-) that disappeared. In addition, the weight percentage of sulfur in the aromatics fraction increased with time from 4.05 to 4.15, confirming the enrichment of this fraction in sulfur-containing sub- tances. 30 The 13 C NMR analyses used to quantify the aromatics C/total C ratio failed to reveal any significant difference before (43.4 %) an after (43.7 %) biodegradation. With respect to the resins, part of the polar compounds of this frac- tion formed during biodegradation remained absorbed on the liquid chro- matography column, and consequently the analyses performed on the eluted resin fraction were not truly representative. This retention of polar compounds of the liquid chromatography column was confirmed by elemental analysis, showing oxygen to drop from 2.75 % (by weight) at the start of the batch culture to 2.35 % at the end of the culture. The determination of molecular weights of the resins yielded the following results: 690 at the start of the batch culture, 740 at the end of the batch, namely very slightly differing molecular weights. 1.2. Examination of oxidation products. Analyzing the aqueous phase sampled at the end of the batch culture (volume sampled = 1 liter), centrifuged and filtered, we found a total organic carbon concentration of 260 mg.l"l. We carried out an esterification (BF3-CH30H) of the compounds of this aqueous phase. The organic extract was evaporated and weighed. The weight of the extracted compounds, related to one liter of culture, was 120.2 mg. In the acidified residual aqueous phase, initial extrac- tion with CH2C1 20 followed by a second extraction with benzene, yielded a new organic phase that contained polar compounds such as alcohols, ketones and phenols, which represented 7.58 mg/1 of aqueous phase after evaporation. Identification by GC/MS coupling of compounds separated by gas phase chromatography was difficult because of the presence of a strong background of poorly resolved constituents, which could be hydrocarbons. Despite these problems, we succeeded in identifying normal and iso acidic compounds in the aqueous phase, in the form of their correspon- ding esters, obtained after esterification of the aqueous phase. The GC/MS coupling enabled us to observe the masses m/e = 74 characte- ristic of n-esters, and m/e = 88 characteristics of iso-esters. 1.3. Changes in microbial flora with time. Three samples were taken during the batch culture, the first at the start of growth, the second during the active biodegradation phase (after 15 hours of culture), corresponding to consumption of the satu- rated hydrocarbons, and the third after 25 hours, in the slowdown period of the biodegradation process, corresponding to microbial attack of the aromatics. In the three different stages investigated, different dominant strains were found, belonging io two genera only, Pseudomonas, and Moraxella, confirming that changes in the crude oil during a biodegradation pro- cess are accompanied automatically by changes in the microbial flora. At the start and middle of the batch culture, we chiefly identified 31 bacteria of the genus Moraxella, indicating that these strains are per- fectly adapted to the hydrocarbons present at that particular time in the culture and, being dominant, they therefore naturally and preferen- tially consumed the hydrocarbons of the "saturates" fraction, as these types of constituentswere biodegraded during this period. At the end of the culture, however, when the "aromatics" fraction was attacked by the microorganisms, only the Pseudomonas strains were dominant. This investigation again confirms that, to observe a significant degradation of hydrocarbons in a crude oil containing a wide variety of compounds, a mixed culture of bacteria is certainly more effective than a pure bacteria, of which the metabolism is only adapted to a given type of constituent. 1.4. Toxicity analysis of oxidation products. During the different ALC 240+ crude oil biodegradation experiments, we always observed a substantial rise in total organic carbon (TOC) * concentration in the aqueous phase with the Tassage of time,.with a regular final concentration around 200 mg.1- . We decided to evaluate the potential toxicity of these solubilized products in the aqueous phase, enriched mainly in aromatics and oxi dation products of certain hydrocarbons present in the crude oil. In particular, the mutagenicity of two samples was determined by the Ames test, the procedure of which is described in detail in Mutation Research 1975, 31, pp. 347-364. The first sample was taken at the start of the culture (with a TOC of 30 mg.1-1), and the second at the end of the batch culture (sample with a TOC of 210 mg.1-1). The correlation between carcinogenic properties and mutagenic properties of 300 compounds was pointed out in Proc. Natl. Acad. Sci., (U SA), 1975, 72, pp. 5135-5139. The principle of the Ames test is to measure the mutagenic properties of compounds that may be carcinogenic in Salmonella bacteria. The two samples were tested in a range from 0.1 to 500 pl on three of the five strains used in the Ames test (TA.1538, TA.98 and TA.100) in order to detect the mutagenicity of products such as HAP, for example. No mutagenic activity was detected in these two samples. It was shown finally that neither of these two samples had any toxic effect on the three strains tested (TA.1538, TA.98 and TA.100). 1.5. Study of the biodegradation of a mixture of pure hydrocarbons. The mixture of pure hydrocarbons consisted of two n-alkanes, hexadecane and octacosane, one isoprenoid, pristane, a two-ring naphtene, decaline, two mono-aromatics, p-cymene and dodecylbenzene, one di-aromatic, dimethyl-naphtalene, one tri-aromatic, phenanthrene, and two sulfur-, containing aromatics, benzothiophene and dibenzothiophene. 32 Experiments were conducted in batch culture in the same conditions as those described for Arabian Light Crude. The mixed culture of bacteria used was the mixture of the strains Pseudomonas and Moraxella isolated and purified, described in Section 1.3. By sucessive cultures in flasks with the pure hydrocarbon mixture as the only carbon substrate, the mixed culture was progressively adapted to grow on these ten hydrocar- bon compounds. The culture was carried out in batch for 61 '/2 hours, and we observed the changes in the biomass, total hydrocarbons, and each compound, and also the organic substances that passed into the aqueous phase. Following a lag phase of about 10 hours, a growth acceleration phase was observed up to the 25th hour, then a linear phase from the 25th to the 35th hour, and finally the slowdown of bacterial growth. At the end of the batch, the dry cell weight was 0.5 g.1-1. The biodegradation process of total hydrocarbons perfectly matched the microorganism growth pattern. After 61 Y2 hours, 82.8 % of the hydrocarbons were degraded. It appears that the three most volatile compounds, paracymene, decaline and benzothiophene, could not be found after extraction, from the very outset of the experiment. These three products must therefore disappear chiefly during extract evaporation operations. However, a small propor- tion passes very rapidly into the aqueous phase in the marine environ- ment, because the latter contained oxidation products of p-cymene among others, as well as benzothiophene. The two n-alkanes, n-hexadecane and octacosane, and the dodecylbenzene were consumed first. For these three products, which practically disappeared by the end of the batch, their respective biodegradation rates after 37 Y2 hours only of culture were 93 %, 87.5 % and 80 %. Pristane only started being attacked after 24 Y2 hours, and was 69.2 % consumed at the end of the culture. During the last 20 hours, while practically no alkanes or alkylbenzene remained in the reactor, dimethyl- naphtalene was biodegraded (disappearance rate 67 %). Phenanthrene and dibenzothiophene were consumed very little if at all. These results perfectly confirm those found with Arabian Light Crude, which showed that alkanes and isoprenoids were attacked first, follo- wed by mono- and di-aromatics. Similarly, it was observed that tri- aromatics and sulfur-containing aromatics were only slightly sensitive or insensitive to the action of microorganisms. The fact that dodecyl- benzene disappeared fairly rapidly is explained by the presence of the linear chain which, like the n-alkanes, is readily accessible to bacteria. The total organic carbon concentration (TOC) measured in the medium was 385 mg.1-1. After esterification and evaporation of the organic extract, esters and some other polar compounds were found in a concentration of 221.5 mg.1-1. We carried out analyses by gas phase chromatography and GC/MS coupling in an attempt to identify these products. Since many products were present in trace amounts, and several of them were eluted simultaneously and combined in a single peak, we encountered considerable difficulty in identifying them on the mass spectrometer. 33 2. CONTINUOUS CULTURES In continuous culture, since this technique serves to check the concen- trations of all the nutritive elements at all times, and to adjust these concentrations to limit thresholds, thus closely approaching conditions encountered at sea, we attempted to quantify the nitrogen, phosphorus and oxygen requirements for the biodegradation of given quantities of hydrocarbons present in ALC 240+. The following operating conditions were used: Dilution rate D = 0.04 h- Temperature 200C Reactor volume 2 liters Agitation 520 rpm Ph of culture 8.1 GHSV 1 (except for quantification of the oxygen requirements, where the GHSV was varied from 1 to 0.25). e ALC 240+ concentration entering reactor -1 always about 2.5 g.1 For a given concentration of an element (nitrogen, phosphorus or oxy- gen) entering the reactor, the experimental time was about one week. Upon each alteration in operating conditions, it was necessary to wait another week for equilibrium to be re-established. When the residual concentration of nitrogen was in excess, the bacterial consumption of this element per mg of hydrocarbons degraded ranged from 0.1 to 0.11 mg. However, when the nitrogen reached a limit with residual contents around 1 mg/liter, the nitrogen requirements dropped to 0.07 mg. The same occurence was observed with phosphorus. In conditions of non- limitation, the biochemical consumption of phosphorus,'around 0.012 to 0.013 mg/mg of hydrocarbons consumed, declined to only 0.005 mg/mg of hydrocarbons consumed when the residual concentration of elemental P reached a limit ( 4.1 mg.1-1). With respect to oxygen, microorganism requirements fluctuated between 1.4 and 1.9 mg oxygen per mg of biodegraded hydrocarbons, for residual dissolved oxygen concentrations between 50 and 7 % of the saturation value. 34 THE AMOCO CADIZ ANALYTICAL CHEMISTRY PROGRAM by Paul D. Boehm, Ph. D. Environmental Sciences Divisions, ERCO (Energy Resources Company, Inc.), 185 Alewife Brook Parkway, Cambridge, Massachusetts 02138 TABLE OF CONTENTS 1. INTRODUCTION 2. METHODS AND MATERIALS 2.1 Sediments and Sediment Cores (Extraction and Processing) 2.2 Plant and Animal Tissue 3. RESULTS AND DISCUSSION 3.1 Overall Findings 3.1.1 Weathering of AMOCO CADIZ Oil 3.1.2 Persistence of Marker Compounds 3.1.3 Residues in Tissues 3.1.4 Environmental Variability 3.2 Surface Sediments (Atlas, University of Louisville) 3.3 Offshore Sediments (Marchand, CNEXO) L'Aber Benoit Sediments (Courtot, U. West Brittany) 3.4 Sediment Cores (Ward, Montana State University) 3.5 Oysters and Plaice (Neff, Battelle) 3.6 Oysters and Fish (Michel, ISTPM) 3.7 Seaweed and Sediments (Topinka, Bigelow Laboratory for Ocean Sciences) 4. Conclusions 5. REFERENCES 35 INTRODUCTION All fate and effects studies of oil spills in the marine environ- ment depend on analytical chemical information concerning the distribu- tion and composition of the spilled oil. This includes petroleum hydrocarbon concentrations and compositions in water, sediment, and tissue samples. In turn, this information can be used to deduce the nature of the weathering process (including evaporation, dissolution, and biodegradation), biological assimilation and depuration, and the mass budget of the oil. Thus the analytical chemistry component of the AMOCO CADIZ research program provides crucial information to many other components of the program in the investigation of the time- dependent fate and effects of this spill. During the six weeks following the grounding of the supertanker AMOCO CADIZ on March 16, 1978, oil came ashore along 320 kilometers of the Brittany coastline (Gundlach and Hayes, 1978). Various shoreline types were impacted (e.g., rocky shores, send flats, coastal embay- ments, tidal mud flats and salt marshes). During the early stages of the spill, oil was transported offshore and deposited in the benthic environment. The fate of petroleum residues deposited in these imp act- ed areas was and continues to be affected by coastal processes which dictate such factors as wave energy and sediment transport, and create environments of differing substrate character (e.g., grain size), chemical status (oxidizing versus reducing), and biological activity (e.g., microbiological biomass). All of these factors and others (e.g., light intensity) combine to determine the weathering character- istics of the residual petroleum assemblage. Biological populations initially impacted by the spilled oil may be subject to chronic exposure to petroleum hydrocarbons associated with (and released from) the substrate to which they are closely linked, or they may undergo rapid or slow depuration of initial resi- dues if no longer exposed to oil, via transplantation or due to flush- ing by "clean" seawater. Such differential exposure histories have been previously observed to profoundly affect the spilled oil residual body burdens in marine organisms (Boehm et al., 1982). Although oil spills have received increasing attention from the scientific community during the past decade, there have been few opportunities to examine the chemical compositional changes in beached or sedimented oil in a variety of coastal environments, over a signifi- cant period of time and to examine uptake (impact) and depuration (recovery) of petroleum by marine organisms. A detailed examination of the chemical changes in oiled substrate suggests both the anticipated residence time of deposited oil, and the potential for biological damage of the petroleum residues. Rashid (1974) examined compositional changes of Bunker C oil from the ARROW spill in Nova Scotia at dif- ferent coastal locations. Other than this study only site-specific studies of the geochemistry of petroleum weathering (e.g., Mayo et al.f 1978; Blumer et al., 1973; Teal et al., 1978) have been undertaken. 36 Uptake and depuration by organisms have been the subjects of many laboratory experiments (e.g. Neff et al., 1976; Roesijadi et al., 1978) but relatively few real spill scenarios (e.g. Boehm et al., 1982; Grahl-Nielsen et al., 1978). This report is intended to present an overview of the chemistry program along with enough supporting data and interpretations for each program element to make this a self contained document. After a methods section, a summary of the general findings is presented. Discussions of the analytical chemical and biogeochemical findings of each of the six specific investigations follow; the last section draws conclusions from the study as a whole. Much of the raw analytical data has been omitted.here for brevity. Tabulations of analytical data are available either from the individual principal investigators or from the chemistry group. This data has formed the basis of several publications to date (Calder and Boehm, 1981, Boehm et al., 1981, Atlas et al.# 1981, Winfrey et al. 1981) as well as several manuscripts in preparation. Additional interpretative details are found in these manuscripts. METHODS AND MATERIALS As part of the NOAA/CNEXO research program to examine the long- term fates and effects of the spill, we obtained samples of frozen intertidal surface sediment, benthic sediment, sediment cores, oysters, flatfish and macroalgae from a number of U.S. and French investigators (Table 1). TABLE 1. Summary of AMOCO CADIZ chemistry program. Chemical Composition, Weathering, and Concentrations in Surface Intertidal Sediments (Atlas; Calder): 1978-1981 2 - Chemical Composition, Weathering, and Concentrations in Subtidal Sediment (Marchand, Courtot): 1978-1979 3 - Chemical Composition, Weathering, and Concentrations in % Intertidal Cores (Ward): 1978-1980 4 - Chemical Concentrations and Composition of Oil in Oysters and Flatfish from Abers (Neff): 1978-1980 5 - Chemical Concentrations and Composition of Oil in Variety of Fish and Oyster Tissues (Michel): 1978-1979 6 - Chemical Concentrations and Composition of Oil Associated with Seaweeds (Topinka): 1978-1980 37 2.1 Sediments and Sediment Cores (Extraction and Processing) Samples of surface sediment or specific depth interval sections of sediment cores were solve nt-ex tr acted and fractionated according to an ambient temperature solvent drying and solvent extraction procedure based on that of Brown et al. (1980) as revised by Atlas et al. (1981) and Boehm et al. (1981). The procedure, involving methanol drying and ambient temperature extraction with a methylene chloride/methanol azeotrope, is illustrated in Figure 2.1. The concentrated extract is displaced with hexane and charged to a glass absorption chromatography column (1 cm i.d.) containing 10 g fully activated (1500C) 80-100 mesh silica gel topped with 1 g 5% deactivated alumina and I g activated (i.e. acid washed) copper powder. The column, which is wet packed in methylene chloride, is rinsed with this solvent followed by hexane. A 0.5 ml volume of extract is charged to the column and eluted with hexane (17 ml, f 1) , hexane: me thylene chlor ide (21 ml, f 2) , and methanol (20 ml, f3)- The fractions are collected separately, reduced in volume, desulfurized using an activated (1 N HCI) copper powder slurry, and an aliquot weighed on a Cahn electrobalance. The fl and f2 fractions are then analyzed by fused silica capillary gas chroma- tography (FSCGC flame ionization detector) and a selected set further scrutinized by gas chromatographic mass spectrometry. FSCGC analysis determined the overall composition of the sample by appraisal of the distribution of resolved (peaks) and unresolved (hump) features, as well as the specific quantities of individual n-alkane (ClO to C32) and isoprenoid (CI5 to C20) compounds. GC/MS/computer analyses focused on the list of saturated and aromatic compounds presented in Table 2 to confirm the identities of compounds or to quantify minor, but important "marker" compounds. Details of the GC and GC/MS analytical procedures are presented in Table 3. Quantification of GC traces was according to the internal standard method wherein quantities of individual hydrocarbons are computed. Several other GC-derived parameters were routinely calculated on sample data. One of these was the n-alkane to isoprenoid ratio (ALK/ISO) in the C13 - C19 range: ALK/ISO n - C14 + n-Cly, + n-Clf; + n C17 + n-CIR 1380 + 1450 + 1650 + 1710 + 1812a aGC retention indices of isoprenoids: 1450 = farnesane, 1710 pristane, 1812 = phytane. This ratio, beginning at -.7 in the reference oil is quickly decreased due to preferential bacterial degradation of n-alkanes versus the branched isoprenoids. The carbon preference index (CPI), the ratio of odd chain alkanes to even chain alkanes in the n-C26 to n-C31 range, is defined as follows: 2(n-C27 + n-C9q) CPI n-C26 + 2n-C;28 + H-C30 38 Sediment sarriole Metmariol Cried Secirrient (1) 14aCl ;saturated) (2) Acidity wil,lci 0) 509 ;n 7aflom ;ar or twrrifuq* uoo @3) Extra=. 3X -,v/C'-i2c:z ,2) Imternas standards i@ 3) C43OH1,CH2C:2 (1,11 41 rlusn wiN2 (5) Shake it arntlient -,emoeraturd 40 loun Mth solvent criangs ifter IS and 24 mour3 CH2C7 and M01hyleris Sediment CH:30H/CH-ZC12 C-hloride Extracu CH.ZC:2' Ciscard (1) Corriolne Oiscare Z OrY over N&'ZS04 @3) Concentrate to 100 u, - 141 *%gn Concentrated Extract 1) Disajace witm H*xans Alurninai silica Gal C,jiurnn Chromatogrsariv Hexane @fj I HexanetMOTMylene 'Iletmarl0i -3) Chloride ff2s iWeigh AliauoO ;'Neign Alicuotl @Weiqn Alicluctl Saturated Aromatic Polar NSC, r4ycrocafbons r4ydrocar=ni CormooUrids I i i (3C2 Store (3C2 MS GC2 GC2 pjS FIGURE 2.1. Analytical scheme for sediment samples. 39 TABLE 2. Focus of GC/MS analyses. - ---------- - -------- - --------- Saturated hydrocarbons Pentacyclic triterpanes (hopanes) Aromatic hydrocarbons Alkylated benzenes (C4, C5, CO Naphthalene and alkylated naphthalenes (Cl, C2, C3, C4) Fluorene and alkylated fluorenes (Cl, C2, C3) Phenanthrenes and alkylated phenanthrenes (Cl, C2, C3, CO Fluoranthene Pyrene Benzanthracene Chrysene Benzofluoranthenes Benzo(a)pyrene Benzo(e)pyrene Perylene Aromatic heterocyclics Dibenzothiophene and alkyl dibenzothiophenes (Cl, C2, C3) TABLE 3. GC and GC/MS conditions. GC GC/MS/COMPUTER A. Instrument HP 5840A HP 5985 B. Column SE-30 (saturates) SE-52 1. Liquid SE-52 (aromatics) phase 2. Type Fused silica Fused silica (J&W Scientific) (J&W Scientific) 3. Diameter 0.25 id 0.25 id 4. Length 30 m 30 m 5. Carrier Helium @ 1 ml/min Helium @ 1 ml/min C. Temperatures 1. Oven 40-290 @ 3*/min 40-290 @ 3'/min 2. Injector 250* C 250' C 3. Detector 300- C (FID) 300 (ion source) D. Ionization - 70ev voltage E. Electron 2200 volts multiplier voltage E. Scan conditions 40-500 amu @ 225 amu/sec (1 scan/2 seconds minimum of 5 spectra per peak) 40 The CPI ranges from values of 1, where oil is present, to values greater than 1 if odd chain biogenic terrigenous n-alkanes dominate the higher boiling n-alkanes. Quantification of aromatic hydrocarbons was accomplished using the technique of mass fragmentography wherein the computer stored raw GC/MS data is searched for parent ions (m+) and the total ion currents for these ions is integrated and tabulated. Retention times of the parent ion mass fragmentograms obtained were compared with authentic standards. The total ion current for each parent ion is compared with that for the internal standard (deuterated anthracene) and instrumental response factors applied. Where authentic polynuclear aromatic hydrocarbon (PAH) standards were not available for relative response factor determination, a response factor was assigned by extrapolation. All of the above techniques were applied successfully to the analyses of replicates+ of a NOAA intercalibration , sediment sample, Duwamish I, prior to commencement of the program and to Duwamish II during the program. Additionally, the EPA "megamussel" intercalibra- tion sample was successfully analyzed for PAH levels. 2.2 Plant and Animal Tissues All specimens of wet tissue, freeze dried tissue, and plant material were thawed and homogenized, or in the case of the seaweed tissue were cut into small pieces, prior to placement in a digestion flask. The samples were added to 250 ml Teflon screw top jars. The digestion, extraction, and fractionation schemes were similar to those developed by Warner (1976) except that the digestion was performed using a 0.5 N KOH/distilled water/distilled methanol system heated in a boiling water bath for 4 hours to achieve complete digestion and hence release of hydrocarbons from the cellular matrix (Boehm et al., 1982). Internal standards were added prior to digestion and carried through the entire procedure (Fig. 2.2) (f]. = androstane; f2 = deuterated anthracene or phenanthrene). The digestate was extracted three times with distilled hexane in the jar, the mixture being centrifuged between extractions. The extracts were combined, concentrated to 0.5 ml, weighed on a Cahn electrobalance, and fractionated on an alumina over silica gel column (see previous section). Two fractions corresponding to the saturated or fl (hexane eluate) and the aromatic/olefinic or f2 (hexane:methyl- ene chloride eluate) hydrocarbons were obtained for gas chromatographic and combined gas chromatographic/mass spectrometric analyses (GC/MS). 41 r,ssue Samole 'a 50 grams Wet) 11) Dissect 9:1esil 12) Homogenize (3) Add to Teflon Jar 141 Add internal Hydrocarbon Stancaras (51 4N KOH (aal (6) Flush Jar with N2 (7) Sam 1 (8) Digest (Saoonifv I Overmign- at Room Terricierature Digestate ill Transfer to Seciarstorv Funne! (2) Add Saturwed NaC; 11) Extract 3 Times vvilh Hexane Combined Hexane Extracts (1) Concentrate (2) Dry Over Na2SO4 Concentrated Extract (1) Alumina Clannuo 12) Metriviene Chloride Elution (3) Oisoiace with Hexane Alumina/ Silica Gei Column Chromatograniny Hexaneifl) Hexanili/MOCIV) Metnanol (f3) (Weigh) Meigh) (Weign) (1) Re"Ponify or (2) Gal Permeation HPLC Cleanuo Saturated (1) Aromatic Hydrocarbons Pow NSO Hydrocarbons M PCs Comaounas Gc2 GC2 Store Gc2/mS Fioure 2.2. Analytical scheme for tissue samples (after Warner, 1976; Boehm et al., 1982). 42 3. RESULTS AND DISCUSSION 3.1 Overall Findings Several general trends in the data presented in the following sections should be noted here along with several considerations of the use of marker compounds as "fingerprints" to trace aged AMOCO CADIZ oil in environmental samples. 3.1.1 Weathering of AMOCO CADIZ Oil The chemical composition of spilled oil from the tanker changed markedly over the first days to weeks, both at sea and once associated with sediment (Atlas et al., 1981; Calder and Boehm, 1981; Boehm et al., 1981). The changes are well documented in Figures 3.1 and 3.2 and are summarized in Table 3A. For comparison, the background saturated and aromatic hydrocarbon composition of sediment samples is illustrated. The non-impacte8 sediments contain.- 1) an unresolved complex mixture (UCM) of hydrocarbon material in both fractions, 2) terrigenous n- alkanes (odd chain) in the saturated fraction, and 3) pyrogenic PAH compounds in the aromatic fraction. Weathered AMOCO CADIZ oil is identified as such in the sections that follow based on the following: 1) The presence of large UCM in f, and f2 fractions with residual triterpenoid peaks. 2) The presence of isolated isoprenoid hydrocarbon compounds in the resolved (peak) part of the GC trace (in samples during the first year post-spill only). 3) The dominance of alkylated phenanthrene (C2, C3, C4), di- benzothiophene, naphthalene, and fluorene (in earlier samples) compounds in the aromatic fraction and a dominance of these aromatics versus pyrogenic PAH (i.e. fluoranthene, pyrene, benzanthracene, chrysene, benzopyrenes, etc.). 3.1.2 Persistence of Marker Compounds The most persistent compounds in the saturated (fl) fraction are the pentacyclic triterpanes (PCT); in the aromatic (f2) fraction the alkylated phenanthrenes (P) and dibenzothiophenes (DBT) are most persistent. To examine the PCT compound distribution, GC/MS analysis of the fl fraction was necessary (e.g. Figs. 3.3 and 3.4). This results in a "terpanogram" yielding information on the relative concen- tration of eight PCT compounds used by several investigators as indica- tors of presence and origin of petroleum (e.g. Dastillung and Albrecht 1976; Pym et al., 1975). Two PCT time series (Fig. 3.5) reveal that the PCT fingerprint is rather constant throughout the December 1978 to March 1980 time period 43 Ax ljCM C STAGE 2 4EA"HERINr SA- it 0 STAGE 3 NEAT@ERING 4-- W,d- lj Cm L I E STAGE 4 WEATMERING !S-- H--, UC. F BACKGPC@NC S11,11- A---' it FIGURE 3.1. Weathering patterns of saturated hydrocarbons in AMOCO CADIZ oil. 44 1A REFERENCE MOUSSE (A,o@aml z t7z 8 STAGE I WEATHERING tAromavu) AlkV t OBT UCM C STAGE 2WEATHERING iAromavul ucu 0 STAGE 3 4EATmERING (Aromaticsi :2 UC111 E PYROLYTIC PAH SOURCE @A,o-vc,_ FL-Fluorantren. z Z I PYAaPY-4 C -Ch" Y8#nzof1uor,n,1,.,. EIFEP, BAP-B,nmov,-, B-Be zofl,orerd @A -A. FIGURE 3.2. Weathering patterns of aromatic hydrocarbons in AMOCO CADIZ oil. 45 TABLE 3A. Weathering of AMOCO CADIZ oil. RAPID 1. Loss of volatile (<n-Cl5) hydrocarbons due to evapora- tion of: a. alkanes b. aromatics - benzenes, naphthalenes, biphenyl (one- to two-ring aromatics) 2. Relative and/or absolute increase in unresolved complex mixture. MODERATE 1. Microbial degradation of n-alkanes; preferential attack of n-alkanes versus branched alkanes (i.e., decrease in ratio of alkanes to isoprenoids). Loss (to solution or other processes) of most resolved saturated hydrocarbon GC traces. 3. Emergence of triterpanes as major molecular markers in saturated fraction. 4. Increase in UCM, with formation of secondary (bimodal) UCM distribution. 5. Loss of fluorenes (two aromatic rings, one saturated ring) and alkyl naphthalenes. 6. Increase in abundance of polar fraction. LONG 1. Persistence of alkylated phenanthrenes and alkylated dibenzothiophenes. 2. Increase in polar fraction. 3. Loss of long chain n-alkanes and isoprenoids. 46 T- T[r ''VW SAY OF MOPLAIX SEDIMENT FIGURE 3.3. GC/MS selected ion searches for pentacyclic triterpanes (ho- panes) in AMOCO CADIZ reference and November 1978 weathered oil in sediments. 3 4 5 7 191.0 10TAL ION TI 94 G6 G'S S' -6 '3 '4 7 9 69 70 71 72 7 7 7 FIGURE 3.4. Triterpane (m/e 191) mass chromatogram of weathered oil. 1,2 = Trisnorhopane (C27H46), 3 = Norhopane (C29H50), 4 Hopane (C30H52), 5,6 = Homohopanes (C31H54), 7,8 Bishomo- rL hopanes (C23H56). 47 A, I... ,,, PA ...... DPCI, I,;,, 197H .1,.Iy Iqj9 1, P IMP Ll I I I I I I I 1 2 1 4 5 6 7 R 1 2 3 A 5 6 1 8 1 2 3 4 5 G I R 1 3 4 5 6 7 A STATION 3- ILE GRAND MARSII [email protected],-,. [email protected]@ 1978 1979 M.-h 1900 1 7 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1 7 3 4 5 6 7 R 1 7 :1 4 5 r, 1.8 STATIONG- AREA WnAC'If FIGURE 3.5. 191 terpanograms. (1,2 = Trisnorhopanes, 3 = NorhopanO, 4 Hopane, 5,6 = Homohopanes, 7,8 = Bishomohopanes) at the two stations. Note that the PCT fingerprints of AMOCO CADIZ and TANIO oils are quite distinct (Fig. 3.6), notably in the ratios of compounds 1, 2 and 5, 6. Thus it appears that PCT fingerprints offer a good means to trace AMOCO oil in highly weathered samples when most other identifiable molecular characteristics have been lost. 1 2 3 4 5 6 1 a 1 2 3 4 5 1;7 n AMOCO CADIZ I ANIO STA I ION I I FIGU1,Rx', 3.6. 191 terpanograms of crude oils. 48 As the most persistent aromatic compounds, the P and DBT compounds (mainly C2, C3, and C4) mark AMOCO CADIZ oil in tissue (oysters) and sediments through mid-1980. The final (June 1981) sediment sampling failed to reveal significant P and DBT levels in any of the stations. The latest (1981) status of the oyster P and DBT levels is unknown. However, through most of the data to be discussed, the P and DBT compounds dominate the f2 distribution. The ratios of C2P/C2DBT and C3P/C3DBT, used by Overton et al. (1981) to differentiate oils, in this spill remain in the 0.3-0.6 range. The use of this ratio is discussed in the text. 3.1.3 Residues in Tissues As stated, the P and DBT compounds are most readily associated with oyster tissue samples in the two-year period following the spil- lage. The branched alkanes (isoprenoids) also persist throughout this period. 3.1.4 Environmental Variability A major question in oil spill studies and for that matter environ- mental studies in general is the question of patchiness of pollutant distributions and the variability due to patchiness in chemical meas- urements. To shed some light on this subject two sets of measurements are available. Two principal investigators (Atlas and Ward) obtained samples at the same time and location in several instances, Atlas sampling the top 3-5 cm, Ward sampling an entire sediment core but subdividing the top 0-5 cm section. The total hydrocarbon values (Table 4) reveal wide disparities where contamination is very heavy (pooling of oil in the Ile Grande) but reasonable to excellent agree- ment in most cases. (Note also that additional replicate analyses are available for sediment samples in Section 3.2 as well). TABLE 4. Analysis of sampling variability. TOTAL HYDROCARBONS (fl + f2) (Pg/g) STATION DATE ATLAS (SURFACE) WARD (0-5 cm) Ile Grande 12-78 650/1300a 1,100 3-79 4,700 700 L'Aber Wraclh 12-78 400/400a 770 3-79 870 1,100 7/8-79 390 290 11-79 1,100 1,100 aReplicate samples. 49 3.2 Surface Sediments (Atlas, University of Louisville) The frequency of sampling for surface sediment (0-3 cm) is shown in Table 5. Ten primary locations were sampled repeatedly (Fig. 3.7). Results of total hydrocarbon determinations for the ten stations, as these concentrations varied with time, are presented in Figures 3.8 through 3.12. Also included in these figures are source evaluations for each sample hydrocarbon assemblage, based on GC information. The biogenic (B) category indicates that terrigenous odd chain n-alkanes dominate the fl. GC trace. The pyrogenic (P) category signifies an important abundance of comb ust ion-r elated polynuclear aromatic hydro- carbons (PAH) in the f2 fraction as well as the presence of some unresolved material (UCM) in both the f]. and f2- In those samples labeled B or B/P the primary sources of hydrocarbons are as indicated although a small fraction of the hydrocarbons may consist of petroleum. Figure 3.13 summarizes these source criteria. Only GC/MS analysis of each sample would definitely eliminate the small chance of a false negative (i.e. not finding AMOCO oil where there were traces). The error bars in the figures indicate that two determinations were made for the December 1978 samples (Table 6). All other determin- ations were based on one replicate. Note that the coefficient of variation ranges from .01 (1%) to .94 (94%). The higher variability is observed in samples with the lowest and highest ('@,1000 ppm) absolute concentration levels, the former due to natural patchiness, the latter owing to "pooling" of oil in heavily impacted stations. GC/MS results are available for stations 3, 5 and 7 throughout the study period and are presented graphically in Figures 3.14 through 3.33. These semi-log plots illustrate quantitatively the aromatic composition of all samples normalized to C3 dibenzothiophene or where C3DBT is absent to pyrene. C3DBT was used to normalize the data as it is assumed that these compounds are the slowest to weather of all of the aromatic hydrocarbons. All AMOCO CADIZ-impacted stations illustrate a normal weathering sequence (i.e. see Fig. 3.1). However, fresh inputs of petroleum were observed to impact the region of stations 7 and 8 in the form of tar chips during November 1979 and stations 2, 11 and 12 in the form of oil from the TANIO spill in August of 1980 (Fig. 3.34). Although a wide range of residual oil concentrations appear in the various samples, several trends in the data seem apparent. Stations 1, 9, and 10 remain unimpacted by the spill throughout the study. Station 2 remains unimpacted until a secondary petroleum input influences its hydrocarbon chemistry in November of 1979 (the timing of the secondary tar impact at stations 7 and 8 also is probably related to leakage from the sunken tanker) and again in August of 1980, the latter relating to the TANIO spill, also readily detected at Stations 11 and 12 at this time. Through March of 1980 weathered AMOCO CADIZ oil is readily detected at Stations 3, 4, 5, 6, 7 and 8. However, the results of the August 1980 samplings indicate that inputs of non-AMOCO CADIZ hydro- carbons (i.e. background) at Stations 6 and 8 become dominant. At stations 3 and 7 where GC/MS data exists, the main AMOCO CADIZ aromatic 50 L E,.(J*.: I AN I I E r S f.M113 It, L FN 1;11 F. V k (41 CARER WRACII MAI 1q. 14, FIGURE 3.7. Surface sediment'sampling locations (Atlas). TABLE 5. AMOCO CADIZ chemistry program; surface sediments (Atlas). Frequency April 1978-October 1978 (Calder) December 1978 20 March 1979 10 July 1979 12 November 1979 is Karch 1980 11 Kay 1981 12 Total 80 Locations Ten Primary Stations 1,2,3 Ile Grande 4 St. Michel-en-Greve 5,6 L'Aber Wraclh 7,8 Portsall 9,10 Trez-Hir 11-14 Other Impact Stations GC/?4S Stations 3 (Ile Grande), 5 (L'Aber Wrac1h), 7 (Portsall) 51 STATION I ILE GRANDE STATION 3 191K -2Q3K .17K -3K ILE GRAND CILEO@ 39K loo a,ow SIOGENIC 8 : DIOGENIC so B,P PYROGENIC X8RONIC) 8.oOo PPyRoGENIc IcNRoh,cl AC - AMOCO OIL AC AMOCO OIL s.ow - 4 EjP Ac Q 4,ow AC STATIZ@ 2 1 )LErRANDE STATION 4 ST. MICHEL EN GREVE 2so 2SOO @C--, 2W AC P" 200 N@0,1 'TANIGI 16o AC 120 a,, AC so a P 90 AC c C.e P 'I0 a P AC . I . I . . I . I I I I . . I I 12,78 3,79 7 'a 11 79 3.W 8,80 6 a, 12@711 3.79 1'7. 1 3,LW &'80 1 81 FIGURE 3.8. (Left) Ile Grande (control) sediment time series. FIGURE 3.9. (Right) Ile Grande (oiled) and St. Michel-en-Greve time series (note scale difference between the two plots). _STATION S '- A3Eq VRAC m STATION 7 AC RORTSALL @ALDER-A 1.000 - -IOL-E-100 -CES 'W ATLAS-S B:8.0 E.11C B : 3IoGs.IC ODD - , PYROGENIC CHRONIC, t6o - -qOGENIC C@AONIC, AC @C --OCO 0-1. - AC AC AC - AMOCO OIL 120 - AC - AC 400 @c C 2oo - Ac 40 - Ac a P P am Ac 'I,C L ABER MAAC 4 _STATION B "TSALL CALD P-a AOLFE-.OOPACES Soo - ,T,AS-j 1 1-0 S- @dEA I-C @c @c :Im - AC 300 - AC Ac 6W 4. 20a - AC 3 @C ,t is i, 79 3 so 3 sc 8: 12178 3,T9 7 79 11"79 3@w ad ao 6d8I FIGURE 3.10. Aber Wrac1h sediment time series (left). FIGURE 3.11. Portsall sediment time series (right). / -C Al AC AC @Al . @P -C --.O.El AC AC AMOCO C -C -MoC. "-L @A C AC AC . @I@C c \V@ .C. 52 STATION 9 ISO TPEZ miR 8,P 100 so sip '0 8GIOGENIC 20 sip PPYROGENIC JCMPONIC) AC AMOCO OIL 9TION IQ sm 1:0 TAEZ N1 A 100 B,P so S.P a/P .0 - 20 S,P 12,76 3,79 8;79 11'75 3 90 S, w a, III FIGURE 3.12. Trez Hir sediment time series. AC LJ AC .......... . FIGURE 3.13. Hydrocarbon compositions forming the basis of source classification categories. P 0. --C .-S-7 - P sip ' /'@- @A.IN;O I - P .1111, - @P a/ 53 TABLE 6. Replication of hydrocarbon concentration data (based on December 1978 analyses of two replicates). STATION R 0-/K 1 56 0.57 2 113 0.08 3 1,000 0.52 4 135 0.60 5 401 0.01 6 217 0.06 7 159 0.10 8 358 0.21 9 18 0.71 10 11 0.94 2,130 ng, g 1.0- A A B C D E F G H I J K L M N 0P0 R S T U V W Z AA FIGURE 3.14. Aromatic hydrocarbons, station 3, December 1978; normalized to C3DBT. (A = napthelenes, B = CjN, C = C2N, D = C3N, E = C4N, F biphenyl, G = fluorenes, H = CIF, I C2F, J = C3F, K = phenanthrenes, L = ClPh, M = C2Ph, N C3Ph, 0 = C4Ph, P dibenzothiophenes, CjDBT, R = C2DBT, S = C3DBT, T = fluorene, U = pyrene, V benzo(a)anthracene, W = chrysene, X = benzofluoranthene, Y benzo(a)pyrene, Z = benzo(e)- pyrene, AA = perylene) 54 1.0- 11,000ng/9 I-r A 0 C 0 E F G H I J K L M N 0 P 0 R S T U V W X Y Z AA It It IL 410ngig 1.0- A 8 C D E F G H 1 J' A L M A U' V' X Y Z AA 1.0- I I T A 6 C 0 E F G H j K LM . 0 0 R S T U V W X Y Z AA FIGURE 3.15. (Top) Aromatic hydrocarbons, station 3, March 1979; normal- ized to C3DBT. FIGURE 3.16. (Middle) Aromatic hydrocarbons, station 3, July 1979; nor- malized to C3DBT. FIGURE 3.17. (Bottom) Aromatic hydrocarbons, station 3, November 1979; normalized to C3DBT. (See Figure 3.14 for key.) 55 6,10ong/g 1.0- 'P A 8 C 0 E F G H I J K L M N 0 P 0 A S T U V W X Y. Z AA 1.0- ?7nq/jp T T I i r ABCDEFGH:j'K'LMN'O'P'O'A'S'T UV wx YZAA 1.0- 5346.4 mg/g I T I A 8 C 0 E F G H I J K L M N 0 P 0 R S T U V W X Y Z AA FIGURE 3.18. (Top) Aromatic hydrocarbons, station 3, March 1980; normal- ized to C3DBT. FIGURE 3.19. (Middle.) Aromatic hydrocarbons, station 3, June 1981; nor- malized to:C3DBT. FIGURE 3.20. (Bottom) Aromatic hydrocarbons, station 5, April,1978; nor- malized to C3DBT. (See Figure 3.14 for key.) 56 4,160 n919 1. r I T T r T r A 6 C 0 E F G M I J K L M N 0 P 0 R. S T U,v w x I Z AA 1.0- 480 ng/g. 8T 0 9 A 6 C 0 E F G J K L M I C) P Q S T U V W X V Z AA 1.0 880.gI/g 1P A C E F G I R S T U V x Y Z AA FIGURE 3.21. (Top) Aromatic hydrocarbons, station 5, October 1.978; nor- malized-to C3DBT. FIGURE-3.22. (Middle) Aromatic hydrocarbons, station 5, December 1978; normalized to C3DBT. FIGURE 3.23.. (Bottom) Aromatic hydrocarbons, station 5, March 1979; nor- malized to C3DBT. (See Figure 3.14 for key.) 57 1.0 94.9/9 A D E I J K L M N 0 P 0 A S T 0 V W X Y Z AA 1.0 460 nq@q A 6 C 0 E L M N 0 P 0 R S T U V W X Y Z AA IL 1.0- A 8 C 0 E F G H I J K L M N 0 P 0 A 5 T U V W X Y Z P@ I I It 11 __j FIGURE 3.24. (Top) Aromatic hydrocarbons, station 5, July 1979; normal- ized to C3DBT. FIGURE 3.25. (Middle) Aromatic hydrocarbons, station 5, November 1979; normalized to C3DBT. FIGURE 3.26. (Bottom) Aromatic hydrocarbons, station 5, March 1980; nor- malized to C3DBT. (See Figure 3.14 for key.) 58 1.0 ngj ABCDEFG'H'I'J''K'L4'NOP@RSTUV@X'Y'Z'AA C3D8T-280nq/q A 8 C 0 E F H I J K L M N 0 P 0 A S Zu t 1.0- C30BT 130 ngg T G H I J K L A N 0 P 0 R S T U V W X Y I AA FIGURE 3.27. (Top) Aromatic hydrocarbons, station 5, June 1981; normal- ized to pyrene. FIGURE 3.28. (Middle) Aromatic hydrocarbons, station 7, December 1978; normalized to C3DBT. FIGURE 3.29. (Bottom) Aromatic hydrocarbons, station 7, March 1979; nor- malized to C3DBT. (See Figure 3.14 for key.) 59 1.0- C30OT:177ng/g A B C 0 E F G H I i KL M N@O P 0 R S T.U V W X Y Z AA COBT-75n9/9 A 8 C 0 E F G H K' M N 0 P 0 R S T U V W X Y Z AA L -11ng,g C3D ST A B C 0 E F H I J K L M N 0 P 0 R S T U V W X Y I A@ FIGURE 3.30. (Top) Aromatic hydrocarbons, station,7, July 1979; normal- ized to C3DBT. FIGURE 3.31. (Middle) Aromatic hydrocarbon s,. station 7, November 1979; normalized to C3DBT. FIGURE 3.32. (Bottom) Aromatic hydrocarbons, station 7, March 1980; nor- malized to C3DBT. (See Figure 3.14for key.) 60 220ngio 1.0- A B C 0 E F G H J K L M N 0 P 0 R S 7 U V W X Y Z AA FIGURE 3.33. Aromatic hydrocarbons, station 7, June 1981; normalized to pyrene. (See Figure 3.14 fo-r key.) A- SATURA FES A 8 AROMATICS, @-)'V PW. ae@4 FIGURE 3.34. TANIO oil. 61 marker compounds, the C3 dibenzothiophenes, and C3 and C4 phenanthrenes, persist but the pyrogenic PAH compounds have replaced any AMOCO CADIZ oil traces at Station 5 in L'Aber Wrac1h. The last sampling, June 1981, reveals total disappearance of traces of AMOCO CADIZ aromatic marker compounds at stations 3, 5, and 7. By June 1981 the only unequivocal presence of AMOCO oil is seen at station 3 in Ile Grande, although it has been extremely weathered. Only pentacyclic triterpanes can be linked to the residual AMOCO oil. GC patterns suggest that petroleum still affects stations 4, 6, 7, and 8, but in only minor quantities relative to other inputs. Thus, for the most part, less than three and one half years has been required to allow normal background inputs to resume their sedi- mentary dominance at all but the most heavily impacted (in terms of post cleanup oil concentrations) and lowest energy (i.e. most protected from waves) environments (i.e. station 3 in the Ile Grande). further interpretive details are presented in Atlas et al. (1981). 3.3 Offshore Sediments (Marchand, CNEXO) L'Aber Benoit Sediments (Courtot, U. West Brittany) In this phase of the analytical chemical program the levels, the persistence, and the precise chemical nature of petroleum hydrocarbons in the offshore sediments of the Bays of Morlaix and Lannion were examined as well as those of L'Aber Benoit sediments (November 1978 only). A summary of the samples analyzed appears in Table 7' and in Fiqure 3.35. TABLE 7. AMOCO CADIZ chemistry program; 2. Offshore surface sediments (Marchand) and Aber Benoit sediments (Courtot). Frequency: April 1978 6 July 1978 14 November 1978 13+7 February 1979 13 TOTAL 53 Locations: Aber Benoit (November 1978) Baie de Morlaix Baie de Lannion GC/MS: Four Time Series (18 Samples) 62 LANNION WRLAIX F A* CA.E. E...' V FIGURE 3.35. Offshore surface sediment and l'Aber Benoit sampling locations (Marchand, Courtot). Hydrocarbon concentrations and source class 'ifications for the entire data set are shown in Table 8. Individual aromatic hydrocarbon determinations by GC/MS appear for several time series in Tables 9 through 13 and for two of the L'Aber Benoit samples in Table 14. An instructive way of viewing the time series information is presented in Figures 3.36 and 3.37. At both the Terenez/ morlaix and Ile Grande time series, concentrations increased between April and July 1978. In the case of the Terenez samples, the increase is due to offshore transport of weathered oil as evidenced by 1) an increase in absolute concentrations, 2) a decrease in the ALK/ISO ratio, and 3) an increase in phenanthrenes (total P, Cl, C2, C3, C4P) and dibenzothio- phenes, without an accompanying increase in the pyrogenic PAH compounds (m/e 202). However, the Ile Grande benthic samples show an increase in total hydrocarbons along with increases in the aromatics including the pyrogenic PAH. This latter finding indicates that both petroleum hydrocarbons and combustion-related PAH material are being transported to and deposited in the offshore sediments near Ile Grande by a similar mechanism, most likely in association with suspended matter from riverine plumes. Figure 3.38, a plot of phenanthrene and its alkyl homologues at the Morlaix Station (Station B), reveals that while the source of the phenanthrenes is petroleum in July 1978, as evidenced by the greater abundance of alkylated compounds versus the parent (unsub- stituted) compounds, the input in February of 1979 is largely pyrogenic (i.e. greater amounts of parent phenanthrene). This illustrates both the usefulness of detailed GC/MS-derived data and their subsequent presentation in alkyl homologue distribution plots. 63 TABLE 8. AMOCO CADIZ sediment sample,-source classification (GC) (offshore sediments). ----------------------- ------------ ------ ---- SAMPLE '40. FL F2 . Ng"g) .............. AC100 4/58 84.1 3/4 90.4 AC 3 6 9 3/5B 252.9 3 3, 172.5 AC426 4/59 35.5 3 133.6 ACI 0 3 4,'2/ 58 200.6 4,12 397.0 AC 3 6 5 4/58 107.9 4/2 503.4 AC429 58/4 37. 3 5/3 31.6 AC42 1 109.0 2/1 123.7 AC138 2/4 408.8 2/4 502.8 AC37L 4/2 19.4 2/4 97.7 AC453 2/4 L42. 5 2/4 36.0 AC56 2 47. 2 2 86.0 AC139 4 5.9 4/2 L7. 2 AC381 2 54.4 2 L19. j AC458A 4 5 8.1 AC4585 4 4.4 5 7.9 ACL27 2 36.9 2/4 58.5 AC370 4/2 18.6 3/4/2 48.B AC452 2/4 13.8 4 41.5 AC132 2/4 165.8 4/2 211.7 AC362 2/4 24.5 4/2 44.7 AC141 2 L5.6 2 39.2 AC396 4/5B 6.0 3/4 13.6 AC432 4 35.4 4?2 26.9 AC134 58/4 78.3 2/4 171.9 AC451 4/2 42.2 4/2 50.7 AC44 1 38.5 2 69.1 AC121 2 17.5 .2 44.8 AC384 2 32.1 2/1 173.6 AC112 2/4 122.5 2 239.3 AC389 2/4 56.4 2/4 151.1 AC445 2/4 29.0 2/4 64.8 AC107 2 53.2 2/4 176.7 AC376 2/4 5.9 2/4 8.9 AC436 2 27.7 2 60.6 AC53 1 28.9 2 31.5 AC118 2/4 B2.2 2/4 97.2 AC377 4 31.9 4 132.1 AC438 2/4 15.8 4 32.1, AC51 1/2 96.4 1/2 L02.4 AC114 2/4 56.4 2/4 249.6 AC379 2/4 27.2 2/4 80.7 AC440 2 103.5 2/4 104.5 AC48 2/1 21.0 4@ 14.3 AC125 2 59.1 2 27.8 AC378 2 63.3 2 191.0 AC479 2 36.5 2 37.9 AC40 and 50 series sampled 4/78 AC100 series sampled 7/78 AC300 series sampled 11/78 AC400 series1sampled 2/79 L'Aher Benoit Sediments: ABT 2 158.2 2 180.8 ACC 4 25.1 4 29.3 AB25 2 29.2 2 29.7 AB29 4 13.6 4 17.5 AB16 2 22.2 2 28.6 AB21 2 455.1 2 440.7 AB4 2 75.6 2 76.9 -------------------- 64 TABLE 9. Terenez/Morlaix time series (9tation A). AGIPHATICS (pg/9) AROMATICS (pg/g) SAMPLE DATE TOTAL RESOLVED ALK/ISO TOTAL RESOLVED AC 42 4/78 109.0 6.1 0.79 123.7 7.4 AC 130 7/78 408.6 11.7 0.10 502.8 17.2 AC 371 11/78 19.4 1.3 0.10 97.7 2.4 AC 453 2/79 142.5 2.0 0.51 36.0 1.6 NCIN C2M CIN CO F ClP C2P C3r P CIP c2p C3P CO AC 42 nd 10.3 40.9 115.5 161.0 16.8 20.1 S4.3 196.9 125.7 95.4 1S3.3 274.9 138.3 AC 136 2.9 6.3 62.5 423.2 730.7 nd 37.4 230.2 905.8 20.4 151.4 593.5 1678.9 882.9 AC 371 nd nd nd nd nd nd nd 4.4 88.0 7.4 9.0 12.4 41.9 27.6 AC 451 2.1 4.6 9.7 10.6 18.8 nd 1.0 7.8 14.8 5.2 5.9 is.0 22.0 21.9 DST ClDBT C2D8T C3DBT FL PYR CHRY BF MOP B(a)P PERL AC 42 19.2 113.6 714.4 1086 195.0 171.4 265.7 298.9 83.5 64.8 24.S AC 138 12.4 383.9 3363.0 6388 3.0 47.1 77.4 120.9 87.2 4S.4 20.3 AC 371 nd 4.0 54.8 179.0 9.1 7.6 16.4 .20.9 9.8 3.7 1.8 AC 453 2.3 6.7 63.8 93.5 4.8 4.2 9.6 9.1 6.0 3.4 1.5 KEY3 nd - none detected. N - napthalene, CI-C4M - alkylated naphthalenes, F fluorene, Clr-C3r alkylated fluorenes. P -.phenanthrene, Cl-C4P - alkylated phenantbrenes, DST - Dibenwthiophene, ClDBT-C2D6T alkylated dLbenzothiophenes, Fl - fluoranthene, PYR pyrene, CHRY C@ysene, SP benzofluoranthene, B(a)P Senzo(elpyrene, B(a)P Senvoia)pycene. PERL perylene. TABLE 10. Morlaix time series (station B). hLIPHATICS (jjglg) AROMATICS (pq/q) SAMPLE DATE TOTAL RESOLVED ALK/ISO TOTAL RESOLVED AC 103 7/78 200.6 12.6 2.5 397.0 7.4 AC 365 11/78 107.9 S.3 5.0 503.0 7.5 AC 429 2/79 37.3 2.0 0.02 31.6 11.9 N CIN C2N C3NC4N FCLF C2F C3r P Cip C2P C3P c4p AC 103 17 18.9 48.6 135.9 220.7 16.3 23.*S 72.8 301.0 117.8 95.4 160.1 312.S 273.9 AC 365 9 14.4 33.8 55.0 111.2 9.9 14.9 36.9 81.3 114.'i 65.3 58.9 143.3 103.6 AC 429 6 10.0 19.7 28.0 50.@ 16.8 11.6 23.3 7.4 185.6 108.6 52.6 26.0 34.9 DST CIDOT C2DDT C3DST FL PYR CHRY BF B(e)P 8(&)P PEAL AC 103 15.2 100.8 772.3 1348 240.2 203.8 345.2 350.7 146.4 L66. 3 87.0 AC 365 13.2 32.7 235.2 440.3 202.2 175.6 254.0 324.6 115.0 130.9 65.4 AC 429 11.3 7.7 10.3 5.0 325.9 302.9 207.4 202. 5 112.5 115.0 37.7 KEY: nd none detected. " - napthalene, CI-CO - alkylated naphthalenes, F - fl uorene, CIF-C3F alkylated fluoreneg, P phonanthrene, Ci- C4P .alky lat ed phenanthre@@s, DST - Dibenzothiophene, ClrJBT-C2DBT alkylated dibenzothiaphenos, Fl - fluoranthene, PYR pyrene, CHRY Crysene, BF henzofluoranthene, B(P)P Benzo(e)pyrene, B(a)P Denzoja)pyrene, PERL pex ylene. 65 TABLE 11. St. Michel en Greve/Lannion time series (station C). ALIPHATICS (pq1q) AROMATICS (pg/9) SAMPLE DATE TOTAL RESOLVED hLK/IS0 TOTAL RESOLVED AC 44 4/78 38.5 1.9 0.38 69.1 6.6 AC 121 7/78 17.5 0.2 0.09 44.3 0.6 AC 384 11/79 32.1 0.5 0.14 173.6 1.6 N C1N C2N C3N C4N P CIF C2F C3F p C1p C2P C3P C4P AC 44 8.6 6.3 12.6 25.0 51.7 nd 3.3 12.4 89.5 10.9 29.8 48.3 115.0 79.4 AC 121 6.2 4.8 10.2 14.0 12.4 1.6 2.2 13.7 57.5 19.8 17.7 55.6 75.4 33.9 AC 384 2.4 2.1 6.3 59.6 121.1 nd 4.2 35.8 73.1 2.8 28.7 80.6 76.6 36.4 OBT C1DOT C2D8T CDBT FL PYR CHRY ap B(e)p B(&)P PERL Ar 44 8.7 26.6 253.5 378.9 11.2 10.2 22.7 31.4 21.5 11.1 6.9 AC 121 3.0 21.2 211.1 348.7 22.1 17.1 34.3 35.1 15.3 10.6 3.2 AC 384 3.3 86.2 337.8 322.5 3.0 3.5 7.5 6.7 3.8 1.5 0.5 KEY: nd - none detected. N - napthalene, CI-C4N - alkylated naphthalenes, F - fluorena, CjF-C3F alkylated fluorenes, P - phenanthrene, CI-C,P - alkylated phenanthrenes, DBT - Dibenwthiophene, CIDOT-C2DBT alkylated dibenzothiophenes, Pl - fluoranthene, PYR - pyrene, 01RY - Crymene, BF - benwfluoranthene, S(e)P - Benzo(e)pyrone, B(a)P Senzc(&)pyrene, PERL - perylene. TABLE 12. Ile Grande time series (station D). ALIPHATICS (pg/g) AROMATICS (pg/g) SAMPLE DATE TOTAL RESOLVED ALK/ISO TOTAL RESOLVED AC 48 4/79 21.0 0.3 3.4 14.3 0.6 AC 125 7/78 59.1 0.9 016 27.8 0.3 AC 378 11/78 63.3 1.9 0.4 181. 0 12.9 AC 439 2/79 36.5 0.8 5.4 37.0 0.1 N CIN C2N C3N C4H F CIF C2F C3F P CIP C. 2P c3p C4P AC 48 nd nd nd rid rul rid nd 9.9 25.1 14.9 16.5 20.2 35.2 21.3 AC 125 11.8 5.1 143.7 686.2 825.1 12.8 68.0 286.0 674.3 63.3 309.9 539.7 729.0 229.0 AC 178 4.2 3.1 112.2 472.5 569.0 nd 20.2 152.4 386.7 13.9 127.8 251.7 473.0 227.9 AC 439 0.8 <1.0 <1.0 1.0 nd nd nd nd rld 3.7 5.9 5.1 4.5 2.4 DBT CIDHT C2DBT C3DBT FL PYR CHRY BF B(e)p B(a)P PERL AC 40 nd 13.2 80.7 145.6 19.8 18.3 34.1 17.2 16.7 6.1 44.7 AC 125 1033 785.4 2420 2917 55.2 39.4 107.7 99.0 52.7 16.0 6.1 AC 378 27.3 274.0 1481 1945 3.0 9.4 na 15.0 19.3 2.2 2.0 AC 439 nd 2.5 12.4 15.5 <1.0 <1.0 na 1.0 1.0 1.0 nd KEY: nd - none detected. na - not analyzed. N - napthalene, CI-C4N - alkylated naphthalenes, F - fl wrene, ClF-C3F alkylated fluorenes, P phenanthrene, Cl-C4P - alkylated phenanthrenes, DST - Dibenzothiophene, CIDBT-C2DBT alkylated dibenZothiophenes, Fl - fluoranthene, PYR - pyrene, CHRY - Crysene, BF - t)enzofluoranthene, B(e)P - Den w(e)pyrene, B(a)P - Benzo(alpyrene, PERL - perylene. 66 TABLE 13. Lannion I and II time series. LhNNION I TIME SERIES (STATION 9) SAME DATE N CIN C2R C3N C4N r CF C2r CIr F CP C P CF C F AC 107 7/78 nd 8.7 32.8 87.4 167.8 4.5 10.1 34.1 95.3 22.8 41.3 60.9 117.7 71.1 AC 436 2/79 1.6 1.9 2.9 1.9 nd nd nd nd nd 3.7 5.9 5.1 2.3 nd DOT ClDBT C2D6T C3OBT rL PYR CHRY BF SWP 0(a)P PERL AC 107 9.9 75.4 295.6 416.4 23 19.6 na 42.6 20.6 19.4 9.5 AC 436 nd 1.5 22.4 40.2 1.6 na 12.3 3.7 2.5 2.0 LAHVION 11 TIME SERIES ISTATION F) SAMPLE DATE N CIN C2N C3N C4N F Cir C2F C3P P CIP C 2P c3p c4p AC 118 7/78 nd nd 9.S 16.3 26.3 I.D 2.2 5:0 18.6 7.9 9.3 19.8 37.1 21.0 AC 377 11/78 5.8 4.1 7.3 6.1 nd 1.5 1.1 2.3 9.0 12.5 7.3 7.6 17.9 7.8 DOT COST C2DBT C3DBT FL PYR CRRY Br B(e)P B(&)P PERL AC 118 nd 10.2 90.1 181.2 10.6 8.5 na 22.6 12.8 6.7 2.2 AC 377 nd nd nd nd nd nd na nd nd nd rid KEYt nd - none detected. na - not analyzel. N - napthalene, CI-C4N - alkylated naphthalenes, F - fluorene, ClF-C3F alkylated fluorenes, P - phenanthrene., CI-C4P - alkylated phenanthrenes, DaT - Dibenzothiophene, CIDBT-C2DBT alkylated dibentothiophenes, F1 - fluoranthene, PYR pyrene. CHRY Crysene, BF benzoflwranthene, B(e)P Benzo(e)pyrene, H(a)P Benzo(a)pyr me, PERL perylene. 67 TABLE 14. L'Aber Benoit GC/MS results. AB 25 AB 21 N nd ..3 ClN nd 11 C2N 4 104 C3N 30 220 CO 55. 307' 4 CiF 4 25 C2F 11 113 C3F 59 311 p 3 14 ClP 30 50 C2P 54 4.40 C3P 84 800 CO 54 450 DBT 5 29 ClDBT 34 200 C2DBT 166 1350 C3DBT 195 2000 Fluoranthene 6 19 Pyrene 4 23 Benzanthracene nd 69 Chrysene 15 53 Benzofluoranthene 11 43 Ben zo (e) pyr ene 7 24 8enzo (a) pyrene 4 12 Perylene 2 10 Fl (Total) 29 460 F2 (Total) 30 440 KEY: nd none detected. 68 I Doo- T-1 H1d-1- ALK, ISO 8001 -0.8 200 IM. -06 3 so A4K ISO 400. F04 200- -0.2 10.000 3,700 DST DST ..20., '.0w 'IS 202 2,000- -90 3.000. 1.50D- 6c 2.DDO 1.0004 '0 .:78 7 78 11 78 2.79 4 78 7.70 11o78 78 2. 79 300 2S01 FIGURE 3.36. (Upper left) Terenez/Morlaix sediment time series. FIGURE 3.37. (Upper right) Offshore Ile Grande sediment time series. FIGURE 3.38. (Bottom) Alkyl homologue distributions of phenanthrene series. 69 3.4 Sediment Cores (Ward, Montana State University) An extensive series of sediment cores was obtained and analyzed by GC and selected samples analyzed by GC/MS for detailed aromatic hydro- carbon profiles. Samples were analyzed in support of anaerobic petro- leum biodegradation experiments (e.g. Winfrey et al., 1981). Three impacted sites and three control sites representative of beach, aber (estuarine) and marsh environments were selected (Table 15, Fig. 3.39). TABLE 15. AMOCO CADIZ chemistry program, inter- tidal cores (Ward). Frequency: December 1978 16 March 1979 28 August 1979 16 November 1979 10 May 1980 12 Total 82 Locations: oiled: AMC-4 (Portsall) - beach L'Aber Wrac1h - estuary Ile Grande (South-Oiled) marsh Unoiled: Ile Grande (North-Control) marsh Trez Hir - beach Aber Ildut - estuary Other Stations: Station 11 Station 12 Baie de Morlaix Port de Concarneau GC/14S: Several selected cores 70 BAIE DE WHLAIX L'AtIt K WRACI I 4MC4 "'O.TSA-1 AhEll Itt.llf 11.LZ IIIII ALSO KIH T DE CUNCAHNL AU FIGURE 3.39. Sediment core sampling locations (Ward). The basic set of data, where GC and GC/MS data exist, is illustra- ted by the data in Table 16. However, secondary data products are presented here to illustrate the basic findings of this program segment. Illustrative GC traces from March of 1979 are shown in Figures 3.40 and 3.41. While the hydrocarbon composition of the control estuary (L'Aber Ildut) is comprised mainly of biogenic compounds the marsh mudflat and beach both contain anthropogenic inputs. The Ile Grande "control" has been impacted by the AMOCO oil as its GC profiles closely resemble those for weathered oil. However, the Trez Hir (beach) site consists mainly of compounds of a pyrogenic origin. The impacted sites all illustrate AMOCO oil in various states of weather- ing, the Ile Grande (marsh) site containing the best "preserved" oil. This is also indicated by secondary treatment of some of the core aromatic data (Fig. 3.42) where oil in both L'Aber Wraclh and Ile Grande appears to be less weathered at depth. Data from the three impacted cores and the control core are shown in Figures 3.43. 3.52, 3.58, and 3.63. Each figure depicts the depth of penetration of AMOCO CADIZ oil throughout the December 1978 to May 1980 time period. The C3P/C3DBT ratio is presented as is the level of the non-petrogenic fluoranthene + pyrene (m/e = 202) total. Accompanying these figures are graphs of the down-core variations in gross hydrocarbon parameters (i.e. fl, f2, ALK/ ISO ratio) and detailed aromatic compound families. Figures 3.44 to 3.51 depict details of the L'Aber Wraclh cores, Figures 3.53 to 3.57 the Ile Grande 71 TABLE 16. L'Aber Wrac'h sediment core (March 1979). ALIPHATICS (ug/g) AROMATICS (ug/9) SAMPLE DATE TOTAL RESOLVED ALIK/ISO TOTAL RESOLVED 0-5 cm 3/79 530.0 27.8 565.0 16.4 5-10 3/79 113.7 8.4 318.4 19.4 10-15 3/79 95.4 2.8 118.1 4.0 .is-20 1/79 4.4 1.1 46.9 4.4 20-25 3/79 10.1 1.6 N CJN C2N C3N CO F CiF C2P C3F P CIP C2P c3p C4P 0-5 20.4 23.0 42.5 292.2 288.9 18.9 53.7 137.2 434.4 100.6 200.5 294.0 516.6 120 9-10 2.6 11.9 50.3 181.6 426.7 13.4 40.8 226.3 440.2 156.5 185.0 278.4 348.7 102 10-15 4.9 10.7 21.8 87.4 159.1 9.3 24.5 91.2 176.1 134.1 101.3 165.6 153.6 117 15-20 5.8 11.5 20.5 36.7 78.0 10.6 13.0 14.3 86.0 146.1 80.7 50.5 21.3 35 20-25 nd 2.6 14.3 17.9 Ind 6.9 7.0 25.0 25.5 100.0 61.4 29.3 10.1 26 DST CIDO T C2DBT C30BT FL PYR CHRY BF 8(e)P B(a)P PERL 0-5 40.0 87.5 1466 1598 116.6 157.7 170.4 136.3 89.0 64.8 24.5 r-10 24.3 135.6 1360 830.7 132.9 172.2 185.2 219.7 124.2 128.2 33.7 10-15 17.4 93.7 498.4 491.6 156.4 178.6 174.4 230.9 118.6 137.0 36.2 15-20 10.7 16.7 32.3 65.2 164.3 155.7 169.6 215.3 100.2 122.3 29.5 20-25 5.0 8.0 10.1 5.0 98.9 74.9 93.2 76.7 94.3 105.6 24.4 KEYi nd - none detected. N - napthalene, Cl-C4N - alkylated naphthalenes, IF - fluorene, ClF-C3P - alkylated Eluorenes, P - phenanthrene, CI-C4P - alkylated phenanthrenes, DST - Dibenzothlophene, ClDBT-C2DBT - alkylated dibenzothiophenes, Fl - fluoranthene, PYR - pyrene, CHRY Crysene, EIF benzofluoranthene, B(e)P - Renzo(e)pyrene, 8(a)P - Senzo(a)pyrene, PERL perylene. III) I M IM., (C) M."'I. M"'111"I (11) G I I "A( N.@' @I. MwIfIm (t) Od"I 13'a"j, IF)( 1 8-1, FIGURE 3.40. Saturated hydrocarbons in sediments from oiled and contro.1 sites. 72 IA) 0,1.1 E,I,..,y M1,011.1 18) cow"'I Elhwv Mudflat L IC) 00.1 @.Il ma,sb Mudhe (01 C...[,.l S.11 M-1, M.,10.1 ..... .......... . ... (E) 0.1.1 8-:1, IF) C'...1-1 B-1, AMOCO CAOIZ OIL :W ... - M..". 7 3c)- 161 A8ER NRACH A."I 27.1979 izi so- _j @11T@ C CI c2c3cd v C, C2 C3 C. 0 C. C2C3 0- ISO- ILE GRANOEISOUTH) ABER WRACS mwc. 1979 Much 1979 0 5, -5 - 1600 ... 2'00 J ,_r@u C I CiC3 4' "C I.C2 Cj C@ 0 C I C C@ CICZC3C4 CIC2C3C, 0 CIC2CI F ILE GAANOE ISOUTM@ ABER WRACH Mw@ 1979 M.r@ 1979 IS-20 10-15cm 100. c,c2c3c.,P C,Czc2c@,o C,CIC, IN clc?cjc@ o C,C..C3r@,0 c,::c3 FIGURE 3.41. (Top) Aromatic hydrocarbons in sediments from oiled and con- trol sites. L FIGURE 3.42. (Bottom) Comparative aromatic compound concentration pro- files derived from GC/MS-histogram presentation. 73 MEMBER 1978 MARC. 1979 :PHC c3p 2021.9 g@ ZP@C CIP 021@q,ql 1-91 C3087 91 C308r O-S 0-5 !00 32 mo 3S Soo S-10 1-0 39 260 5-10 440 @2 -305 Z 10-15 28 .39 490 lo-15 200 31 -335 2, 12--300 15-20 SO @j 3 x 21-26 17-_3w 2o-25 3o 2o2 17, 26-31 8-- 31-36 1, -30Q AUGUST 1979 -NOVEMBER 1979 FIGURE 3.43. Aber Wrac'h zp.c c,p 202[@,g, :PHC CIP 202o@ql W@9) Z30BT '_Qj C308T sediment cores. a-: 311 13 110 "M - -300 52 S3 loo -2 @w 33 3w o-2 57--loo 2-3 93a 32 jlo 3- 1 am 26 36o M- 196o 4-5 920 - -3oa LP@C C3@ 202(@%qi 5_;o 235 o'59 7!o 1.9.9, C30BT 0-3 270 o 76 '.'oo 11OCOlAOllOll P OGENIC BloGENIc o-15 @i - 200 400 '9'9 600 800 1000 '9'9 2000 3000 8OOO._. Saturate- Aromatics - - - - - - ALK/lSO _-20 20 OST 202 220 252 0.50 1.0 ALVISO 4.75 FIGURE 3.44. (Left) Aber Wrac'h core, December 1978. FIGURE 3.45. (Right) Aber-Wrac'h core, December 1978. 208 400 ma/a 600 Soo 1000 20 00 3000 L Saturates Aromatics -------- - - - ALK/lSO to. ------------ p 220 OBT__ 202 228 252 0.50 1.0 1.50 2.0 5.0 ALVISO FIGURE 3.46. (Left) Aber wrac'h core, March 1979. \\g@ FIGURE 3.47. (Right) Aber Wrac'h core, March 1979. 74 200 400 600 200 400 19`9 60 0 800 101 to- 20 Saturates 90, Saturates Aromitics - - - - - - - - Aromatics --------------- --ALK/ISO - -- ALK/130 0.50 1.0 1.50 0.50 1.0 1.50 2.0 2.50 ALK''ISO ALK/ISQ FIGURE 3.48. (Left) Aber Wrac'h core, August 1979. FIGURE 3.49. (Right) Aber Wrac'h core, November 1979. 10,00 2000 300g, 200 400 600 800 10 `20 -20 Saturates -------- Aromatics -------- - - OBT ALVISO 202 228 252 0.50 1.0 1.50 2.0 0.50 1.0 1.5 2.0 ALUISO ALVISO FIGURE 3.50. (Left) Aber WracIh core, November 1979. FIGURE 3.51. (Right) Aber Wrac'h core, May 1980. DECEMSER 1978 MARCH r979 !:PHC C3P 2020g,g) !:PHC CIP 202(@q,g@ @;19,9) C3CBT @9:9) C308T 0-5 1,100 28 25 0-5 700 .32 45 5-10 62 - - 5-10 300 @45 240 10-15 14 - - x 10-15 63 643 350 15-18 23 - - a 15-20 139 37 310 20-25 34 so 230 FIGURE 3.52. Ile Grande AUGUST 1979 MAY 1980 (oiled) sedi- !:PHC C3P 202(@gfgj ZPHC C3P 2020g,gl ment cores. (Mwgl C30ST Wq/gJ C3067 0-5 1,100 22 130 0-2 - 5.900 a 830 3,400 42 1,ioo S-10 550 42 170 2-10 - 10-15 "0 - - 1-2 3.8w 15-20 100 - - 2-3- 3.700 3-4 m 22.WO 52 1.300 4-5 3.800 AMOCO CAOIZ OIL 8.810 PYROGENIC BIOGENIC 75 20 400 60c 800 200 400 600 10 10 Ca 20 Sitaralas 2 0 - - - - - - - - - - - - - - - - Aromatics --- ALKiiso Aromatics --- ALKASO 0.50 1.0 1.511 2.0 0.50 1.0 1.50 ALKASO ALK ISO 1000 2000 3000,8000 200 400 600 800 lo@ 10 j20 N 20. Aromolics - - - - - - - -3811roill 08T ALK/180 2-02-i2l 252 1.0 1 5 2.0 03.25 ALK/180 2000 4000 81/1 5000 Saco 5000 10- 62 20 Aromatics - - - - - - - - ALK/196 0.50 1.0 1.511 2.0 2.50 ALVISO FIGURE 3.53. (Upper left) Ile Grande south core, December 1978. FIGURE 3.54. (Upper right) Ile Grande south-core, March 1979. FIGURE 3.55. (Middle left) Ile Grande south core, March 1979. FIGURE 3.56. (Middle right) Ile Grande south core, August 1979. FIGURE 3.57. (Bottom) Ile Grande south core, May 1980. 76 FIGURE 3.58. AMC-4 sediment cores. DECEMBER 1978 MARCH 1979 :PHC C3P 202(@g g !:PHC C3P 202,g 9, wgV C308T W9 g) C308T 0-5 131 48 37 0-5 7 3o 7 S 5-io 140 5-10 150 - - U-15 10-15 130 - - 15-20 15 - - 15-17 20-25 45 - - AUGUST 1979 NEW NPUT) NOVEMBER 1979 !PHC C? 2021@g,i :PHC C? 202i@q gi WWO C3DBT W94) C3D8T 0-5 46.OOQ 3 25 0-5 120 .28 32 5-10 8.800 1pp 25 268q@ 0-q@5 680 - - 60qS_ 9 MAY 1980 !:PHC C3P 202tg:g, W919) COST AMOCO CADIZOIL .-5 q1 2qJq1q1 2qJq@ PYROGENIC 8-13 24 SIOGENIC cores, Figures 3.59 to 3.62 the AMC-4 cores and Figure 3.64 a L'Aber Ildut core. Most of the cores were subdivided into sections of 3-5 cm in depth. However, two finer subdivisions from L'Aber Wrac'h - Novem- ber 1979 (1 cm segments down to 5 cm), Ile Grande - May 1980 (top 10 qmqm subdivided plus 1 cm sections down to 5 cm) were made. Penetration of oil was observed down to 10-15 cm in L'Aber Wrac1h sediments with concentrations decreasing with depth when viewed in 5 cm sections. Note however, that while petroleum aromatics were decreasing in concentration with depth', the pyrogenic PAH compounds increased with depth. Finer subdivisions of the core indicate greater variation within the core than the 5 cm sections would indicate (Fig. 3.52). An increase in vertical penetration of oil was observed for the Ile Grande site between December 1978 and March ,1979. A fresher layer of oil is found at the 15-20 cm depth (see Figs. 3.53 and 3.54) where naphthalenes, dibenzothiophenes, and to a lesser extent phenanthrenes, are more abundant than in surrounding layers. The gross hydrocarbon concentration changes at this level are not nearly as dramatic as are the petroleum aromatics, thus confirming that the "bulge" in Figure 3.42 is due to the less weathered nature of the buried oil. The finely divided May 1980 core (Fig. 3.52) indicates a higher petroleum content probably owing to a secondary input or to sampling variability which resulted in much higher levels (5q-10 parts per thousand) during May 1980. The down core distribution of hydrocarbons is quite non-uniform as well with a preserved layer of fresher oil at 3-4 cm. 77 The AMC-4 cores appear dominated by well mixed AMOCO oil through- out the 0-20 cm depths. Lesser amounts of pyrogenic PAH vis-a-vis the Aber and marsh sediments are due to the sandy nature of the AMC-4 samples. A new large input of oil is seen in August 1979 resulting in some down-core concentration variation. Chemical descriptions of the "control" site cores are shown in Figure 3.63. Note that non-petroleum PAH are widely observed in these sediments and that non-AMOCO CADIZ-impacted sediments contain 50-300 ppm of chronic hydrocarbon pollutants. q2q00 400 moll q600 Sao 200 400 600 1q0q[ q2 20 Saturates 20 Aromatics Saturates - Aromatics ALqKqIqSqO AqLqK/qIqSqO 0.50 1.0 1.50 0.50 q1.0 1.5 2.0 ALqKIqISqO ALK/qIqSqO 100q0 ng/g 200q0 2q0q0 4q00 'uQ/Q 3000 5000, 400q0q0 1q0 q10 qP DOT 702 228 252 2q0 Saturates 2q0L Aromatics -------- ---ALK ISO 0.50 1.0 1.50 2.0 A0qWISqO FIGURE 3.59. (Upper left) A12qW-4 core, December 1972q8q. FIGURE 3.60. (Upper right) AMC-4 core, March 1979. FIGURE 3.61. (Lower left) AMC-4 core, March 1979. FIGURE 3.62. (Lower right) AMC-4 core, August 1979. 78 ABER POUT ILE GRANDE INORTH, MARCH 1979 MARCH 1979 ZPHC C3P 202(@q,g) LPHC C3p 202mg 91 ,@g 9) C30ST Wq,gl F30ST 0-5 300 6 11 0-5 250 5-10 Go - - T 5-10 200 10-15 140 - - 10-15 100 15-20 77 - - 15-20 100 20-25 110 - 230 20-25 10 25-29 290 - 870 TREZ HIR MARCH 1979 A110COCAOIZOIL INC C3P 2021@9,g, PYROGENIC @g 91 C30ST BIOGENIC 0-5 59 26 640 5-10 30 - - 10-15 91 - - FIGURE 3.63. Miscellaneous sediment cores. 1000 2000 .20 OBT 262 228 Z52 FIGURE 3.64. Aber Ildut, March 1979. 79 3.5 Oysters and Plaice (Neff, Battelle) Samples of oysters and plaice from several impacted regions (L'Aber Wrac1h, L'Aber Benoit and Baie de Morlaix) and two supposedly unimpacted locations (Brest and Loctudy) were analyzed (Table 17, Figure 3.65). The results of the oyster time series analyses are summarized in Table 18. Both the "gross" hydrocarbon parameters as well as the petroleum-associated aromatic hydrocarbons are presented. Though not "clean", the control (Brest) oysters are several times lower in gross concentration throughout the time period and an order of magnitude lower in aromatic hydrocarbon content than either of the impacted sites. It is not apparent if the levels have decreased substantially in either of the Abers, though aromatic levels are 3 to 4 times lower a year and a half after the spill. For comparison, levels of several of the non-petrogenic PAH components (i.e. m/e 252) are presented. TABLE 17. AMOCO CADIZ chemistry program; oysters and plaice (Neff). Frequency: December 1978 4 April 1979 6 July 1979 7 February 1980 9 June 1980 11 Total 37 Location: L'Aber Benoit - Oysters; Plaice Muscle/Liver L'Aber Wrac'h - Oysters; Plaice Muscle/Liver Loctudy - Plaice Muscle/Liver; Oysters (7/79 only) Brest - Oysters Baie de Morlaix - Oysters (7/79 only) GC/MS L'Aber Benoit Oysters L'Aber Wrac1h Oysters Control Oysters 80 L ABEF1 WFIACH L' ABER BENOIT ALSO LOCTUOY FIGURE 3.65. Oysters and Plaice sampling locations. TABLE 18. Petroleum hydrocarbons in oysters (Crassostrea gigas). PETROLEUM b HYDROCARBONS P DBT 252 LOCATION DATE (Ug/g) (Ug/g) (ug/g) (Ug/g) L'Aber Wrac1h 12/78 660 12 22 0.04 4/79 1,200 15 12 0.02 7/79 590 5 10 0.03 2/80 820 10 16 0.60 6/80 (#l) 440 4 6 0.40 6/80 (#2) 560/570d - - - Brest 12/78 260 4 4 0.07 (control) 4/79e 1,100 11 10 0.01 7/79 91 0.3 0.3 - 2/80 150 0.4 1.1 0.6 6/80 93 0.6 0.7 0.2 L'Aber Benoit 12/78 690 - - - 4/79 800 15 15 1.0 7/79 - - 2/80 430 14 9 1.1 6/80 .520 3 5 0.2 aSum of phenanthrene and alkyl phenanthrenes. bSum of dibenzothiophenes and alkyl dibenzothiophenes. "Sum of m/e = 252. dReplicate analyses. eOrigin of sample unclear. 81 The GC traces for the impacted oysters are consistent throughout the study. The aromatic hydrocarbons (Figs. 3.66, 3.67) are dominated by the alkylated dibenzothiophenes and alkylated phenanthrenes th 'rough- out. The alkyl naphthalenes and fluorenes, significant in December of 1978, are removed from the tissues by June 1980. Aromatic hydrocarbons in the control oysters (Fig. 3.67), while less concentrated, are dominated by the same compound series, though the compositions in the controls remain consistent with time (i.e. no loss of flucrenes or naphthalenes). GC/MS traces of the oysters confirm the importance of the dibenzothiophene series (Fig. 3.68). Saturated hydrocarbon GC traces are illustrated in Figures 3.69 and 3.70 for impacted and control oysters respectively. The saturates of the L'Aber Wraclh samples are dominated by branched alkanes (e.g. isoprenoids) and a large low boiling UCH (Cll - C20). The UCM in Ehe controls is less pronounced yet significant, and while the isoprenoids are abundant indicating some weathered petroleum, a higher boiling -smooth n-alkane distribution (i.e. paraffins, n-C20 - n-C30) is Of equal importance. Figures 3.71 to 3.76 show some representative aroma- tic and saturated fraction data from oyster samples taken from L'Aber Wrac'h and the control station. The results of the plaice analyses are summarized in Table 19. The absolute concentration data does not address the source of the observed levels which for the most part are not linked to AMOCO CADIZ oil. The muscle tissues exhibit some petroleum-like GC traces includ- ing some UCM material and smooth n-alkane distributions with the presence of UCM material primarily responsible for the higher levels shown in Table 19. Liver tissue in all samples is much higher in absolute hydrocarbon content (Figs. 3.77 and 3.78). The f, (saturated) traces are characterized by a high molecular weight UCM (cycloalkanes), and an n-alkane distribution in the C22 to C23 region, while the f2 traces are characterized by polyolefinic material, including the biosynthesized compound squalene. These fl and f2 distributions are characteristic of fish livers from many geographic regions (Boehm, 1980; Boehm and Hirtzer, 1981) and are probably not related to any particular spill@event. 82 A Alk I P .,.o nor B FIGURE 3.66. Aber Wrac'h impacted oysters aromatic hydrocarbons; A December 1978; B June 1980. A -0 FIGURE 3.67. Brest control oyster aromatic hydrocarbons; A December 1978; B - June 1980. 83 C C C b C1 @C b1b C ' b b a -4 la Al; cc C C M/e C C b b b M/e I . I b a a mle A- TI P I T I -2@- -3r, 37 79 16 -10' A't 4'a A':3_ 4'4 A@ '9 A"7 49 49 90 C;,t q,;- FIGURE 3.68. Section of GC/MS total ion chromatogram of aromatic fraction of oyster sample illustrating major alkyl dibenzothiophene o (DBT comp nents (a = CIDBT: b - C2DBT; C C3DBT). =b 84 UC. 11CM FIGURE 3.69. Aber Wraclh impacted oysters saturated hydrocarbons; ns; A December 1978; B June 1980. ,A4 B_ ot@ FIGURE 3.70. Brest control oysters - saturated hydrocarbons; A Decem- ber 1978; B - June 1980. 85 10.000- 10,000- 1.000- 1.000- ng!g nglg 100: 100- p V W X Y Z AA 0 R S T UV Z A 13 C D E F G H I J K L M N A 5 C D EF G H I J K L M N 0 P FIGURE 3.71. (Left) Aber Wrac'h, Crassostrea gigas, aromatic hydrocarbons, December 1978. (See Figure 3.14 for key.) FIGURE 3.72. (Right) Aber Wrac'h, Crassostrea gigas, aromatic hydrocar- bons, June 1980. (See Figure 3.14 for key.) TABLE 19. Summary of Plaice hydrocarbon data. HYDROCARBONS STATION DATE TISSUE (p9/9 dry wt) L'Aber Wrac'h 5/79 muscle 90 7/79 Muscle 33 2/80 muscle 77 6/80 Muscle 186 7/79 Liver 1,350 2/80 Liver 1,200 6/80 Liver 1,640 L'Aber Benoit 5/79 Muscle 147 7/79 Muscle 17 2/80 Muscle 104 6/80 muscle 48 7/79 Liver 1,030 2/80 Liver 1,860 6/80 Liver 2,500 L,octudy 4/79 Muscle 2/80 Muscle 41 6/80 muscle 38 2/80 Liver 1,300 6/80 Liver 1,900 86 10,000- 1.000- 0 10.0- x CIO Alk-4 C-10 C I IN-61 A,ka,,C -I I n9/9 M .919 FAR PA IS - P- "'Y * "-- 1380..650- 100- A 8 C D E F G H I J K L M N 0 P 0 R S T Q V W X Y Z AA A S C 0 E IF G H I J K L M N 0 P 0 R S 10.0- 10.0- 0 "919 .9A 1.0- to- M x 0 A B C 0 E F G H I J K L M N 0 P 0 R S Y Z AABB A a C D E IF G H I J K L M W FIGURE 3.73. (Upper left) Brest, Crassostrea gigas, aromatic hydrocarbons, December 1978. (See Figure 3.14 for key.) FIGURE 3.74. (Upper right) Aber Wrac'h, Crassostrea gigas, saturated hydro- carbons, December 1978. FIGURE 3.75. (Lower left) Aber Wrac'h, Crassostrea gigas, saturated hydro- carbons, June 1980. (See Figure 3.74 for key.) FIGURE 3.76. (Lower right) Brest, Crassostrea gigas, saturated hydrocar- bons, December 1978. (See Figure 3.74 for key.) 87 AROMA TICS 7 IF, r Aka SATURATES C FIGURE 3.77. Plaice Liver, control. AROMA I ICS SAtIMAIFS FIGURE 3.78. Plaice Liver, oil-impacted. 88 3.6 Oysters and Fish (Michel, ISTPM) An analytical chemical program in support of the early post-spill (March 1978 - March 1979) programs of the Institute Scientifique et Technique des Peches Maritimes (ISTPM) was undertaken (Tables 20 to 22 and Fig. 3.79). Samples of freeze-dried oysters and fish (various species) were analyzed by.GC and several samples by GC/MS. The results of the analyses are tabulated in Tables 23 to 25. Based on the nature of the GC traces, sources of observed hydrocarbon distributions are derived: fresh AMOCO CADIZ oil, weathered oil, and biogenic hydrocar- bons. Often combined sources are apparent (e.g. weathered oil/biogenic hydrocarbons). Two oyster time series, summarized in Table 26, indicate that initial heavy oil impacts on the tissues are reduced over time.but certainly not eliminated. GC traces illustrating the change in aroma- tic hydrocarbon composition with time (Fig. 3.80) show that again the alkylated phenanthrene (P) and dibenzothiophenes (DBT) dominate the assemblage through February of 1979. Fish tissues do not reveal significant oil impacts. For the most part the hydrocarbons consist mainly of biogenic compounds (e.g. olefins) with an occasional UCM and again the presence of DBT and P compounds probably, though not definitely, related to AMOCO CADIZ oil (see Table 24). An attempt at decontamination via oyster transplantation yielded lower levels of hydrocarbons (Table 23; sample 143) indicating that once removed from a polluted substrate the oysters can depurate their oil burden significantly. Thus the oysters exhibit similar area-wide uptake of AMOCO oil, initially at the 3000 ppm level, rapidly reduced to the 300-700 ppm level and to the 50-200 ppm level a year after the spill. However, identifiable oil residues remain. Fish samples show only sporadic uptake of any oil indicating that the oil has not significantly impac- ted coastal fish, or that once impacted the fish rapidly depurate and/or metabolize petroleum. b..E (It LAN@ 'AU-W14ACft(f1 IENIE'Wl IHA.t Ik .0- .-1 A.L.1111-1.1 111.1 M 1. UAIL Dk ('Ahl,klk N'JN R.) ."It '. . I A 11 FIGURE 3.79. Oysters and fish sampling locactions. 89 TABLE 20. AMOCO CADIZ chemistry program, freeze-dried fish and oysters (Michel, ISTPM) . Frequency March 1978 1 Oyster April 1978 1 Oyster + 7 Fish May 1978 1 Oyster + 6 Fish June 1978 3 Oysters + 1 Fish July 1978 4 Oysters September 1978 4 Oysters October 1978 3 Oysters + 4 Fish November 1978 3 Oysters December 1978 4 Oysters + 5 Fish January 1979 1 Oyster February 1979 2 Oysters March 1979 3 Oysters TOTAL 30 + 23 53 Locations Various TABLE 21. ISTPM oyster sample summary. No. Date Sampling Location 5 5.4.1978 Aber Benoit - Prat ar Coum 71 23.3.1978 Baie de L'Arguenon 143 24.5.1978 Essai de decontamination 176 22.6.1978 Baie de Morlaix - Penze R.G. (Le Ven) 178 20.6.1978 Aber Benoit - Prat ar Coum 184 22.6.1978 Baie de Morlaix - Calot (transfert) 212 20.7.1978 Aber Benoit - Prat ar Coum 223 21.7.1978 Baie de Morlaix - Le Frout (Le Ven) 234 18.7.1978 Baie de Morlaix - Penze R G (Le Ven) 242 18.7.1978 Baie de Morlaix - Penze (B.I. Brannelec) 295 20.9.1978 Baie de Morlaix - Penze R D (Cablet) 297 20.9.1978 Baie de Morlaix - Penze R G (Le Ven) 311 20.9.1978 Baie de Morlaix - Penze R D (Gallion) 327 18.9.1978 Aber Benoit (Garo - Hanssen) 349 19.10.1978 Aber Benoit (Garo - Hanssen) 357 19.10.1978 Baie de Morlaix - Penze R D (Kerarmel) 359 19.10.1978 Baie de Morlaix - (B I Brannelec) 399 16.11.1978 Baie de Morlaix - Penze R D (V. Bernard) 400 16.11.1978 Baie de Morlaix - (B I Brannelec) 406 16.11.1978 Baie de Morlaix - Penze (Cadoret) 420 15.12.1978 Baie de Morlaix - Penze R G (Le Ven) 436 15.12.1978 Baie de Morlaix - Penze R D (Cadoret) 440 15.12.1978 Baie de Morlaix - Penze R G (Vallegant) 442 15.12.1978 Baie de Morlaix - R D Ile Noire (Kerarmel) 446 31.1.1979 Baie de Morlaix - Penze R D (V. Bernard) 471 27.2.1979 Baie de Morlaix - Penze R D (Gallion) 473 27.2.1979 Baie de Morlaix - Penze R D (Vallegant) 514 30.3.1979 Baie de Morlaix - Penze R D (Cadoret) 517 30.3.1979 Baie de Morlaix - Penze R D (Ker Armel) 518 30.3.1979 Baie de Morlaix - Penze R D (Cadoret) 90 TABLE 22. ISTPM fish sample summary. No. Nature Date Sampling Location 36-3 lieu jaune 13.04.1978 Portsall (3' W Amoco) 40-7 roussette 13.04.1978 Roscoff (Bank ar Forest) 41-3 lieu noir 13.04.1978 Portsall (8' N Amoco) 58-1 lieu jaune 13.04.1978 Portsall (21 E Amoco) 93-2 mulet 27.04.1978 Portsall 100 flet 24.04.1978 Baie de Lannion 118 lieu jaune 29.04.1978 Portsall (3' E Amoco) 151 maquereau 23.05.1978 Baie de Douarnenez 170-1 lieu jaune 11.05.1978 Riviere de Trequier 170-2 lieu jaune 11.05.1978 Riviere de Trequier 198 tacaud 16.05.1978 Baie de Lannion 200 lieu jaune 16.05.1978 Baie de Lannion 203 maquereau 3.05.1978 Baie de Lannion 209 mulet 30.06.1978 Aber Wrac'h 377-2 plie 26.10.1978 Ile de Batz 379-3 mulet 24.10.1978 Baie de Morlaix 415-2 sole 6.12.1978 Baie de Lannion 419-2 grondin 6.12.1978 Baie de Lannion 446-4 plie 15.10.1978 Baie de Morlaix 449-2 sole 15.10.1978 Baie de Morlaix 454-1 plie 5.12.1978 Aber Benoit 454-6 sole 5.12.1978 Aber Benoit 455-4 plie 20.12.1978 Aber Wrac1h TABLE 23. Results of ISTPM oyster analyses. Total Hydrocarbons Source Sample No.a (fl + fq; ug/9-Ary wl@.) (from GC)b 5 2700 1 71 610 2/1 143 180 2/3 176 530 2 178* 1600 2 184 380 2 212 270 2 223 400 2/3 234 640 2 242 80 2/3 295 340 2 297 270 2 311 290 2/3 327 630 2 349 420 2 357 320 2 359 50 2 399 100 2 40.0 95 2 406 70 2 420 150 2 436* 1000 2 440 90 3/2 442 150 2/3 446* 105 3/2 471 50 3/2 473 140 2 514 140 2 517 170 2 518 140 2 GC/MS results available (Table 25). aSee Table 21 for location and data of each sample number. bl = fresh AMOCO CADIZ oil 2 = weathered oil 3 - biogenic hydrocarbons 91 TABLE 24. Results of ISTPM fish analyses. 'IN)t I Hydrocarbons 1;0"rce sample No.'2 L@L_+_J_Z; uj'q dry wt) (from GC)b 36-3 is 3 40-7 39 3 41-3 29 3 58-1 18 3 93-2 45 2/3 100 70 3/2 118 45 3/2 151* 170 3/2 170-1 18 3 170-2- 31 2/3 198 31 3/2 200 69 2/3 203 37 3 209 30 3 377-2 a 3 379-3 25 3 415-2 154 3/2 419-2 15 3 449-2 13 3 454-1 21 3 454-6 20 455-4 6 GC/KS results available (Table 25). aSee Table 22 for location and data of each sample number. b, - fresh AMOCO CADIZ oil 2 - weathered oil 3 - biogenic hydrocarbons TABLE 25. GC/MS results of selected analyses of oyster and fish tissues (ng/g). Oysters Fish #178 #436 #446 #151 4170-2 N nd nd nd nd 2 ClN nd nd nd nd nd C2N nd nd nd nd nd C3N 290 nd 52 nd nd C4N 1400 nd 150 nd nd N 2190 nd 202 nd 2 p 220 30 85 17 40 CIP 1400 nd 220 30 70 C2P 4600 130 350 30 60 c3p 9500 200 1300 20 50 CO 10000 100 1300 10 40 P 25720 460 3170 107 260 DST 180 nd 16 nd 2 CiDBT 1400 nd 100 nd 24 C2DBT 6400 320 480 nd 120 C3DBT 9600 580 1410 nd 100 DST 17580 900 2006 nd 246 m/e 202 700 100 320 30 41 m/e 228 900 30 330 nd 20 m/e 252 500 nd 400 nd nd N - naphthalenes P - phenanthrenes DST = dibenzothiophenes 202 = fluoranthene + pyrene 228 = benzanthracene + chrysene 252 - benzofluoranthenes + benzopyrenes 92 TABLE 26. Petroleum hydrocarbons in oysters. EPETROLEUM LOCATION DATE (ppm) Aber Benoit April 5, 1978 2,700 June 20, 1978 1,600 July 20, 1978 270 September 18, 1978 620 September 19, 1978 410 Baie de Morlaix June 2. 1978 530 July 1978 70-600 October 19F 1978 60-230 February 27, 1979 240 March 30, 1979 150 Aik,6 P 1.11 I)BY FIGURE 3.80. Baie de Morlaix impacted oysters aromatic hydrocarbons; A - 5 April 1978; B - 27 February 1979. 93 3.7 Seaweed and Sediments (Topinka, Bigelow Laboratory for Ocean Sciences) In support of an investigation on the impact of the spill on macroalgal population recovery and growth, a series of plant and adjacent sediment samples was analyzed by GC to determine if and to what extent AMOCO oil was associated with the plants (Table 27). The data presented in Table 28 in conjunction with a consideration of Figures 3.81 and 3.82 illustrate that while several of the plant samples do contain weathered oil (see Fig. 3.81) the n-alkane, penta- decane (n-C15), is the most abundant biogenic component in all sam- ples. The distribution of biogenic components in general (Fig. 3.82) can be seen as contributing markedly to the total hydrocarbon levels even in the "oil-impacted" tissues. GC/MS results of an "oil impacted" plant's aromatic hydrocarbon fraction (Table 29) indicate that again the P and DBT family series are the most abundant aromatic compounds present. In this sample the P compounds are, in total, more abundant than the DBT series, but the C3DBT are still the most abundant group (8400 ppb). TABLE 27. AMOCO CADIZ chemistry program; seaweed samples (Topinka). Frequency June 1979 15 Plant + 7 Sediment August 1979 2 Plant May 1980 3 Plant Summer 1980 12 Plant TOTAL 32 Locations Various GC/MS One Seaweed Sample 94 TABLE 28. Summary of analytical results; seaweeds, summer 1980. Gravimetric GC Sample fl (Pg/g) f2 (P9/9) n-CI5 (pg/g) Status HC-4-1 41 74 38.5 1/2 HC-4-2 10 16 10.7 2 HC-4-3 11 14 4.8 2 HC-5-1 31 29 5.0 1/2 HC-5-2 61 72 35.0 1/2 HC-5-3 61 52 25.0 1/2 HC-5-4 11 12 21.3 2 HC-5-5 39 23 58.0 2 HC-5-6 17 16 25.3 2 HC-5-7 10 7 7.2 2 HC-5-8 12 9 34.0 2 HC-5-8 8 12 9.0 2 (Repeat) HC-5-9 5 15 3.7 2 Status codes: I = weathered petroleum 2 = biogenic TABLE 29. GC/MS results of seaweed aromatic fraction analysis (sample HC-5; Tregolonou, Fucus vesiculosis; 4 June 1979). Concentration Compound (ng/dry weight plant) C3-fluorene 610 Phenanthrene (P) 420 Clp 920 C2P 2400 C3P 6400 C4P 5300 ZP 15,440 Dibenzothiophene (DBT) 40 ClDBT 300 C2DBT 4000 C3DBT 8400 EDBT 12,740 m/e 202 (fluoranthene + pyrene) 930 m/e 252 (benzofluoranthenes 1800 + benzopyrenes) 95 A SATUHATES 8 AROMATICS 7 !t I FIGURE 3.81. HC-4-1 seaweed hydrocarbons oil-impacted. A SATURATES hjI, El AHWMAHCS FIGURE 3.82. Seaweed hydrocarbons control. 96 CONCLUSIONS A number Of Specific conclusions concerning the levels of AMOCO CADIZ petroleum hydrocarbons in various environmental compartments, the changing chemistry of the hydrocarbon assemblages, and the persistence of petroleum in these compartments are presented here. 1) Upon introduction into the environment, the oil weathered rapidly with evaporation and biodegradation changing the o'Ll's chemistry markedly even prior to landfall. 2) Oil impacted a variety of intertidal sedimentary types arid a number of secondary impacts were noted at many stations. 3)' Oil was'buried in most sedimentary environments with burial and/or penetration down to 15 cm in fine-grained sediments and deeper (",20-30 cm) in sandy sediment. 4) Oil remained less biodegraded in sandy beach environments Ithan in fine-grained sediments in which heavily biodegraded oil was characteristic. 5) The presence of TJCM material, pentacyclic triterpanes, and alkylated phenanthrene and dibenzothiophene compounds remain as characteristic chemical features of AMOCO CADIZ oil in sediments. 6) Less weathered oil appeared to be buried (10-20 cm) i n fine- grained sediments as evidenced in samples taken one year after the spill. 7) Offshore sediments were impacted after the shoreline impact, probably through processes involving beaching, sorption on intertidal sediments, and offshore transport of these sedi- ments. Samples taken after the spill in April 1978 do not reveal AMOCO CADIZ oil, thus indicating a lag (weeks to months) in offshore deposition. 8) Surface intertidal sediments taken in June 1981 show that "normal" background inputs, both of biogenic and chronic pollutant origins, have replaced AMOCO CADIZ oil as major components of the hydrocarbon geochemistry. Only at the most impacted stations at Ile Grande marsh and within the sandy beach sediment at AMC-4 (Portsall) do identifiable AJAOCO residues persist. At Ile Grande the aromatic marker compounds are absent, but hopanoid compounds (triterpanes) and a large UCM persist. 97 9) Oysters were initially heavily impacted by oil (several thousand ppm) and after two years (June 1980) traceable AMOCO CADIZ residues are still evidenced by homologous series of isoprenoid alkanes, phenanthrenes and dibenzothiophenes. Petroleum residues persist approximately at the 100 ppm level. 10) Fish do not appear to have been directly impacted (chemical- ly) by the spillage to any significant extent. 11) Compositional profiles traceable to AMOCO CADIZ oil are likely to "disappear" from all sediments within another year (i.e. 1982; four years after the spill) although this should be confirmed by direct measurements and attention to molecular marker compound distributions. REFERENCES Atlas, R. M., P. D. Boehm, and J. A. Calder, 1981, Chemical and biolog- ical weathering of oil from the AMOCO CADIZ oil spillage, within the littoral zone: Estuarine Coastal Mar. Sci., Vol. 12, pp. 589- 608. Blumer M., M. Ehrhardt, and J. H. Jones, 1973, The environmental fate of stranded crude oil: Deep Sea Res., Vol. 20, pp. 239-259. Boehm, P. D., 1980, Gulf and Atlantic Survey (Gas I): Atlantic survey for selected pollutants: Final Report, NOAA/NMFS Contract NA-90- FA-C-00046, National Marine Fisheries Service, Sandy Hook, New Jersey. Boehm, P. D. and P. Hirtzer, 1981, Gulf and Atlantic survey (GASI) Ches- apeake Bay to Port Isabel, Texas: survey for selected organic pollutants in finfish: Draft Final Report, NOAA/NMFS, Sandy Hook, New Jersey. Boehm, P. D., D. L. Fiest, and A. A. Elskus, 1981, Comparative weather- ing patterns of hydrocarbons from the AMOCO CADIZ oil spill observed at a variety of coastal environments: i n AMOCO CAD I Z - Fates and Effects of the Oil Spill, Proceedings of the Interna- tional Symposium, 19-22 November 1979, pp. 159-173. Boehm, P. D., J. E. Barak, D. L. Fiest, and A. A. Elskus, 1982, A chem- ical investigation of the transport and fate of petroleum hydrocar- bons in littoral and benthic environments: the TSESIS oil spill: Marine Environ. Res., (in press). Brown, D. W., L. S. Ramos, M. Y. Uyeda, A. J. Friedman, and W. D. Mac- Leod, Jr., 1980, Ambient temperature contamination of hydrocarbons from marine sediment - comparison with boiling solvent extractions: in L. Petrakis and F. T. Weiss (Eds.), Petroleum in the Marine Environment, Advances in Chemistry Series No. 185, American Chem- ical Society, Washington, D.C., pp. 313-326. 98 Calder, J. A. and P. D. Boehm, 1981, The chemistry of AMOCO CADIZ oil in the LAber Wrac1h: in AMOCO CADIZ: Fates and Effects of the oil Spill, Proceedings cT-the International Symposium, 19-22 November 1979, Brest, France, pp. 149-178. Dastillung, M. and P. Albrecht, 1976, Molecular test for oil pollution in surface sediments: Mar. Poll. Bull., Vol. 7, pp. 13-15. Grahl-Nielsen, 0., J. T. Staveland, S. Wilhelmsen, 1978, Aromatic hydro- carbons in benthic organisms from coastal areas polluted by Iranian crude oil: J. Fish. Res. Bd. Canada, Vol. 35, pp. 615-623. Gundlach, E. R. and M. 0. Hayes, 1978, Investigations of beach pro- cesses: in The AMOCO CADIZ Oil Spill, A Preliminary Scientific Report, IZAA/EPA Special Report, Washington, D.C., pp. 85-197. Mayo, D. W. , D. S. Page, J. Cooley, E. Sorenson, F. Bradlev, E. S. Gil- fillan, and S. A. Hanson, 1978, Weathering characteristics of petroleum hydrocarbons deposited in fine clay marine sediments, Searsport, Maine: J. Fish. Res. Bd. Canada, Vol. 35, pp. 552--562. Neff, J. M., B. A. Cox, D. Dixit, and J. W. Anderson, 1976, Accumulation and release of petroleum derived aromatic hydrocarbons by foui@ spe- cies of marine animals: Mar. Biol., Vol. 38, pp. 279-289. Overton, E. B., J. McFall, S-. W. Mascarella, C. F. Steele, S. A. An- toine, I. R. Politzer, and J. L. Laseter, 1981, Petroleum residue source identification after a fire and oil spill: in Proceedings 1981 Oil Spill Conference, American Petroleum Institute, Washing- ton, D.C., pp. 541-546. Pym, J. G., J. E. Ray, G. W. Smith, and E. V. Whitehead, 1975, Petroleum triterpane fingerprinting of crude oils: Anal. Chem., Vol. 47, pp. 1617-1622. Rashid, M. A., 1974, Degradation of bunker C oil under different coastal environments of Chedabucto Bay, Nova Scotia: Estuarine Coastal Mar. Sci., Vol. 2, pp. 137-144. Roesijadi, G., D. L. Woodruff, and J. W. Anderson, 1978, Bioavailal':)ility of naphthalenes from marine sediments artificially contaminated with Prudhoe Bay crude oil: Environ. Pollut., Vol. 15, pp. 223-229. Teal, J. M., K. Burns, and J. Farrington, 1978, Analyses of aromatic hydrocarbons in intertidal sediment resulting from two spills of No. 2 fuel oil in Buzzards Bay, MA: J. Fish. Res. Canada, Vol. 35, pp. 510-520. Warner, J. S., 1976, Determination of aliphatic and aromatic hydrocar- bons in marine organisms: Anal. Chem., Vol. 48, pp. 578-583. Winfrey, M. R., E. Beck, P. Boehm, D. Ward, 1981, Impact of the AMOCO CADIZ oil spill on sulfate reduction and methane production in French intertidal sediments: Mar. Environ. Res., (submitted). 99 STUDIES OF HYDROCARBON CONCENTRATIONS AT THE ILE GRANDE AND BAIE DE LANNION STATIONS POLLUTED BY THE WRECK OF THE AMOCO CADIZ Henri Dou, Gerard Giusti, and Gilbert Mille Laboratoire de Chimie Organique A, Associe' au CNRS n0126, Centre de St. i6rome 13397 Marseilles Ce'dex 13, France INTRODUCTION A study of the hydrocarbon concentrations in district no. 7 has been made since December 1978 in collaboration with the Marine Station of Endoume (Mes- dames Vacelet, Plante, and Lecampion). The first series of analyses was made outside of the CNEXO-NOAA framework, while our second study was supported by them. Results of the two studies are herein combined. METHODS Nature of the Samples Samples were collected at sites indicated in Figure 1. Sample sites A, D, and F are located in a very polluted zone; B, C, and E are located in a zone where the pollution level is lower since a dam was erected under the bridge to prevent the spreading of oil. Subscripts indicate specific areas samples: 1 - marsh, tidal creek, and 3 - upper mud flat (see Figs. 1 and 2). Samples were collected in December 1978, March 1979, November 1979, and May 1980, using a plexiglass corer (ID - 26 mm) . The 5 am superficial layer was subsampled with a steel spatula. Sediments were immediately frozen, flown 0 to Marseilles, and kept at -30 C. Analytical Techniques To yield the maximum amount of information, we have chosen the systematic soxhlet extraction following Farrington's method. This method is expensive and time-consuming but, for the biologists, At is the only one which gives satisfactory results. Moreover, reproducibility was tested several times and was found to be satisfactory. To avoid the very long separation of alumina- or silica-packed columns, we developed a micromethod of separation using Sep-Pak of Waters. This rapid technique is described in Analytica Chemica Acta. The general analytical scheme is as follows: 101 0 20 40 AL IL IL BI C2 AL E .0e JL 1W ML -IL Al AL va JL AL 3 AL WL OR D im. A" AIL AIDICIBI -Um- -UL JUL MUD on 1:1 SANDY MUD All- AIjC2,A3A3.E31 SAMPLING SITES REFERENCE STATIONSzCl,BI SCHORRE(SALT MARSH) CZ TIDAL CREEK POLLUTED STATIONS a AI,DI SCHORRE(SALT MARSH) A2 TIDAL CREEK L A3 SLI KK E (MUD SLOPE) FIGURE 1. Ile Grande marsh sampling sites. VL CHENAL 50 tidal creek @KHAUTE-SLIKKE SCHORRE higher mud slope salt meadow FIGURE 2. Detailed map of the sampling area. 102 extract-ion (toluene methanol-soxhlet) 2 pentane, water + sodium chlorida weight before saponification AVSP saponification with KOH weight after,saponification APSP separation with SEP PAK hexane elution chloroform elution I r fraction hydrocarbons polar fraction FA FB Each fraction was weighted by gravimetry (Balance: Perkin Elmer AD2Z, 1/10 of ug). In each case, the fraction FA was chromatographed on a capillary column (OV1 or SE52) with a Girdel or a Carlo Erba Fractovap 4160. The fractions FA and FB were also analyzed by high-pressure liquid chromatograph (column RP 18, radial pressure, waters-type detector). Since the FA fraction may contain saturatedr unsaturated, or aromatic hydrocarbons, fluorescence spectrometry (Perkin Elmer 3000) was used especially during the 1980 survey when biodegradation rates were higher. Infrared spectroscopy (Perkin Elmer 125 and 225) was also used systematically to show the absence of carbonyl functions on the FA fractions. During the 1980 campaign, it appeared that soine of the chromatograms of the FA fractions were not sufficiently resolved due to the c absence of saturated hydrocarbons. However, as they represented a substantial weight (A 1 - 257 g; A2 = 14 g) , NMR spectroscopy (C 13 and proton 250 Mz Cameca) was used to denote the presence of heavy-weight compounds with fused 103 rings (aromatic and nonaromatic). The structure of these complex mixtures was not elucidated at this time, but could be studied in the future. RESULTS Results are presented in 'Tables 1, 2, and 3. The weight of the compounds AVSP (before saponification) and APSP (after saponification) are indicated. The value of the ratio AVSP/APSP is characteristic of the capacity of the medium to be biodegraded. A value close to 1 shows poor bacterial activity. As the value increases, better biodegradation is indicated. This fact is due to the functionalized intermediate compounds made by the bacteria and extract in banic udium. The fraction FB is constituted by unsaponifiable compounds. Comparison of PA and FB is not interesting. Only the comparison of FB values for different sampling times is relevant. When biodegradation increases (less linear hydrocarbons), the PS fraction decreases. At Al and A2 station (in very polluted zones), the concentrations of hydrocarbons (linear or substituted) were about null in 1980, but complex organic compounds of heavy molecular weight, which are extractable with hexane (since present in the FA fraction), remain in the sediment. The molecular structure of these compounds has not been established (the petroleum-type asphaltenes are not extracted at the very beginning by pentane). CONCLUSIONS Station C 1 In 1980, in spite of the presence of 0.54 g in the PA fraction, its chromatogram was no longer characteristic of a petroleum-polluted zone (C17, C18, C19 predominant). However, light petroleum pollution is indicated,at -20 cm below the surface. Station B 2 in 1980, various deposits did not allow correct sampling; therefore, only 1978 and 1979 values were determined. In 1979, there seemed to an indication of return to normal state, especially at the rhizome level (bottom). Station Al in 1980, half of the hydrocarbons of 1979 remained. This fraction did not show the presence (by chromatogram) of linear or substituted hydrocarbons; however, its weight cannot be explained only by the unbiodegradable cyclanes or aromatics (33 g). 104 TABLE 1. Hydrocarbon content of Ile Grande marsh sediments. TOTAL SAMPLE WET AV-SP AP-SP re HY DROCARSONS STATION$ EIGHT 1g) (g/kgl Ig/kq) AV/AP iq /kq) IFAI (q.kql 12/78 11/79 12/70 11179 12/78 11/79 12/70 11/79 12/78 11/711 12/70 11/79 Marsh h, III 13A0 25.10 71.60 43.48 67.69 39.56 1.01 1.10 35.70 16.93 32.9" 18.94 rh 64.70 99.50 2.00 0.92 1.37 7.10 1.47 1.26 0.90 2.99 0. 4 *11 3.68 0, NU 12.55 14.60 173.90 243.45 162.50 210.50 1.07 1.15 67.90 102.12 94.611 94.51 rh 59.13 90.10 19,78 1.98 16.SO 0.96 1.20 2.06 9.40 0.46 0. 13 0.23 ca 24.2S 74.60 1.99 0.24 1.40 0.19 1.26 1.26 1.00 0.06 0. 411 0.10 su I .,o ,_:O, 1,: 00 * 1. 1; . S;jj 1.14 1.14 4.000 S.56 1.9 1.75 r 0 7 12 1:21 h @3.5 9 9 1.3 0.8 9 1.63 2.24 1 .6 0.59 0. 11 0.10 C, au 15.35 19.85 13.74 2.43 12.27 1.93 1.12 1.26 8.10 1.19 4. Il 0.43 rh 50.50 132.50 2.41 0.15 1.59 0.12 1.52 1.25 1.33 0.05 0. 26 0.03 Marsh Channel A@ Ps 513.200 10 18,52 201.459 16.70 12.41 1:11 1.73 9 4@25 7@69- 6.259 1 120:10 1.71 1*4 2 0.60 125 1.58 0: 0, 0, 0. 23 0. 48 0. 9 a 1 1 'b. 6 1 .3 0. 0-1 2.03 2.43 10 03 c @,:10 '9.@O 0:97 0 1 , 21 0.02 0. 0. C: 11 '0.'o " 90 14.07 6.32 8.27 4.29 1.70 1.48 5.00 2.17 3.26 1.41 .r 39.80 4NO 3.07 1.69 2.26 0.94 1. 36 2.01 1.50 0.66 0.77 0.09 ca 18.65 76.90 0.60 0.92 0.45 0.40 1. 33 2.3 0.22 0.2S 0.23 0.05 Upper Mud Flat A ps 37.90 1.05 11.48 16.91 11.44 12.91 1.00 1.31 5.60 a. 5.56- 3.45 cs 105.50 7.43 6.66 1.12 3.54 2.40 93 P: 7,500 2 69 2.32 L 0 52 11 ' 11 -l: 0 19 c 2.3 1:43 1:76 1.95 1 Arabian light 100 9 94 9 1.07 321 661 AV-s P : Befo r' 0:;po' ifica tion AWAP Ratio AP-S P Afte r onification FS Fraction B separated by Sep- P" HC Cl elution Wexqht in q Par kg of dry sediment. F9 Fraction A, hexane elutloft:tO,.i hydroaarbons *Th:d0fconcentr.tio n. are ev.l.- ps Surface crust .t rom sediment samples scraped by a Spatula to about 0.05 co depth. 105 TABLE 2. Hydrocarbon concentrations with depth. BIOTOPE AND DEPTH OF TOTAL HC, g/kg OF DRY SEDIMENT STATIONS SAMPLE (cm) 12/78 3/79 11/79 5180 Marsh Ps 39.87 Al 0- 3 32.97 18.84 14.98 3-14 0.47 3.68 0.03 34-37 0.04 Di 0- 3 94-68 94-51 230.60 3-14 8.13 0.23 17.78 14-26 0.48 0.10 0.18 Bi 0- 3 1.90 1.75 3-14 0.17 0.10 Ps 1.50 C1 0- 3 4.17 0.43 0.54 3-14 0.26 0.03 0.15 14-32 0.05 Channel Ps 14.20 A2 0- 3 7.69 10-78 6.59 2.57 3- 9 0.48 0.22 0.29 1.14 9-28 0.10 0.03 0.08 28-36 0.08 Ps 0.07 C2 0- 3 3.26 1.41 3- 9 0.77 0.09 9-28 0.23 0.05 Upper Mud Flat ps 5. 5 6 3.45 0.27-15 A3 0- 3 24.95 2.40 0.60 3- 9 0.65 0.50 E3 0- 3 0.89 3-11 0.08 11-20 0.08 20-27 2.81 ps 0.52 B.1 0- 3 0.19 0.56 3-15 0.22 15-19 0.16 ps = surface cr-ust, sampled by scraping 106 TABLE 3. Chromatographic analysis of the saturated fraction of hydrocarbons from Ile Grande sediments. C C Pr/Ph Predominance x 17/Pr I8/Ph STATIONS 12/78 1T/-79 T2/78 11/79 12/78 11/79 12/78 11/79 Marsh Al su 1 0.5 0.77 0.58 0.54, 0.67 1. xx rh 0.48 0.5 0.41 0.21 0.61 0.35 1. xx Di su 0.60 0.2 0.47 0.11 0.60 0.54 1. xx rh 1.24 1.5 0.93 1.92 0.75 0.66 1. 1.40 ca 3.30 2. 0.78 1.92 0.17 0.83' xx xx - B1 su xx 4.43 xx 2.62 xx 0.54 xx 1.57 rh 0.21 1.31 1.75 4. 7.80 3.2 2.37 xx C1 su 1.42 6.20 3.80 1.11 rh 2.80 6.67 1.66 1.05 Marsh Channel A2 ps 0.35 0.13 0.56 xx zr 0.50 0.26 0.58 1.25 ca 1.00 3.50 2.20 1.03 C2 su 0.13 0.19 0.39 0.44 3.50 23.33 0.99 1.84 zr 0.13 1.25 0.26 2.5 2.28 4. 1.06 1.48 cs 2.33 3.33 5.30 3.5 2.- 0.75 0.89 1.30 A3 ps 4.10 6.00 1.25 1.03 cs 0.09 <0.06 eO.85 xx B3 ps 0.53 0.53 0.94 1.04 cs 0.29 0.14 4.72 1.60 Arabian light 10.30 4.70 0.48 0.88 x predominance 2(C 23 + C25 + C 27 -aliphatic hydrocarhons C 22 + 2C 24 +2C 26 + C 28 xx = Nonextractable Pr Pristane su = Superficial pait; some centimeters Ph Phytane rh = Rhizosphere zr Reduced zone cs = Sandy layer ca Clay layer 107 Station D This site is subjected to a flowing stream which must have facilitated the deposit of hydrocarbons in large quantity, since between 1978 and 1980, an important increase was measured (95 g in 1978 to 230.g in 1980). Station C2 This site indicates that substantial biodegradation is in progress. Station A2 In the PA fraction, we again notice the presence of organic compounds which are neither linear nor substituted. The ratio FA 1978/FA 1980 is very close to that of station Al. Stations B3 and E3 These are reference stations in the less polluted zone. Some remaining pollution is visible, indicating that the dam was.not able to prevent oil from spreading to this side of the bridge. in 1980 at B 31 the chromatogram of the FA fraction did not indicate petroleum type, and there was a return to normal levels. At E 3 in 1980, some minor petroleum pollution remained. Station A3 There are large variations in the hydrocarbon concentrations, primarily at the sediment surface. Two separate samplings in 1980 (5 mm depth) yielded 0.27 g and 15 g of oil. However, at -10 cm, the values became equal to 0.50 g. The analysis of ;the 15-9 fraction shows that linear and substituted hydrocarbons have disappeared. our results are in good agreement with those noted by biologists. The less polluted zones are returning to a normal state, but a seemingly important residue remains in very polluted sites after linear, substituted, and light aromatic hydrocarbons have disappeared. 108 IA 2A %Y fill IB 2B !3 J Chromatogram 1) Station A3, March 1979. (A) surface, (B) bottom. 2) Station A2, March 1979. (A) surface, (B) bottom. 3) Station A3, 1980, surface skim. 4) Station D, 1980, 0-3 cm surface. L-"@4j-@A @11 109 EVOLUTION OF THE HYDROCARBONS PRESENT IN THE SEDIMENTS OF THE ABER WRAC'H ESTUARY by Jean DUCREUX Institut Frangais du Pftrole 92506 Rueil-Malmaison - FRANCE Following the Amoco Cadiz accident, the Institut Francais du Petrole analyzed'various samples in an effort to learn the physico- chemical characteristics of the oil pollutant released, and to observe its evolution over time. These studies dealt principally with samples of "chocolate mousse" and of beach-sand type surface sediments taken at various depths. In March 1978 a study of the physico-chemical evolution of -the hydrocarbons trapped in the subtidal sediments of the Aber Wrac'h estuary was undertaken with the collaboration of the Centre Oc6anolo- gique de Bretagne. INITIAL CHARACTERIZATION OF THE POLLUTANT The description of the Amoco Cadiz's cargo was undertaken on the basis of samples of foam ("chocolate mousse") taken at Portsall on March 22 and 23, 1981. This method of identification is based on the work of Pelet and Castex, and operates accordin g to a breakdown by chemical family, which has already been used by J. Roucache to make a geochemical study of the organic matter extracted from sedime@ts. The pollutant was identified as a mixture of two crude oils -- Iranian light and Arabian light -- physico-chemical characteristics of which are fairly similar ; they contain 45 to 47 % saturated 'hydro- carbons, 31 to 34 % aromatic hydrocarbons, 16 to 17 % polar compounds, and 4 to 5 % asphalt compounds (Fig. 1). The ratio of saturated hydro- carbons to aromatic hydrocarbons is on the order of 1.3. The saturated fraction is constituted of normal paraffins, iso- paraffins of isoprenoid compounds (pristane, phytane, etc... ) and cyclic and polycyclic alkanes or cycloparaffins. With gas-phase chromatography, n-alkanes in the sample taken at Portsall 23 March follow a regular distribution curve ; her top corresponds to the n-C15 and n-C16, after which it tapers off regularly to the n-C35 (Fig. 1). It is likely, incidentally, that compounds over n-C35 exist, but they have not yet been detected. Ratios of pristanes/n-C17 and phytanes/n-C18 in crude oil are, respectively, 0.37 and 0.51. To take into account evaporation pheno- mena and compare with evolved samples, this oil sample was topped at 3400C. An unresolved complex mixture (UCM) appears under the n-alkanes, constituted of isoparaffins and cycloparaffins. Alkane (n+iso) ill CCM CG 50 S PORTSALL 23-03-78 40- hude bfuld 50- 20 IR 10 2 0 L LIM 30 A@ A@,kw U IL FJ 70 S IRAN to ACHA JARI. L.0 17 10 zu 25 30 10 01 L-L@@ 5 APABIE SAOUDITE ARABIAN LIGHT 20 30 IJ Ji-ij FIGURE 1. Saturated hydrocarbons : Portsall 23.03.78; Iranian light and Arabian light. contents determined by mass spectrometry are on the order of 53 Cyclane contents decrease gradually from 14.2 % for the single- membered cycloparaffin rings, to 2.@3 % for six-membered rings. The combination of gas-phase chromatography with mass spectro- metry (GPC/MS) made it possible to distinguish the components of the aromatic fraction : monoaromatic, diaromatic, and triaromatic hydlro- carbons. Aromatic sulfur compounds of the thiophenic type are the berizo- thiophenes, dibenzothiophenes, and naphthobenzothiophenes. Polar compounds (resins) contain oxygenated functions (princi- pally hydroxyls and carboxyls) shown by infra-red spectrometer ana- lysis. Elemental analysis of this resin fraction made it possible to assess contents of the following elements C = 78 % H = 8.9 % N = 1.5 % 0 = 4.5 % In this case, the oxygen content is not calculated by subtracting the total of the other contents from 100 %, but is titrated by the Unterzaucher method. Asphaltene compounds, separated from the oil by cold-hexane precipitation, are known to have very complex laminated structures. metal (such as nickel and vanadium) and sulfur contents were determined on a dry extract before deasphalting (Fig. 2). Mdtals are essentially present in heavy fractions of crude (resins and asphal- tenes) as chelate compounds. Their contents range from 14 to 16.5 PPM for nickel and from 45 to 60 ppm for vanadium, with a ratio Ni/V of 0.27 to 0.31. Sulfur contents are on the order of 2.35 % by weight. These determinations will eventually serve as a point of depar- ture in foll.owing the pathways of the pollutant. It is to be noted that this study is not intended to cover the light evaporated and/or dissolved third of the cargo, of which about 25,000 tons is light aromatic hydrocarbons. Benzene, toluene, and xylene are estimated at respectively 3,300, 4,600, and 3,000 tons. EVOLUTION OF THE PHYSICO-CHEMICAL PARAMETERS OF THE HYDROCARBO1.14S We followed the chemical evolution of the hydrocarbons in three different types of samples taken from three areas of pollution - subtidal sediments (Aber Wrac'h) ; - water surface-stable emulsions (hydrocarbons with water, or "choco- late mousse") ; - intertidal sediments (beaches). The Subtidal Sediments of the Aber Wrac'h A detailed study of these sediments was made by DUCREUX and MARCHAND (1979) (collaboration IFP/COB) . The evolution of the 113 6qE6qi) T L'@ 1 L ISAT Ill.': I E X T R A C T 1 0 N !' C 1! C 1 3 I EX7PAIT I R f7encur Cr. Y: :Ctz.@Z/ E N L E V E TI E N TD USL I B R E SE' D E S A S P H A C C 'i 0qF_ F.':. SATURES MC. ARYIATIqQ@IES RESINES F I I S I C C G I R FIGURE 2- The program of analysis. hydrocarbons trapped in the subtidal sediments of the Aber Wrac'h was followed from March 31, 1978 to October 22, 1979 at nine stations (Fig. 3) in the estuary, taking into account the fairly pronounced marine character of the environment and the sandy, muddy, or sandy/ muddy nature of the sediments. Over a second period, from January 17, 1980 to June 24, 1980, three study sites, Stations 5, 6 and 8, which were deemed representa- tive of the different evolutionary patterns of the hydrocarbons, were again put into operation. The studies undertaken at these three stations only, are included in this report. To simplify the evolutions observed in the Aber Wracq'h, the figures included here are limited to samplings taken on March 31, 1978, November 22, 1978, June 20, 1979, January 17, 1980, and June 24, 1980. Station 5 A reduction was noted in the pollution of this station, situated in the outer area of the Aber where the marine environment is parti- cularly pronounced (Fig. 4). It was very slight, however, since the concentrations observed in June 1980 were over 670 mg of hydrocarbons per kilogram of sediment. This may be explained by the slightly muddy character of the sediment. 114 ABER WRAC'H 03 Ile d'Ehr4 04 OplcuguernejU 5 Anhe An Und4d4 FIGURE 3. The Aber Wrach. 199550 STA ST.8 ST@ 5 343-76 22-1@48 20.0,6.79 24-0@-Sc T FIGURE 4. Evolution of hydrocarbon contents in the sediments of the Aber wrac'h. 115 The relative contents of saturated and aromatic hydrocarbons decreased perceptibly (Fig. 5) in nearly constant proportions, the ratio SAT/ARO decreased slightly probably because of a significant increase in resins -- from 17 to 40 % by weight. After January 1980, however, contents of all these substance remained about the same. /U SA8qAqF T 73.31 40. 30. ST 5 STS 20 STS 10 AW 31-3-78 AW 22-11-78 AW 20 6 .79 AW 1?- 1-80 AW 24-6-80 2q%ARO 40. 30. STS ST ST 6 20. 10. AW 313-78 AW 22q11-78 AW 20-6 -79 AW 17-1-BO AW 24-6-80 POL ST 6 50. ST a ST 5 40. 30. 20 AqA 31q.3-78 AW 22q-11 78 AqW q20 q6 79 AW 17-1q-80 AqW 2q4 q6 q8q0 FIGURE 5. Evolution of the various chemical families (saturated hydrocarbons, aromatic hydrocarbons, polar compounds). The increase in resin contents results from an oxidation degra- dation of the crude in the medium, reflected in a perceptible upward curve of the absorption bands under infra-red at the level of the hydroxyls (3,600 to 2,900 cm- 14q) and the carbonyls -- including esters, and a great predominance of Carboxylic acids (1q,740 116 1,700 cm-1) (Figure 6). The elemental analyses of the resins confirm this tendency towardoxidation over time (oxygen levels in March 1978 were 5.4 % in June 1980, 7 31-03-78 22-11-78 20-06-79 17-01-80 24-06-80 3400 30()o 170 1100 700 Cm 1600 Station 5 - FIGURE 6. Infra-red spectrometry of the resins. The distribution of saturated hydrocarbons determined by gas phase chromatography (Fig. 7) demonstrates the evolution which led to the degradation of the n-alkanes (5.99 % to 3.33 %) to n-C30 (Table 1). It should be noted that by March 31, 1978, this degradation appears to have begun. Thus there was an increase in the ratios of pristane/n-C17 and phytane/n-C18 (Table 1). The isoprenoids degraded less easily than the n-alkanes (TISSOT and al.), but by June 1980 these compounds could no longer be detected. As degradation advanced, the contribution of the biogenic hydrocarbons increased, and n-alkanes of odd carbon numbers between n-C25 and n-C35 -- characteristic of the cuticle waxes of higher plants (Eglington and Hamilton - 1963) -- appeared, confirming the ter- restrial contribution to the contents of organic matter in the sediments. Moreover, the disappearance of the n-paraffins points up a relative enrichment in the unresolved complex mixture of the isoalkane and cyclo- alkane compounds, and generally in the compounds around n-C30. It was possible to demonstrate these last compounds by GPC/MS.by measuring the masses : m/e 217 being characteristic of the tetracyclic 117 HC.SAT. 11C.ARO. RESINES ASPHAL. SAT. PRIS PRIsT PHY3,,@' n-a1c. S Ni V Ni STATIONS Pr6levements i P'Is 't PdF. 't PdS t Pds ARO. 11y'll. @-C17 _,-J'@-C 18 It Pds Pds- __Ij_g/q 1jq/q V PORTSALL 23.03.78 45,1-' 34,55 -15,70 4,60 1,30 0,81 0,37 0,51 22,11 - - - huile brut,. PORTSALL 23 *03 78 37,30 34,10 2-1,30 4,32 1,09 0,44 0,37 0,51 29,05 2,38 14 45 0,31 w 0 1* 42,40 35,80 17,20 4,60 1,18 0,80 1,20 1,55 5,99 2,30 13 45 0,29 2 36,30 37,70 17,OU 8,90 0,9(, 0,67 1,60 2,14 3,33 3,57 35 50 0,70 3 -15, (JLJ 40,20 -'L), 00 4,8U 0,87 - - - - 2,70 20 40 0,50 5 4 jl'@O 32,00 30,E)U 5,90 0,98 - - - - 3,55 39 40 0,97 5 33,ou 33,70 --,7,80 5,50 0,98 - - - - 2,65 21 43 0,49 6 23,8o 29,50 39,30 7,40 0,81 - - - - 2,62 27 45 0,60 F- 7 18,80 26,70 '10,00 14,50 U,70 - - - - 2,22 23 50 0,70 8 .Z@'00 LH, 70 36, IjO 8,40 0,91 - - I - - - - I - - I 00 1 32,70 34,30 24,60 8,41 0,95 0,50 1,62 2,16 4,09 - - - - 2 34,00 35,20 20,10 10,60 0,96 0,65 1,46 2,02 3,91 4,80 400 800 0,50 3 31j'o 30,80 -19,50 8,00 1,03 - - - - - - - - 4 30,@O 37,10 27,80 4,60 0,82 - - - - 2,60 18 44 0,41 1 21J,1U - - - - 1 27,40 3'),UU 4,5U 0,70 2,32 10 35 0,29 6 20,40 29,3U 37,20 13,1() 0,70 - - - - 2,62 6 17 0,3B 7 20,40 28,40 37,()0 14,20 0,72 - - - - 2,28 24 48 0,50 8 22,00 24,80 43,40 q'B0 0,88 - - - - 1 2,90 18 1 34 0,53 1 18,40 18,10 38,10 9,40 1,02 0,22 0,50 0,96 9,74 - - - - 2 29,90 29,20 25,50 15,40 1,02 0,42 2,10 2,67 3,95 - 3 24,60 27,70 36,20 11,50 0,89 - - - - 3,80 - - 8 4 27,00 31,00 32,20 9,80 0,87 - - - - 3,90 40 67 0'('0 5 @%'@o 31,70 37,30 4,90 0,83 - - - - 2,86 14 65 0,21 6 19,20 22,70 40,90 17,20 0,85 - - - - 4,21 36 95 0,38 7 22,00 27,60 34,90 15,50 0,80 - - - - 2,00 24 45 0,53 Li 20,80 27,20 36,60 15,40 0,91 1 - I - - I - 1 3,32 1 22 1 39 0, 56 SAMPLING DATES I = 31.03-78 2 = 02,05.78 3 = 22.11.78 4 22.02.79 5 = 20.06.79 6 = 22.10.79 7 = 17.01.80 8 24.06.80 c TABLE 1 Evolution of the physico- hemical parameters. 3 -03-78 1 30 22-11-78 20-06-79 35 31 24-6-80 3, 29 25 CG STATION 5 FIGURE 7. Evolution of the saturated hydrocarbons. sterane compounds (CnH2n-6) eluted between n-C26 and n-C31, anc.1 m/e 191 characteristic of pentacyclic triterpane compounds (CnH2n-8) eluted over n-C30 (Fig. 8 and 9). This relative enrichi-,lent in cyclo- paraffins is confirmed by the mass spect rometry analysis of thE! fraction (Hood and O'Neal - 1958). Among the saturated hydrocarbons, (n+iso)-alkanes show a rapid diminution after March 31, 1978, and a stabilization thereafter ; cycloparaffins, principally 2-, 3- and 4- membered rings show an enrichment, and 5- and 6-membered rings are fairly stabilized (Fig. 10). 119 - - - - - - - - - - - - - - - - - - - - - - ---------------- C30 C25 C29 70 C25 C21 C27 ro C21 C23 C3 50 1 1 0 1, v @ cm C31 @! C22 Q 40 0 30 3B C32 32 C33 20 C33 20 to J Vr 1.4 r r-,. ---rT --r_r'_--T Fr '3 g 4ZO 5CB 703 bZa S, Z; a 4-@Z) 500 703 800 m/e 191 60] _,%. m/e 217 Penlacycliques @32 C32 twelpanes P, ntacycliques tFiterparles diterpeft" FIGURE B. Combination of gas-phase chromatography and FIGURE 9. Combination of gas-phase chromatography and mass spectrometry of a saturated hydrocarbon mass spectrometry of a saturated hydrocarbon fraction. fraction. 60 50- 40- 30- 20- 10 11 3 71 0 1 2 3 4 5 6 [email protected] sTAT'ON 5 FIGURE 10. Distribution of cycloparaf fins by mass spectrometry. In the aromatic fraction, chromatographic profiles showed - a loss in light hydrocarbons as from March 31, 1978 ; - a positive response in flame photometric detector (FPD), specific sulfur detector, throughout the period studied, which is significant, given the presence of sulfurous aromatic hydrocarbons (thiophenics), and indicative of extensive continual pollution (Fig. 11) ; - a persistence of dibenzothiophene despite a decrease until June 1980; - an unresolved complex mixture similar to that of the saturated hydro- carbons.. Metal (nickel and vanadium) and sulfur contents assessed through .various samplings (Table I) confirm the persistence of the pollution, and do not indicate any significant evolution. The divergences observed may be attributed to variations in concentration of the two crudles present in the polluting mixture. It is to be noted that it is not possible to date to establish the cause of a relatively high peaX at the level of the n-C16 in some saturated hydrocarbon chromatogralm. A similar peak appeared at Station 6. Station 6 Decontamination is virtually nil here. In June 1979 A maximum of over 6,000 mg of hydrocarbons/kg of sediment was registered ; by June 1980 hydrocarbon contents were still over 1,200 mg of hydrocarbons/kg of sediment (Fig. 4). This is true of all stations where the marine cha- racter of the environment is least pronounced that is, in the upper reaches of the Aber. 11 1>7. As can be generally observed in the stations, the most striking evolution is the degradation of the n-alkanes to n-C30. They appear to evolve very rapidly, since the very first sampling taken at this station on March 31, 1978 indicated that the ratios of pristane/n-C17, 121 FPO FID PORTSALL 23-3-78 AW 31-3-78 FPO IP S2 F1 AW 22-11-78 AW 20-6-79 . .. .............................. ..... FPO 'if I"C7, Fla AW 17-1-80 AW 24-6-80 STATION 5 FIGURE 11. Evolution of aromatic hydrocarbons. 122 and phytane/n-C18 were among the highest and that this degradation was among the most advanced (Table I). In figure 5', a continual decrease in saturated hydrocarbons is noted over time (33 % on March 31, 1978 -- 22 % on June 24, 1980). In spite of a maximum aromatic hudrocarbons content observed in June 1979, they underwent the same type of evolu- tion (34.3 % on March 31, 1978 - 24.8 % on June 24, 1980). The result is an increase in the relative contents of resins ; but it should be noted that in March 1978, polar compound contents (24.6 %) were higher at this station, which is located in the outer part of the Aber, than at Station 5 (17.2 %) where the marine character is more pronounced. This level held true until June 1979, when the increase became the same as at Station 5. This phenomenon can be explained by the presence of polar com- pounds of terrestrial origin deposited by the river which flows into the Aber. This contribution of autochthonous organic matter is seen in the presence of n-alkanes of odd carbon numbers (n-C25, 27, 29, 31, 33) in the aliphatic fraction, all the more accentuated by the signi- ficant acceleration in the degradation of the normal fossil paraffins (Fig. 12) The result is a relative enrichment in cyclic satured hydro- carbons (cycloparaffins) -- particularly of 3-, 4-, and 5-membered rings around n-C30 -- which are less easily degradable. Chromatogram profiles of the aromatic hydrocarbons indicate how polluted the station is, principally in the FDP response (Fig. 13), where the persistence of dibenzothiophene and its alkylate derivatives (and also of naphthobenzothiophenes) may be noted. The stability of these compounds made it possible for us to use these chromatograras as the "fingerprints" of the pollution. As a matter of fact, according to the DGMK report, biogenic hydrocarbons of terrestrial origin (recent sediments) are poor ill aromatic compounds (< 5 %), and particularly low in sulfurous corn- pounds of the thiophenic type. The evolutions observed in the resins are similar to those at Station 5 ; an oxidizing degradation increases the oxygen contents (8.5 % in June 1980), and an upward curve is noted in the infra-red absorption bands of the hydroxyls and carbonyls, the latter of which are predominantly carboxylic acid compounds (Fig. 14). Metal and sul- fur contents remain fairly constant over time. Station 8 This station was distinguished in being the farthest from the sea, in an area of sandy mud. Hydrocarbon contents show that the decontamination process at this station was virtually nil (Fig. 4). An anomaly is noted in January 1980, when the hydrocarbon con- tents went over 5,000 mg./kg of sediment. As we shall see below, this is due to pollution from petroleum cuts or fuel oil which was later deposited on top of the pollution under study (Fig. 15 and 16). Aril- through information is lacking on some samples, which were too small, it can be seen that metal and sulfur contents remained very stable over time. 123 31-03-78 31 29 27 20 2S 20-06-79 31 29 27 25 3 6q@3, @A M 31q1 29 25 24-6-80 31 29 27 0qw8q%q@4qt 25 STATION 6 FIGURE 12. Evolution of the saturated hydrocarbons. Examination of the qc44qhq,qromato2qgrams of the "saturated" fractions shows a quantity of biogenic hydrocarbons of terrestrial origin which is not negligible, and which is characteristic of the upstream sta- tions on the Aber which receive large quantities of deposits from the soils. The distribution of nq-alkanes is in fact typical of that ob- served in the extracts from recent sediments in the predominance of qodd-carbqon-numbered nq-alkanes from n-C25 through n-C35 (Fig. 16). These quantities of biogenic hydrocarbons present in the ali- phatic hydrocarbons are demonstrated in the ratio R29q-31 developed 124 FIB ..A FPO PORTSAILL 23-3-78 AW 31-3-78 FIB S2 FPO ............... AW 22-11-78 AW 20-6-79 FIB FPO AW 17-1-80 AW 24-6-80 STATION 6 FIGURE 13. Evolution of aromatic hydrocarbons. 125 31-03-78 2241.-78 20,06-79 17-01-80 24,06-80 3400 300o do 0 1600 STATION 6 FIGURE 14. Infra-red spectrometry of the resins. by Tissot et al ; they were calculated for the earliest samples only R 29-31 = 2(C29 + C31) C28 + 2C30 + C32 The relationship shows that in this case, the n-C29 and n-C31 predominate-Where the value is over 1, the odd-numbered carbons pre- dominate. In Table 2, a compilation of these values for all the stations, it is seen that R is much higher than 1 at Stations 6 and 8. Even in March 1978, this presence of natural compounds modified the distribution of hydrocarbons by family - saturated hydrocarbon contents were low - polar-compound contents (resins and asphaltenes) high. While the ratio of SAT/ARO decreased slightly, as at Stations 5 and 6 (Fig. 17), degradation of n-alkanes and isoprenoids (pristane and phytane) was observed throughout the period, showing clearly which polycyclic alkanes are most resistant to degradation. The chromatcgrams of the aromatic hydrocarbons have very marked profiles under photometric detection (FPD), and a persistence in the unresolved complex mixture (FID) (Fig. 16). Comparison of the results of the GPC with those of the high resolution MS undertaken on the re- ference samples of March 23, 1981 at Portsall, and the sampling taken 126 31-03-78 33 2S 20 22-11-78 20-06-79 27 25 311 2@ LL11 24-6-80 31 27 410,L-1 CG STATION 8 FIGURE 15. Evolution of the saturated hydrocarbons. on June 24, 1980 at this station, shows a degradation (or dissolution) of the alkylated cycloparaffins to C3 and C4. Concerning the thio-- phenic derivatives, some benzothiophenic alkyls disappear to become @4 @10 C5, while the initial alkylate derivatives of dibenzothiophene become C3 ; the naphthobenzothiophenes persisted. In the sampling taken in January 1980, an abnormally high res- ponse is noted -- by Flame Ionization Detection (FID) and by FlaME! 127 FP13 'NJ FIB PORTSALL 23-3-78 FPO S2 S2 Flo AW 22-11-78 AW 20-6 -79 FPO S2 FIB ,AW 17-1-80 AW 24-6-80 STATION 8 FIGURE 16. EVolution of aromatic hydrocarbons. 128 SAT.,/ ZA R C., 0 ST. 5 o ST.6 A ST. 8 PORTSALL 23.3.78 0.5. 0. AW.31-3-71 AW 22-11-78 AW 20.6 - 7 9AW 17-1-80 AW 24-6-80 FIGURE 17. Evolution of the ratio of saturated hydrocarbons to aromatics. TABLE 2. Values of the R29-31 ratio. 2 (C29 + C31) Valeurs de R29-31 = C28 + 2C3o + C32 Station 31.03.78 2.05.78 1 1902 1,23 2 1 13 3 1,10 1,05 4 1,36 1,07 5 1,23 1,22 6 2,34 1,35 7 1,26 1,59 8 4,00 2,78 9 1,06 1,12 129 Photometry Detection (FPD) as well. This is caused by the presence of a light cut (gasoline, fuel oil, etc... ) deposited on top of the Amoco. Cadiz pollution (Fig. 16). Here again the infra-red resin spectrum shows an increase in absorption of hydroxyls and the transformation of esters to acids (Fig. 18). The oxygen content of these polar compounds (1.4-7.5 %) confirms the degradation by oxidation. 31 03 78 22 11 78 20 06 79 17 01 80 2406 80 ok"@, 1100 700 CM 3400 3000 170 1600 Station 8 .FIGURE 18. Infra-red spectrometry of the resins. Stable Hydrocarbon/Water Emulsions or "Chocolate Mousse" When petroleum spreads over a marine environment, the movement of the waves, the wind, and the currents causes a very rapid emulsion with the sea water which is called "chocolate mousse". These stable emulsions are constituted by dispersing sea water in the hydrocarbons (inverse emulsions). This type of emulsion was sampled during the first month after the accident (Fig. 19, Table 3). A program of analysis which is different from that used for the sediments and sands was employed (Fig. 20). After purifying the emul- sions (of sand, algae, etc ... ), they were distilled to measure the water contents of those emulsions with boiling points below and those above 3400C. This method is described by Pelet and Castex. The results are assembled in Table 4. 130 ROSCOFF BRIGNOGAN LANNION KERDENIE GUISSENY MORLAIX ORTSALL BREST 22. 23 mars 78 PORTSALL GUISSENY 3 avril 78 ROSCOFF A avril-78 PORTSAU GUISSBY BRIGNOGAN 18 octobre 78 KERDENIEL 31 janfler 79 TREOMPAN 28 mars 79 KERDENIEL TREOMPAN FIGURE 19. Sampling stations. TABLE 3. Characteristics of the samples. Lie@jx Dazes des POZT5ALL 22.,-.76 aj C@:.-.:6r Et 7CC-rn' 4.:Z.7= 2E.C2.79 KR D E,,'l E L 26.03.79 (P.-cf. = 25 c.) Moreover, on these fractions a simulated distillation curve was plotted (TBP or True Boiling Point) by gas phase chromatography accor- ding to the method of Petroff et al. (1981). OMPA -J@L- P Results of the disti llation (Table 4) and the TBP on the PI-3400C (Table 5) indicate a loss by evaporation and dissolution of about 7- 8 % of the light-hydrocarbons between March 22, 1978 and April 4, 1978. 131 PILLUE ')-D -I [-I T-S--F -ECU P-ERES DR U T L (-.T V XTR4 M rJX@iLET C@4@:' DECANTATION' FILTRATION CENTRIFUGATION DESHYYLUATION Na2SO- PRODU IT EPURE PqDD'JIT EPUR--- ETtTAGE 340*C POIDS S, Nil V 1. 3;0+ 1 R ASPHALTENES I POIDS PAR CHR0114TOGRAPHIE COUCHE @!INCE H-..ARO.',AT]qJES_ I RESVIES FC @ is k! IR FIGURE 20. The program of analysis. TABLE 4. Results of distillations of samples of emulsified crude. (----rr4Uv..er1ts PORTSALL SENY ROSCOFF PORI I IBRIGNOGAN mars 1978 3 avril 78 978 22 mars 78 4 avr[t I dans 71.7 68.7 67,7 51 60 57 57 340% 24.3 20,0 8.9 13,15 15,95 13.05 8,7 ',voids 11C IT @ 340% 75.71 1 80.0 91.1 84,05 86,95 91.3 86,10 Two samples (Guisseny, March 23, 78, and Brignogan, April 4, 1978), lost considerably more. The latter was initially the point of highest Pq0 D4S JIT EPUR . V @E 'IES IR contents. Its chromatographic profile (TBP) is also very different from that of the samples taken at Portsall on March 22 and 23 (Fig. 21). 132 TABLE 5. Simulated distillations of PI-3400C. 2 poid: Tempiratures 'C cumuli PORTSALL PORTSALLI GUISMY PORTSALLI GUISSENY ROSCOFF BRIGNOGAS '22.3. 23.3 23.3 4.4 4.4 3- 4 4.4 1 162,2 199.1 223,6 216,3 210.2 227.3 239,2 5 93,4 218,6 248,2 233,4 229,7 245,5 72.9 10 207,8 228,7 260,8 246 7 245,9 257,8 285.9 20 224,6 245,0 273,8 264.2 266.2 271.4 300.5 30 237,0 256,1 285,9 275,8 279.9 284,8 310.5 40 251,4 271 7 292.4 286,3 290,7 292,9 319,6 so 267,9 281,7 300.6 294,7 300.9 301,6 326.9 60 282,7 290.6 306,5 303,5 308.9 308,9 135,0 70 295,2 301,3 317,0 313,2 319.2 318.8 343,6 80 310,6 313,2 326.0 322,9 329.9 328.1 353.9 90 330,6 331,3 343,0 340,7 346,8 343.6 371.0 '100 392,1 390,3 400.9 459,8 4 422.3 460,0 F-7 7 -J le 77 17 -7-7 7 r FIGURE 21. "TBP" PI-34VC. 133 The chromatogram of the saturated hydrocarbons shows that the n-paraffins had disappeared (Fig. 22). Although there are divergences 4%--1 4.4.76 FIGURE 22. Chromatogram. of the saturated hydrocarbons taken at the Brignogan station April 4, 1978. in such parameters as metal and sulfur contents, the infra-red s- ,pec- trum definitely confirms that it is crude ' from the Amoco Cadiz. We believe that these differences may be due to the fact that we were dealing with oil sheets that had been treated to a greater or lesser degree. These observations should be compared with those made by Aminot et al-at the same time at a station just offshore from this one, which showed an abnormal loss of dissolved oxygen. He explained it as in-situ biodegradation of the hydrocarbons, which appears to be the only logical explanation. Distillation of the 340+ fractions shows little in the way of interpretable differences (Table 6). The sulfur and metal (nickel and vanadium) contents show, by their stability, how important these com- pounds are as pollution markers (Table 7). TABLE 6. Simulated distillations of 340+ residues. 2 pq:ids Tempiratures *C 'OGAN dis il Ji. PORTSALL PORTSALL GUIqSEVY PORTSALL GUISSqM ROICOFF BRIqO 22.3 22.3 22.3 4.4 4.4 3.4 4.4 1 291,4 283,8 281,8 305,1 303,2 295.8 266.6 5 346 .0 325.2 327, 1 338,9 342.9 336,1 318,1 10 370.2 346,9 349.6 359.4 367,9 360,8 342.2 20 408.8 383,3 386.2 394.2 405.3 399,5 386.3 30 446,0 420,3 422,6 431,3 443.8 439,5 421,7 40 484,1 457,5 458,5 463.7 478,3 472.7 458,8 50 526.9 497,8 q497.2 499.3 51q4.6 q515,7 503,6 60 543q.4 q540,2 539.3 558q.6 560.1 54,35 q1 6q1 qZ 62q,1 q1 62,3 qZ 6q1,8 q1 56q,8 1 61,7 qZ q547q,3 547q.5 548,6 549,6 q: 568q.6 548 569q04 The breakdown by chemical family is shown in Table 8. It appears that the evolution of the crude in all of the emulsions is a slow process. A slight decrease in the ratio of saturated to aromatic 134 TABLE 7. Sulfur, nickel and vanadium contents. S Ni V Ni/V Z Pds Vg/& Ug/9 FORTSALL 22.03.78 2.33 16,5 62 0.27 PORTSALL 23.03.78 2,38 14.0 45 0,31 GUISSENY 23.03.78 2.38 16 50 0,32 2.22 14 ROSCQFF 3.04.78 48 0.29 PORTSALL 4.04.78 2,30 16 58 0.28 ce CUISSENY 4.D4.78 2.18 20 65 0,31 BRIGNOGAN 4.04.78 2,30 22 68 0,32 TABLE 8. Evolution by chemical family. Pds pds pds 7 Pds Z SAVAROS. H.C.saturt@ F.' ar=at. Risines Asphaltines, PORTSALL 22.03.78 39.06 35.66 21.71 4,57 1.06 + PORTSALL 23.03.78 37.28 34.12 24,29 4.32 1.09 GUISSERY 23.03.78 41.69 34.11 16.40 7,80 1,22 PORTSALL 22.03.78 47,65 31.29 16,77 4.28 1,52 (21,05) PORTSALL 23.03.78 45,12 34.55 15,70 1 4,60 1,30 (20.30) GUISSENY 23.03.78 39.45 31 24,90 1 4,60 1,27 (29,50) 0 PORTSALL 4.04.78 46,75 30.12 19.18 3.95 1.55 (23.13) W GUISSENY 4.04.78 46,38 31.76 17,65 4,21 1,46 (21,86) BRIGWGA,'; 4.04.78 34.10 31,50 25.60 8.80 1,08 (34,40: ROSCOFF 3.04.78 43,69 34,01 16,96 1 5,34 1,28 1 (22,30 KERDENIEL 18.10.78 40.00 33.50 19,50 1 7,DO 1.19 (35 cz) (26,50) TREO-'TAN 31.01.79 33t9O 39 19,20 7.90 0,87 TREOYA,' 28.03.79 31,10 38.90 22.60 7.40 0.80 (30,0) KEFLDENIEL 28.03.79 38,40 36.90 '18,10 6,60 1,04 (24,70) KERDENIEL 28.03.79 27,60 26,30 35,90 10,20 1,05 1 (25 cm) (46.10) hydroc arbons is seen, however, as well as a slight relative increase in the polar compounds. The infra-red spectra of these compounds show, in fact, a slight upward curve in the absorption band of the cartonyls 135 between the end of March and early April 1978 (Fig. 23). These results corroborate those of Roussel and Gautier at Antifer.(1979). PORTSALL 22.3.78 PORTSALL 23.3.78 GUISSENY 23.3,78 ROSCOFF 3.4.73 PORTSALL 4.4.78 GUISSENY 4.4.78 BRIGNOGAN 4.4.7F 3600 3200 2600 1700 i6oo 1100 7;0 criF' FIGURE 23. Infra-red spectrometry of samples of crude (R 340+) and extracts. 136 The pathways followed-by the crude are shown in the triangular diagram of the saturated and aromatic hydrocarbons, and the polar compounds (resins + asphaltenes) (Fig. 24). Pj Portsmll 22.:'3. 7@ P, Portnall 23. @@3. 7 @ rqS Port3all 4.)4.70 emulsions 0qNGuisanny 23 . C3 . 7 ". CA CutsRony 4.D4.7-1 ARO. BOrignogan 4.74.7a Rnosc6ff 3. @ 2 7 1 Tj Tr6omDesn 31 1 @ 1 7 90 10 Y T% Tr6ompan [email protected] sands K, Kardeniel (25 C-) Y,2 Kerdnniel 20 03 , 7@ 80 K, Kardeniol 28,03.7" (75 c-) 56qX V Y @y 7 -@o AW9 Abpr wric'h 0 StiLlon 9 IV I;V, 60 40 qW Y qX", ,IV Y-4qN 50 11 YIY j. I 44 40 60 6qV ... .... '2qAI %) 'o 7 qW. "Y 2qV so V .. .......... ;qX I... no 70 6o 50 40 30 20 10 6qW. )r) (RES.+ ASPH.) FIGURE 24. Triangular diagram of the distribution by chemical family. For the sake of comparison, we have added the corresponding values in samples of polluted sands and a subtidal sediment taken in the Aber Wrac'h (AW 9). Intertidal Sediments - Polluted Beach Sands The program of analysis employed for the study of these sands is identical to that used for the subtidal sediments of the Aber Wracq'h (Fig. 2). The charaqcteristics of these samples are seen in Table 3. The sulfur contents of these polluted sands (2.5-2.6 %) are aq-11 slightly higher than those of the emulsions even the most advaqrqiced (2.3-2.4 %8q), but are about the same as those in samples taken from the Aber Wrac'h. This evolution is explained by the disappearance of 137 certain chemical species which are easily degradable and/or soluble, such as n-alkanes, light aromatics, etc... leaving a higher relative concentration of sulfurous species, which are mote resistant to degra- dation. Metal concentrations (nickel and vanadium), and their ratio, did not vary significantly through the one-year pariod of the study (Table 9). This behavior was noted above in the discussion of the Aber Wrac'h samples. TABLE 9. Sulfur, nickel and vanadium contents. S Ni V Ni/V % pds Vig/g Pg/9 KE ELD EN I EL 18.10.78 2,55 14 45 0,31 (35 an) TRF.C@ITAN 31.01 .79 2,65 17,5 83 0.2 1 E5 TREO.'TAN 28.03.79 2,66 19 65 0,2 KFADENIEL 28,03.79 2,69 20 85 0.23 X I KERDENIEL 28.03.79 )a 65 0 (25 cm) I As we did with the samples of emulsified mousse, as described above (Fig. 24), we recorded the saturated and aromatic hydrocarbons, and the polar compounds (resins and asphaltenes) on the triangular diagram (Table 8). The most notable evolution took place in the latest samplings of polluted sand. But it is difficult to isolate the factors contri- buting to evolution in the beach sands : time, extent of dispersion of the crude, how long the oil was on the sea, the support material (sand, mud, rocks). This is all the more true of a sampling taken a year after the catastrophe, which may have undergone a very complex history of burial before being picked up again by the water during a storm or a spring tide. Even with these reservations, however, it seems that the triangular diagram shows that the crude follows several pathways in its evolution : - a very short and stable pathway, as we saw above in emulsions on free water ; - a pathway in which a relatively slow disappearance of saturated hydrocarbons (n-alkanes) and aromatics (Mono- and diaromatics) results in a moderate increase in polar products, when the crude is trapped in sand ; - an evolving pathway followed by crude which is trapped in more or less muddy subtital sediments of the Aber Wrac'h, as demonstrated above. The chromatograms of the saturated hydrocarbons in the polluted samples taken in 1979 all show a general degradation of n-paraffins to n-C30, confirmed by an increase in the ratios of isoprenoids to n-alkanes (n-C17 and n-C18). This degradation seems slower in polluted beach sands than in the sediments of the Aber Wrac'h. The mass spectrometry study of the (n+iso) distribution, and that of the 1-- to 6-membered rings of cycloparaffins confirms this 138 evolution (Table 10, Fig. 25) . But a slight alteration is seen in Sur- face samples (Kerdeniel, March 28, 1979), which may be explained by the reemergence of masses of only slightly degraded crude, which have been trapped in sand, during a storm. TABLE 10. Distribution of the cycloparaffins by the mass spectrometry. Z Vol. Z Vol de cyclanes 5 Prqilqivements qParaffines q1 I(n - is.) I cycle 2 cycles 3 cycles q1 4 cycles 5 cycle. 6 cycles PORTISALL 52.92 14,20 3,07 8,20 5,87 2,81 2,33 23.03.78 55,66 14,29 13,04 8,05 5,30 2,07 1,59 23.03.78 1 196q1 ROSCOFF 53,13 14,09 13,50 9,17 6,22 .2.69 1,19 40;72q4.76 PORTS&LL q147,49 14,44 14,90 10,16 7,46 3,31 2,34 4.04.78 q1 GUISSENY 146,33 16,88 15,78 10,43 7.73 2,29 C,55 4.04.78 ERICNOCAN 31.90 22,15 19,72 12,24 8,52 3,64 1,83 4.04.76 TREWeAN 35 17,12 18,13 13,20 9.7. 4,26 2.54 31.01.79 TREOMPAN 30,93 17,75 19,42 14.24 10,21 4,59 2,86 28.03.79 KERDENIEL 44,52 15,58 16,17 10,92 7,52 3,11 2,18 28.03.79 KERDFNIEL 31,88 11.23 15,97 16v07 13,99 6.80 4,06 28.03.79 (25 c= prof.) 'TA r1r" @60 qM F n@o A'J 31 . 0 1 .9 50q- qrqR I qr, 4 qr 1 q7 A 40q- qGqUqjq1qSq9FqPq;Y 4.q04.q7qn 30q- prqiqRTqSALL [email protected] 20. 10q- 7q1.q91.7qn 6qMqITqSALL [email protected]@q,7q1 0 q1 2 3 qNqombrqn dn nqoyqmuqx qInqAqphqOnqn@q) FIGURE 25. DiStributionof the cycloparaf fins by the mass spectrometry. 139 It should be noted that a sample taken at the same time and same place, but at a depth of 25 cm, shows an advanced stage of degradation, comparable to that observed in samples from the Aber Wrac'h at the same time. Compared with these subtidal sediments, oxidation degradation of the trapped crude in beach sands is slight and slow. This is seen in the infra-red spectra of the resins, where the absorption bands of the carbonyls are less marked (Fig. 26). KERDENIEL 18.10.7B TREOMPAN 31.1.79 TREOMPAN 28.3.79 KERDENIEL 28.3-79 3600 3200 2600 1700 1600 1100 750 CM FIGURE 26. Infra-red spectrometry of the resins. CONCLUSION The samples studied fall into three categories - subtidal sediments (Aber Wrac'h) ; - oil/water emulsions or "chocolate mousse" - intertidal sediments (beach sands). In the slightly muddy sediments of fine sand in the stations lo- cated in the outer part of the Aber Wrac'h, where the marine character is pronounced, a decrease in global contents of extractable compounds is observed, whereas in those located in the upper part of the Abers the decontamination process is slow, probably inhibited by the muddy nature of the sediments. 140 The degradationsobserved in these sediments results in - - the progressive disappearance of saturated hydrocarbons, principally the normal paraffins ; - the disappearance of the light aromatic hydrocarbons - the oxidation of the polar compounds (esters, acids, etc...). The compounds which persist are : - the saturated polycyclic hydrocarbons and the heavy aromatics - sulfurous aromatic hydrocarbons of the thiophenic type ; - resins and asphaltenes, resulting in stable metal (nickel and vana- dium) and sulfur contents. In the sediments samples in the area of the Aber, terrigenic de- posits are superposed on the Amoco Cadiz crude, res.ulting in - an increase in polar compounds -- resins and asphaltenes. The ill- crease in asphaltene contents is due to the presence of pigment (green) of chlorophyllaceous origin ; - the appearance of n-alkanes of add carbon numbers (n-C25 throug].-I n-C33). The most striking evolution in the "chocolate mousse" samples is the loss of light hydrocarbons due to evaporation and dissolution. Volatile compounds under C14 were not considered in this report. The samples of polluted sand taken from the beaches a year after the accident show a degradation phenomenon principally affecting the saturated hydrocarbons, and among these, principally the n-paraffins. There is an increase in the contents of polar compounds. But our in- formation is not adequate to state at what point in the history of these samples, the degradation was most intense. 141 REFERENCES CITED Aminot, A., 1981, Actes du colloque, Brest, Nov. 1979 : Amoco Cadiz, Cons6quences d'une pollution accidentelle par les hydrocarbures, CNEXO, Paris, pp. 223-242. Rapport DGMK.. Rapport de recherche 150. Mdthode de diffdrenciation des hydrocarbures biog6nes et des hydrocarbures d'origines pdtroli6res. Ducreux, J., Marchand, M., 1981, Actes du colloque, Brest, Nov. 1979 Amoco Cadiz, Cons4quences d'une pollution accidentelle par les hydrocarbures, CNEXO, Paris, pp. 175-216. Eglington, G., Hamilton, R. J., 1963, The distribution of alkanes. Chemical Plant Toxonomy. T. Swain Ed., Acad. Press, pp. 187-217. Hood, A., O'Neal, M. J., 1958, Preprint of I.P., Hydrocarbon Research Group and ASTM. Committee E14 Joint Conference on Mass Spectro- metry, University of London Senate House, Pergamon Press. Pelet, R., Castex, H., Juillet 1972, Atlas de r6f6rences de pollutions p6troli6res. Rapport IFP n' 22 422. Petroff, N., Colin, J. M., Feillens, N.f Follain, G., Juillet-Aoat 1981, Revue de l'Institut Franqais du P6trole, Ed. Technip, Vol. 36, n' 4, pp. 467-484. Roucache, J., Huc-, A. Y., Bernon, M., Caillet, G., Da Silva, M., 1976, Application de la chroma.tographie couche mince A 1'6tude quanti- tative et qualitative des extraits de roches et des huiles. Revue IFP, Vol. XXXI, pp. 67-98. Roussel, J. C., Gautier, R., 1981, Actes du colloque, Brest, Nov. i979 Amoco Cadiz, Cons6quences d'une pollution accidentelle par les hydrocarbures, CNEXO, Paris, pp. 135-147. Tissot, B., Pelet, R., Roucache, J., Combaz, A., Utilisation des al- canes comme fossiles g6ochimiques indicateurs des environnements g6ologiques. Rapport IFP, r6f. 23 440. Unterzaucher, 1940, Ber. Deut. Chem. Ges., 73 B, pp- 391. 142 THE AMOCO CADIZ OIL SPILL DISTRIBUTION AND EVOLUTION OF OIL POLLUTION IN MARINE SEDIMENTS by Michel Marchand, Guy Bodennec, Jean-Claude Caprais, and Patricia Pignet Centre Oceanologique de Bretagne - CWEXO BP 337, 29273 BREST CEDEX, France INTRODUCTION In March 1978, the supertanker AMOCO CADIZ was stranded on shallow rocks off Portsall .(north Brittany), 2.5 km from the coast. Two hundred twenty-three thousand tons of a mixture of Arabian light crude oil (100,000 t) and Iranian light crude oil (123,000 t) flowed into the sea without interruption from 17 March to 30 March. The maximum extent of the oil slicks is presented in Figure 1. At this point, about 360 km of coastline were polluted by the oil. The analyses of oil in seawater, measured by UV fluorescence spec- troscopy (Marchand and Caprais, 1981), revealed that the oil spill G-W 3* T ,fi AMOCO CADIZ CARTE D'EXTENSION MAXIMALE EN MER DES NAPPES D HYDROCARBURES DU 17 MARS AU 26 AVRIL 1978 FT . ..L..X 0 12 24 56 4C m 'FOUAPINE W&J 48. 48*N C, S, 40 3. FIGURE 1. Maximum extent of oil slicks on the sea surface, 17 March to 26 April 1978. 143 affected a very large section of the western English Channel. One month after the AMOCO CADIZ wreck, the most polluted areas were located in the coastal zones, such as the Aber zone (38.9 + 6.7 ug/1) , the Bay of Morlaix (11. 5 + 5. 1 ag/1) , and the Bay of Lannion (10. 7 + 3. 0 ug/1) . Analysis of samples from various depths revealed that the contamination extended throughout the water column. The 490N parallel roughly con- stituted the northern limit of pollution. Beyond this limit, oil in surface seawater was not observed (1-6 + 0.5 gg/1). In March 1979, one year after the AMOCO CADIZ stranding, hydrocarbon concentrations returned to a normal level (below 2.0 ug/1); however, some residual traces of pollution were still observed near the Abers and at the bottom of the Bay of Lannion (about 2.0 ug/1). We also began a chemical follow-up study of oil pollution in marine sediments. Some data have already been presented during the interna- tional symposium held in Brest (France) in November 1979 (CNEXO, 1981; Ducreux and Marchand, 1981; Marchand, 1981; Marchand and Caprais, 1981). in this document, results of our study are presented in three parts: (1) oil pollution in sediments collected from the Western English Channel one month after the wreck, (2) specific study in the Bays of Morlaix and Lannion to determine the distribution of oil pollution in surface sediments and at various depths, and the evolution of oil contamination over one year, and (3) specific study of the Aber Wrac'h to determine oil evolution from 1978 to 1981. MATERIAL AND METHODS Surface marine sediments were collected in the western English Channel with a Shipek grab. In coastal areas, small Ekman and Hamon grabs were used. The samples, after freezer storage, were dried by using a Soxhlet apparatus or by stirring with chloroform. The organic extract was concentrated to dryness, then dissolved with 10 ml of carbon tetrachloride. A first indication of petroleum pollution in sediment was obtained through a direct analysis of nonpurified extracts by IR spectrophotometry (Perk @n Elmer 397). Quantitative measurements were carried out at 292o cm- corresponding to the presence of hydrocarbons and polar compounds. The data also reflect coextracted natural substances (fats, fatty acids, etc.) from sediments. The IR spectrophotometer was calibrated with a mixture of Arabian and Iranian light crude oils. Hydrocarbons were analyzed after cleanup of organic extracts on activated Florisil (200 OC) in glass columns (15 cm x 0.6 cm i.d.) . Hydrocarbons were eluted with 15 ml of carbon tetrachloride and measured by IR spectrophotometry. For some samples, organic carbon was determined with an auto-analyzer LECO WR-12. In a joint study with the French Petroleum Institute concerning the Aber Wrac'h sediments (Ducreux and Marchand, 1981), we compared the gravimetric determinations and the IR spectrophotometric analysis of nonpurified organic extracts. Results of the two methods are similar (Fig. 2). We also compared the IR spectrophotometric results obtained on nonpurified and purified organic extracts from some Aber Wrac1h sediments. In this case, correlation was significant (Fig. 3). 144 Off E PPM I FIMOCO CRD I Z SEDIMENTS DE L FIBER WRHCH ISM.91 HE-qURE DES EXTRR I TS ORGFIN 1631 IES SRRVIMETRIE CEXTI SPECTROPH13TOMETRIE IR INC] C EXT 3 =0. 971 HC 3 + 13.713 4- C13EFF I C I ENT DE DETERM I NRT I 13N 9. 98 SWIM. + 4. + HC C PRIM 3 O.W I 9.92 ITZ00.23 FIGURE 2. Correlation between gravimetric determinations and IR spec- trophotometric measurements for organic extracts. CHC3 PURIFIES CPPM3 HMOCO CRD I Z SEDIMENTS DE L HESER WRRCH MESURE DES EXTRRITS ORGRNIGUES PRR SPECTROF4iTOMETRIE I.R EXTRRITS NON PURIFIES ET PURIFIES SUR FLORISIL I Soo. OR. + I Omam. 9 + Y = 0.0 X'- 41.69 C13EFFICIENT DE DETERMINATION 0.84 + + CHC3 NMJ PURIFIES 0.00 --4 1 F11 illi I 0.00 1000.00 2200.00 3000.00 H220.20 FIGURE 3. Correlation between IR spec tropho tome tr ic measurements for nonpurified and purified organic extracts. 145 OIL POLLUTION IN THE WESTERN ENGLISH CHANNEL (APRIL 1978) One month after the wreck, sediments were collected during an oceanographic cruise (R/V SUROIT) to assess sea bottom contamination of the western English Channel. The sampled sediments were coarse- to medium-grained calcareous sands (more than 70% CaCo 3 ). In the Bays of Morlaix and Lannion and near the Aber zone, the content of calcium carbonate in the sands was somewhat lower (50-70% CaC6qO Organic carbon content was generally low, from 0.02 to 0.6 percenq? *q(m- = 0.18% + 0.13%). The oil concentrations in the sediment ranged f rom 10 to qf,100 ppm (nonpurified organic extracts) (Marchand and Caprais, 1981). Generally, the zone of contaminated sediments reflected offshore and coastal areas impacted by the drifting slicks (Fig. 4). The pollution of the sea bottom was a result of the diffusion of oil into the water column. Off Sept-Iles, a gradient was observed from the coast q@o the open sea (210, 52, 42, 34 ppqm) . At the 490N parallel, from west to east, one could observe an increasing and decreasing gradient (21, 19, 48, 102, 54, 52, 24 ppm). The highest petroleum accumulation in marine sediments were located in the coastal and sheltered zone of the Abers (100 to more than 10,000 ppm) and in the Bays of Morlaix and Lannion (10 to more than 1,500 ppm) (Fig. 5). -4T3 4*30' 4- 3*30' 3 '30- 0 0 0 7 14 0 2110q@ 0q"0q@2q@34 030 024 30 42o 48 q__a - 50 _4q72q7 - q;q@6qiq4 6qt 2 24 19 *21 q70 540 102 16 *72 219 e24 e26 o64 76 *13 e27 *39 058 o79 64, ,()o 30 0,41 \ 0 L-- 72 7 *1 2 1117 52 0 0 o39 P.1 'P*" 20q@ 7 0q@q:'Lpqz. \ Noll- '3qQ q18 qWq.qSq, A \.A e sediment FIGURE 4. oil pollution i n marin s (April 1978). Concentrations expressed in ppmq. 146 U. d. 8.1. P I. d. DAIE 0q- DAIE DE LANNION DE MORLAIX Cod. P.-Ml Known are*$ of OR accumuLation in Sediments I Abor Benoit 2 Abef Wroc'h 3 Ftmbro do Pt..i 4 Riviire do Mortaix S Point, do Pri,,L 6 Boi* do Lannion FIGURE 5. Known coastal areas of oil accumulation in sediments. BAYS OF MORLAIX AND LANNION (JULY 1978 TO FEBRUARY 1979) A survey was undertaken from July 1978 to February 1979 to follow oil degradation within bottom sediments of the Bays of Morlaix and Lannion. In August 1978, specific sampling of some stations was made by the BRGM. (Bureau de Recherche Geologique et Miniere). At these stations (Fig. 6), several-meter-long cores were taken by vibracorinqg to ascertain the vertical distribution of oil in sediments. Oil Pollution in Surface Sediments The sedimentary description of sediments collected from July 1978 to February 1979 is given by Beslier (1981) and Beslier et al. q(1981q). The weathering of hydrocarbons in some polluted samples was studied by Boehm et al. (1981). In July 1978, the hydrocarbon concentra8qltions (determinations made on purified organic extracts) ranged from 8 pp8qm to more than 1,500 ppm. Average concentrations are presented in Table 1. Complete data are given by Marchand and Caprais (1981). We also used the parameter of total hydrocarbons/organic carbon (HC/OC) to demonstrate pollution of surface sediments. This qcatio (HC/OC x 10 4) ranged from 48 to 6,065. Marchand and Roucache 2q(1981) showed in a study of another oil spill in Brittany (BOHLEN wreck) t0qPqiat a 147 I.I.WEI ILE C...OE Ift 125 IJ6 ILE DE SATZ 4 S.'s VE A 115 M .116 .151 ""@;41A .107 4 1, DE PRIVEL 113 -1:10 A A IN .106 ,:145 -112 SA DE 1AOqLAIA .131 *138 .10 .124 .127 1123 .132 Oc L.RE. LE CALLOT X22 ,21 StMCMEL EN GRIEVES A120 1@ 101 TIE LE- .14 .1 .34 M3- A 100 -AA FIGURE 6. Sampling stations in the Bays of Morlaix and Lannion. TABLE 1. Average hydrocarbon concentrations in surface sediments in the Bays of Morlaix and Lannion (July 1978). A R E A SEDVnT Number of Hydrocarbon observations concentrations (PPMT- BAY OF NY-)PLjklX 25 311 1 418 @brlaix river coarse sand to 6 7-67 1 88 sandy mud A East area coarse sand to around Primel f ine sand 8 600 656 Central area c oarse sand to 7 116 100 West area muddy sand Penz6 river coarse sand to muddy sand 4 14S 96 ---------------------------------------- ------------------- -------------------- BAY OF LVINION 1 22 188 1 213 North West area fine sand 2 41 43 South Ile Crande coarse sand 2 298 WO marsh Central area coarse sand to 9 182 138 fine sand South East area fine sand 9 204 298 148 ratio of more than 100 is an index of oil pollution in surface sediments (Table 2). In this study, only four samples had a ratio.less than, 100; 80 percent of collected samples had a ratio more than 200; ai,'Id 28 percent had a ratio more than 1,000. This simple parameter confirmed that the surface sediments were highly contaminated. Three sedimentary oil accumulation areas were recognized in the Bay of Morlaix: the Morlaix River, the Penze River, and the east area of the bay around Primel. In the Bay of Lannion, pollution was located beyond a line joining Beg An Fry and the Ile Grande marsh (Fig. 5). On the whole, from July 1978 to February 1979, the decontamination process was related to two essential factors: sediment type and the energy level of the geographic zone. Among muddy and fine-grained sands, a slow decrease in oil content was observed, whereas in areas more exposed to high energy conditions, as around Primel in the eastern part of the Bay of Morlaix, the fine- and coarse-grained sand bottom was rapidly cleaned (Table 3). TABLE 2. Ratio of hydrocarbons/organic carbon (HC/OC) in unpolluted and polluted surface sediments (from-Marchand and Roucache, 1981). Number of HC HC LOCATION samples (ppm) r10 REFERENCEES UNPOLLU`1'ED SEDIDNIENTS English Channel (France) - estuary of Seine 3 30 - 40 1 16 - 31 TISSIER, 1974 - bay of Veys 2 31 - 51 26 - 53 TISSIER & OUDIN, 1973, 1975 Iroise sea and bay of 23 3.6-109.5 21 - 70 MAPCHAND & ROLICAGiFl Audierne (France) 1981 NW Atlantic 9 1.3 - 19 10 - 41 FARRINGTON & TRIPP, 1977 Gulf of Mexico 60 1.5-11.7 9 - 23 GEARING et at.1976 ----------------------------- ---------------- -------------- ---------------------- POLLUTED SEDINLEWS English Channel (France) - Estuary of Sein 3 230 - 920 1232 - 430 TISSIER, 1974 - Estuary of Seine 3 70 - 170 276 - 1-83 TISSIER & CUDI(N, 1973, 197S N-W Atlantic (coastal zone) 6 113 - 2900 120 - 180 FARRINGTON & 1,7RIPP, 1977 Bay of Narragansett (USA) 4 350 - 3560 313 724 FARRINGTON & QUVN, 1 1973 Bay of Narragansett (Us,\) 520 - 5410 1350-15590 VA1N VLEET & QUDZN, 1977 Bay of Nbrlaix and Lannion 45 9 1S47 .48 - 6065 present study M : 951tl@137 +a 1-49 TABLE 3. Average hydrocarbon concentrations (ppm) in the sediments in the Bays of Morlaix and Lannion from July 1978 to February 1979. A R E A S -Jul. 1978 November 1978 February 1979 Bay of 1%rlaix 311 418 168 1 246 172 262 Morlaix river 267 88 185 - 82 169 80 Penzd river 155 22 97 East area (Primel) 600 6S6 S3 S4 19 21 Bay of Lannion 281 262 130 113 116 ill Oil Pollution in Core Sediments In August 1978, 24 sever al-meter- long vibracores were taken to as- certain the vertical distribution of oil incorporated into bottom sediments. Complete data concerning sedimentary characteristics'and oil chemical analysis are presented by Marchand and d'Ozouville (1981). A comp.arison of grain size analyses of surface (Hamon) grab and vibracore samples revealed some differences. The vibracoring technique disturbed the first 10 cm of surface sediment layer as expressed by a shortage of fine particles. In another respect, we noticed that hydrocarbon levels in surface layer - sediments were low, from 1 to 50 ppm (Fn = 22 + 19 ppm) , except for two stations in the Bay of Morlaix (AF 132: 106-ppm; AF 139: 174 ppm) - These values were lower than those reported for surface sediments collected by the Hamon grab in July and November 1978 (Table 4). These observations are an indirect confirmation that oil is inclined to be adsorbed on fine sedimentary particles. Hydrocarbon analyses of deep layers in the sediment cores did not reveal any significant vertical penetration of oil. Organic extract (EXT) and/or hydrocarbon (HC) concentrations very often remained homogeneous in the whole of the cores. The most important levels were found in the first 20-30 centimeters, corresponding to the layer sampled by the Hamon grab. A diving survey undertaken during the same time by d'Ozouville et al. (1979) gave the same conclusion: the depth of oil penetration was usually given to less than 7 cm, possibly related to the dep th of biological reworking. Table 5 presents some results showing the absence of a deep diffusion of oil into bottom sediments. Reference data relating hydrocarbon content to sediment type for 17 cores are presented in Table 6. 150 TABLE 4. Comparison between hydrocarbon concentrations (ppm) observed in surface sediments according to sampling techniques. HAMON GRAB VIBRATORY BORING STATION July 1919 November 1978 August 1978 10Z 291 276 36 103 392 228 49 112 914 - so 114 207 2 367 25 4 122 79 - 22.S 124 72 - 16 132 292 56 100 138 1547 130 29 139 60 - 174 151 2T1 79 4 154 - 8 227 < 5 i7ft 403 - 447 132 t100 41 - 6 TABLE 5. Hydrocarbon concentrations (ppm) in some sediments cores from the Bays of Morlaix and Lannion. Stationj A F A F A F A F Pth 112 122 132 139 (cm)@ 0 - 10 22.5 100 174 10 - 20 so < 5 20 - 30 < 2 11 30 - 40 < 2 19 40 - 50 4 < 5 < 2 SO - 60 60 - 70 < 2 < S 151 TABLE 6. Average concentrations of organic extracts (EXT) and hydrocarbons q(qH4qQ in sediment cores collected from the Bays q(q)f Morlaix and Lannion. Depth of (EXT) (H 1 oring (m) Sediment n PPM q(n) pp C4ql -_ q1 Al: It)' I.S mud (a) 231 � 80 (7) 26 � Is AF 10-) 7.2 mud (14) 220 � 73 q(11) 23 � 23 Aq@- q@04 3.2 medium to coarse sand and (8) 16 � 15 ma6rl A: 12 4.0 fine to coarse sand - (6)q(1q0 < 2 - 4 AI: 114 1.3 medium to coarse sand and - (S) < 2 - 4 ma6rl A I1 6.8 fine sand - AI: 1-'4 S.5 fine to coarse sand - 8 < 2 - 6 2.1 mediLIM to fine sand (8) 14 i 9 q0q0 AI 1 1.5 matrl and silt (3)(2) 26� 4 3 < 2 I. 13 s 2.1 fine sand 7 (') 14� 7 I 13 1-) 2 .7 fine to coarse sand - 6 < 2 A 1137 2.0 mediunt to fine sand (9) 10� 10 Al I@Sl 2.0 fine to coarse sand - (8) < 2 5 A[ It,-, 2.1 medium to fine sand (7) 13i 3 AF ISU 2.3 coarse to fine sand (7) 26� 6 Al' 191 Z.S medium to fine sand (7) 10� 6 AF 237 2.3 fine sand to sandy mud (6) S excepted surface layer, (n) number of observations, (EXT) IR spectrophotometric determinations of non-purified organic extracts, (H2qQ IR spectrophotometric determinations of purified organic extracts on florisil. ABER WRAC'H (1978 TO 1981) The two Abers (Benoit and Wrac1hq), located 8 km east of Portsall, were heavily impacted during the spill. These Abers are small estuaries, 10-15 km long and about 1 km wide. A study from March 1978 to March 1979 (Marchand and Caprais, 1981) revealed that the bottom sediments throughout the Abers were heavily polluted. Concentrations were more than 100 ppm and, as in a muddy area in the Aber Benoit, sometimes reached higher than 10,000 ppm. After one year, Marchand and Caprais (1981) showed that the natural decontamination process was related to the nature of the sediment and the energy level of the geographic zone. The fine- and medium-grained sands located in the exposed, downstream part of the Aber Beno *it were well decontaminated (average hydrocarbon content reduced from 700 to 27 ppm). On the other hand, in mud-dominated areas, the sediment acted as an oil trap (oil content above 10,000 ppm) and decontamination was not observed. For the Aber Benoit, oil pollution of mud-dominated zones such as Loc Majan will be long term. In the Aber Wrac'h, the evolution of oil pollution in the bottom sediments has been followed since March 1978. The location of sampling stations is presented in Figure 7; analyses for hydrocarbon and organic carbon concentrations are presented in Tables 7 and 8, respectively. Organic carbon concentrations ranged from 0.08 to 3.32 percent. Sediments are much more homogeneous than those of the Aber Benoit. Composition, with the exception of station 3 located at the mouth of the Aber, varied from slightly muddy to muddy sands. 152 FIGURE 7. Sampling stations in the Aber Wrac'h- ABER WRAC'H 04 LANq6EDA IKM 8 TABLE 7. AMOCO CADIZ oil pollution in the sediments of the Aber Wrac1h. Data collected from 1978 to 1981. OC = organic carbon; CaC6qO = carbonate calcium; EXT = gravimetric determination oql organic extract; OIL = IR spectrophotometric determination of nonpurified organic extract; HC hydrocarbons, IR spectrophotometric determination of purified organic extract; (Sq) = surface; (P) = 10-15 cm depth. DATE T SAMPLING (mon tha) 9TALTION 1 2 3 4 5 6 78 9 March 31 Oc M 0.53 2.42 0.23 0.96 1.06 1.03 0.66 1.94 0 1978 0.5 EXT (ppm) 2130 1122o 711 2185 3160 765 -397 19 1 OIL (ppm) 2051 12000 773 2450 37o6 839 2259_ 953 2063-1 CaCO 3 2 - - - - 18.8 20.8 7.06 1.5 Oc 0.66 - 0.34 1.59 0.71 0.78 0.610 May 5, q:15 1.;, 197 EXT 2400 - 800 11490 2000 1660 1190 460 216o OIL 3023 - 1020 117SO 2970 1380 1236 503 2-1)b November 22o EXT q;90 4030 q:30 3740 2600 718.) 1600 t,80 85,, 1978 8.25 OIL 1109 4144 48 3598 2481 7 1 1105 671 87.1 February 22. Oc 0.56 f-53 - 0.75 n.85 1.12 1.00 O.bO - 1979 11.25 KXT 1740 266o so 1340 1360 3 02 0 10110 870 178o OIL 11189 2679 113 1301 1268 21,56 1410 12h6 1677 June 20, EXT 1550 1200 80 480 1450 6070 1260 1450 $50 19, 9 15.25 OIL 11458 1124 74 1,12 1)178 4900 1325 1445 '4@g October 22. 1979 19.25 OIL 715 1314 BO 1237 1047 2712 2496 485 5q0 Januq7ryo20. 98 22.25 OIL 1408 2567 - 1796 1473 1861 1399 5694 Ioq' June 1980 Oc 0.41. 1.74 0.08 1.15 0.81 ?.4o "90 .67 @I 0o 27 011. 442 1892 42 122q@ 1075 1788 11 q;ht- I So IIC (PPm) 175 991) 48 11'22q-q- 602 ;Qt% qOqc qoq'q8q3 0. qFq14 q0. 9q0 q0q.q7q0 q2q. q1q1q) q)qZ q(qSq) q1q). .1 q1q1 January 1981 34 q?q.q1q'q!0q'q)q2q. qIQ(qP 1 q;0q(q0q' qoqll. 321 562 347 456 23q2q0q(qS)3Iqb7(S'8q;8q;q7qtqiq(qS)6qj 3q-q1q4 qIq70qq(qIq'q)6qP3qh9(qIq" qIqiqZqiq(qrq) I( ) q'(S: q(q' qo q( q)6q! q1q1qC 62 255 q105 20q5 8q3 qS 11q64. 6q1q3 qs q21q-q'q. q7 00qM I -( :4q@qrqv 2q- OIL 432 1 o 7q,q4 (S) 25 970(S) 6q5q1q0qzq) 1517(qs) Igo q9q4q"q(qSq) q9 1q9 q(qS I I March 16., 30q3(qp) 6 q2 7 ( 1q1) q14q71q0q' q20 `0 q( qP) 7 1q1) 1q1.3q, qkqrq) q1981 16 HE: 229 431(S) 551(Sq) q12qR I q(Sq) q674 (qS) 40 4 qIqIqSq(S) 31q2(S) % 3r q@q@08qTq'q.82q1 q3 q64 90q(q1q'q) 40 (Pq) 6qFq8 7 (q1q) 722 qP) 1 q'q2 P 707kqr) ;80qJ qa 23, Olt. 308 q8q5q1q(qSq) 41q3(S) 'q8q64 (qS) q1 qh23 S) q540(qS) q1q1q7(qS) 4q11 qUq. 19,25 q1q,04(q?q) q95(P) 541(P)1320q(P) 96 581 HC I qI(qP) 331(P) 50 362q(S) q108q(S) q(q1q6(S) 74qt) qS)2ql 2q00(Sq)q)8ql 135q(S) 326 q171 (qP)ql 42q(P) 486 q(-qPq)q-q660 qPq)q. 4.17 6qQ31(P) 153 TABLE 8. Organic carbon content of the Aber Wrac'h sediments. (n) number of observations. SAMPLING STATION. n OC(%) S 1 5 0.60 t 0.15 (25 2 4 1.63 � 0.65 (40 3 3 0.22 � 0.13 (59 4 S 1.07 � 0.32 (30 5 5 0.83 � 0.15 (17 6 6 1.16 � 0.88 (S4 7 6 1.50 � 1.14 (76 %) 8 6 1.22 t 0.87 (71 %) 9 4 0.90 � 0.12 (14 %) In this discussion, three areas of Aber Wrac1h are described in terms of oil degradation:: (1) the mouth of the estuary (fine-grained sands, station 3), (2) the downstream part (stations 1, 2, 4, and 5), and (3) the upstream part (stations 6, 7, 8, and 9). The evolution of oil pollution in sediments from 1978 to 1981 for each station is given in Figure 8; the change in average oil concentrations for each of the three defined areas is given in the Table 9 and Figure 9. In March 1978, 15 days after the AMOCO CADIZ wreck, concentrations ranged from 773 ppm to 12,000 ppm. At the mouth of the Aber which is well exposed to high marine energy, hydrocarbon content dropped from 773 ppm in March 1978 to 25 ppm in March 1981, illustrating a fairly rapid decontamination of these fine-grained sands. Since the sediments of the Aber Wrac1h are relatively homogeneous, the decontamination process is mainly related to the energy level of the zone. In the downstream part of the Aber, the sediments were more polluted in March 1978 (average about 5,000 ppm) than the sediments collected in the upstream part (average about 1,500 ppm). For the first 39 months after the spill, a' natural but slow decontamination was observed with some temporary increases in January 1980 (about 1,800 ppm) and March 1981 (about 780 ppm). In June 1981, the average residual oil content was about 600 ppm. In the upper part of the Aber, the sediments were initially less polluted, but since this is a low-energy area, a decrease in hydrocarbon content was not observed until January 1981. As had been observed in the downstream portions of the Aber, significant increases in hydrocarbon levels were observed during January 1980 in upstream areas. Since March 1981, oil levels have decreased; average residual oil concentrations were about 670 ppm. Three years after the wreck, the petroleum pollution of the Aber Wraclh sediments seems to be relatively uniform. Figure 10 gives the observed decontamination rate of the sediments. At first approximation, station 3 (located at the mouth of the Aber) is well decontaminated (3% of the initial residual oil level observed in March 1978). However, sediments within Aber Wrac1h remain quite polluted with about 20 percent of the residual oil content still remaining in March 1978. 154 110111 M HC EFIF"I EVMJJT113N DES TENEWS RESIDUELLES D WDR13CMBLRES DRG LES SED I MENTS DE L MEER WRFKH In 002. 6 I 7 49 I' 8 3 T EM153 18 i i i 3 a 9 12 Is Is 21 24 27 30 33 3155 39 42 45 FIGURE 8. Evolution of residual oil pollution in Aber Wrac1h sediments. TABLE 9. Evolution of oil pollution (ppm) in the Aber Wrac1h sediments from 1978 to 1981. DA I t. C' Ma r I ch 311 Ma Nov,22. Fchr.22, Jun 20, Oct 22.@ Jan-20 Juno, lianuary Imarch 16 June Z3 197 1y 198 198 0 1981 198 (month! 8 978 19 8 1979 1979 19 9 0 1 1981 LOCATION __O.5 1 .5 8 . I's II . .15 15.2S 19. 2S 19. 25 22.21 34 36 39.2S Month (St. 31 773 IWO 148 113 74 80 - 42 - 25 - Downsti cai:i p.i i t nos 1 5914 29 15 1709 1043 tO93 181 1 1158 421 783 609 !@168S t5053 11.103 16ol �445 !285 t531 �595 !110 129S 1290 Upstieam 1,.,it 15.18 1340 857 17" 2149 .:565 .25412 1390 .:747 9001 t672 102 (St. b", S'! !736 �708 184 5 7 8 1646 - -21 3 t409 - 250 6S7 Ali stat qt, ut 1w Aber 1,:- 'W 3290 3300 1986 1718 IS9(j 13Z9 2161 1274 1084 842 b40 tj "t .3(.:;l .3839 3S8 �575 1377 �847 t14193 t489 1085 W3 t472 155 100 EM. 14C CPM3 EVMAIT113N DE5 TENEI.M RE51DLEAM MOYENNES D HYDFMCRRMAqE3 WINS LES 901MOT5 DE DIFFERENTS SECTEURS DE L FIBER WFIRCH to 000.. 3206 33 SFM.ES VRSEUX 10 0 F1MM4T CST Se7o9tgl T13UTES 5TRTIONS RVFL CST h2s4oU 41 too 13 SO SRELES FINS EMSMJCHURE CST 33 113 T EMOIS3 3 6 9 12 is Is 21 24 27 30 M 35 39 H2 4S FIGURE 9. Evolution of average residual oil pollution in sediments from the Aber Wrac1h. C P"CENTFSE I 100.110. TMD( DE DECIINTRMINRT113N DES SEDIMWS DE L FIBER WRFKH 130.1111 15111.013 52 46 LIM.110 40 39 33 25 FIBER WFIRCH ST h2s4tSi6i 20.00 to 71819 3 14 to 10 EM513IJOiURE EST 33 2.010 6.00 12.00 18.03 24.06 30-00 36.00 42. 00 T EM0153 FIGURE 10. The rate of oil decontamination in sediments from the Aber 064 3 f 3 Wrac'h. 156 ACKNOWLEDGMENTS We would particularly like to thank Dr. Cabioch and personnel of the Station Biologique in Roscoff (France) for their cooperation and assistance throughout the work performed in the Bays of Morlaix and Lannion. The support of field scientists Serge Berne, Laurent d'Ozouville, and Anne Beslier is also gratefully acknowledged. This work was supported by the Ministere de 1'Environnement et du Cadre de Vie (France) and the National Oceanic and Atmospheric Administration (U.S.A.). REFERENCES CITED Beslier, A., 1981, Les hydrocarbures dans les sediments subtidaux au nord de la Bretagne: distribution et evolution: These de 3e cycle, Universite de Caen, 204 pp. Beslier, A., J. L. Berrien, L. Cabioch, J. L. d'Ouville, Cl. Larsonneur, and L. Le Borgne, 1981, La pollution des sediments sublittoraux au nord de la Bretagne par. les hydrocarbures de 1"AMOCO CADIZ: dis- tribution et evolution: Cf. CNEXO (1981), pp. 95-106. Boehm, P. D., D. L. Fiest, and A. Elskus, 1981, Comparative weathering patterns of hydrocarbons from the AMOCO.CADIZ oil spill observed at a variety of coastal environments: Cf. CNEXO (1981), pp. 159-174. CNEXO, 1981, AMOCO CADIZ, fate and effects of the oil spill: Colloque International, Centre Oceanologique de Bretagne, Brest (France), 882 pp. Ducreux, J. and M. Marchand, 1981, Evolution des hydrocarbures presents dans les sediments de l'Aber-Wraclh d1avril 1978 a juin 1979: Cf. CNEXO, pp. 175-216. Marchand, M., 1981, AMOCO CADIZ, bilan du colloque sur les consequences d'une pollution accidentelle par hydrocarbures, Brest, november 1979: Publ. CNEXO, Rapp. Scient. et Techn. n 0 44, 86 pp. Marchand, M. and M. P. Caprais, 1981, Suivi de la pollution de l'AMOC0 CADIZ dans lleau de mer et les sediments: Cf. CNEXO, pp. 23-54. Marchand, M. and L. d'Ozouville, 1981, Etude de la pollution par les hy- drocarbures de VAMOCO CADIZ des sediments des baies de Morlaix et de Lannion: Rapport de Contrat CNEXO/NOAA, Centre Oceanologique de Bretagne, Brest (France), (in press). Marchand, M. and J. Roucache, 1981, Criteres de pollution par hydrocarbures dans les sediments marins. Etude appliquee a la pollution du BOHLEN: Oceanologica Acta 4(2), pp. 171-181. 157 AMOCO CADIZ POLLUTANTS IN ANAEROBIC SEDIMENTS: FATE AND EFFECTS ON ANAEROBIC PROCESSES by David M. Ward Michael R. Winfrey I, Eric Beck I and Paul Boehm 2 1) Department of Microbiology, Montana State University, Bozeman, Montana 59717 2) Energy Resources Company, Inc., Cambridge, Massachusetts 02138 INTRODUCTION It was estimated that much of the oil spilled after the wreck-of the AMOCO CADIZ impacted intertidal and subtidal sediments (Hann, et al, 1978; Gundlach and Hayes, 1978). Considerable differences exist between sediment and aquatic environments which could have dramatic effects on the persistence of spilled oil and its effects on the native biology of coastal environments. Recent investigations have shown that intertidal and subtidal sediments are anaerobic except in the initial few millimeters near the surface (Sorensen, et al, 1979; Revsbech, et al, 1980a, b) . Since oxygen is known to be of extreme importance in the microbial biodegradation of hydrocarbons (Atlas, 1981; Hambrick, et al, 1980; DeLuane, et al, 1981; Ward and Brock, 1978) it is likely that the persistence of hydrocarbons would be much greater in anoxic sedi- ments. This could create a source of relatively unweathered petroleum for secondary pollution events. One of the majpr objectives of this study was to investigate the extent of pollution by AMOCO CADIZ oil in anaerobic coastal sediments. Evidence for weathering and potential biodegradation of sediment hydrocarbons under aerobic and anaerobic conditions was also obtained in chemical and microbiological studies. Sediments are also important sites where extensive mineralization of organic mat 'ter and recycling of nutrients occurs (Fenchel and J$r- gensen, 1977). The effect of oil on sediment microorganisms and pro- cesses has been examined in some studies (Walker, et al, 1975; Knowles and Wishart, 1977) but only a few studies have examined the effects on mineralization (Griffiths, et al, 1981, 1981 (in press)). Because of the extreme thinness of the oxygenated zones of coastal sediments an- aerobic processes are important in mineralization and nutrient recycl- ing (Sorensen, et al, 1979). The effects of oil on anaerobic processes have not been studied. Many studies on sediment chemistry and microbio- logy support the model for anaerobic microbial food chains in marine sediments presented in Figure I (see Mah, et al, 1977; Fenchel and Jor- gensen, 1977; Rheeburg and Heggie, 1977,; Bryant, 1976). Within anaero- bic zones polymeric organic matter is fermented principally to H 2' CO 2 and acetic acid. Acetate and H 2 are the main energy sources for spec- ialized anaerobic bacteria which terminate the food chain. The activi- ty of these terminal groups is thought to be important in 'influencing fermenting bacteria to produce mainly acetate, H 2 and CO (Bryant, 1976). The importance of these energy sources in marine seAments has 159 POLYMERIC ORGANIC MATTER FERMENTING BACTERIA V CO 2 + H 2 +ACETATE METHANE-PRODUCING SULFA TE-REDUCING BACTERIA -so + BACTERIA 4 S 04 *C H4 + CO 2 *C 0 2 + H 2S Figure 1. Simple model for anaerobic microbial food chains in marine sediments. The model does not include minor electron sink fermentation products or other possible anaerobic groups such as denitrifying bacteria. The asterisk indicates the major product from the methyl position of acetate. only been confirmed in this and other recent studies (Winfrey and Ward, submitted; Sorensen, et al, 1981; Banat and Nedwell, personal communi- cation). Sul f ate- reducing and methane-producing bacteria share the potential to utilize these fermentation products and may compete in marine sediments. Detailed investigations to characterize anaerobic processes in Brittany sediments were made as a part of this study, but discussion here is beyond the scope of this report. In summary, meth- R anogenesis is only significant in the competition for acetate and H 2 in sediments where sulfate is depleted (e.g., deep subtidal sediments, Winfrey, et al, in press). In intertidal sediments where sulfate is high at all depths (see below), methanogenic bacteria may be restricted 160 to energy sources other than acetate or H (such as methylamines) , and sulfate reduction dominates as the signiAlcant terminal process (Win- frey and Ward, submitted). As another major objective of this study we examined terminal processes of the anaerobic food chain, which should indicate the activity of the overall food chain, for evidence of changes due to AMOCO CADIZ oiling. The processes investigated were those which dominated the food chains, principally sulfate reduction and acetate metabolism. METHODS Location,of Sampling Sites Sites were selected to represent beach, estuary and salt marsh sediments in the lower intertidal region which were significantly oiled by AMOCO CADIZ pollutants. Similar sites in areas unoiled 'or lightly polluted were selected as controls. Locations are shown in Figure 2. Observations on the chronology of oil movements were used to determine the extent of oil impact and times at which oiling first occurred (Centre National pour l'Exploitation des Oc6ans, 1979; Gundlach and Hayes, 1978). The oiled beach site was between stations B and C at AMC-4 (Gundlach and Hayes, 1978). This site was opposite the wreck and was heavily oiled immediately after the spill. The control beach site was 100 m east of the west access to the beach at Trez-hir. This beach faces the Bay of Brest and was not reported to have been oiled. The oiled estuary site was a mudflat at Aber Wrac'h, 200 m south of the stone wall at EPA-7 (Calder, et al., 1978). AMOCO CADIZ oil was de- tected in this area in medium thickness two days after the oil spill. The control estuary site was a mudflat on the south bank of Aber Ildut approximately 3 km west of Breles. Oiling at this site was prevented by two booms extended across the mouth of the Aber. The oiled marsh site was an intertidal mudflat in Ile Grande near AMC-18 (Gundlach and Hayes, 1978). Thick accumulations of AMOCO CADIZ oil reached this site bylthe eighth day following the spill. The control marsh was a natural mudflat in the Ile Grande marsh, just south of the main intertidal channel. A barricade at the bridge adjoining east and.west marsh areas prevented oiling at this site. Sample Collection and Processing Sediment cores were collected with hand-pushed plexiglass tubes (60 cm x 37 mm ID), stoppered, and transferred to the lab in an upright position. Cores for sediment hydrocarbon analysis were kept frozen until processing. Processing of samples for biological activities was done at the Centre Oc&anologique de Bretagne in Brest or at the Station Biologique at Roscoff within 8 h after collection. All manipulations for analysis of biological activity were carried out using strict an- 161 ILE GAAkDE ABER AMOCO CADIZ MORLAIX STUDY r AREA BREST FRANCE T111 "IR > Figure 2. "Locations of sampling sites. aerobic techniques designed for cultivation of bacteria with extreme sensitivity to oxygen (Hungate, 1969). Cores were sectioned into de- sired intervals and subcores removed by a No. 4 cork 2borer or by a 3 ml syringe with the end of the barrel cut off (50.3 mm ). The 0-3 cm in- terval was used for all the oil and mousse addition experiments. For these experimen 'ts an anoxic slurry was made by mixing the core section with 20% (V/V) anoxic artif 'ical seawater (ASW, Burkholder, 1963). In experiments on hydrocarbon metabolism slurries of other depth intervals were made in the same 'way. Subsamples (2.0-2.5 ml) from core sections or slurries were tran sferred to 2 dram glass vials (Acme Vial and Glass Co.) and sealed under, a stream of helium. The helium was passed over heated copper' filings to remove any traces of oxygen. Vials were sealed with 00 butyl rubber stoppers (A.H. Thomas). Unless noted be- low, all isotope additions (1.0 ml) were taken from sterile anoxic stock solutions with a 1 ml helium flushed glass syringe (Glaspak). Mousse, oil and hydrocarbon additions were added to vials containing ,sediment under a flow of helium gas using a I ml pipet 12 hours before microbial activities were assayed. Benzene and toluene were added with a 5.pl syringe (Hamilton). Measurement of Microbial Activities All incubations were done at ambient temperatures (20-24"C). Com- parisons between different samples were made by a two sample t test, using the ANOVI program of MSUSTAT (Lund, 1979). 162 Sulfate Reduction Sulfate reduction assays were set up using a modification of the technique of Ivano35 (1964). Each vial of sediment received approxi- mately I pCi of Na SO in 1 ml of anoxic ASW. Samples were mixed and incubated for 2.0 1. Ae reaction was stopped by the addition of 0.5 ml of 2% zinc acetate followed by 0.2 ml of formalin. Samples were assayed and rates determined in @4e Montana State University lab as described by JOrgensen (1978). H 2 S was distilled to traps containing 2% zinc acetate. Radioactivity was determined by counting the zinc acetate trap (5 ml) in 10 ml Aquasol (New England Nuclear) on a Beckman LS-100C liquid scintillation counter. Correction for quenching was by the channels ratio method. Methane Production Methane production was measured by quantifying the increase in methane in the head-space of vials containing sediment. A 0.2 ml gas- eous subsample was removed at desired intervals -and analyzed by flame ionization gas chromatography (see below). Acetate Metabolism 4Acetate metabolism was measured by adding approximately 0.5 pCi of 12-1 Cl-acetate in 1.0 ml of sterile anoxic sulfate-free ASW. Samples were mixed and incubated for 2.0 h unless otherwise statei-4 The reac- @4on was stopped by the addition of 0.2 ml formalin. CO2 and/or CH 4were measured in samples of the gas headspace (see below). Hydrocarbon Metabolism All radiolabelled hydrocarbons except benzene and toluene were di- luted in benzene to the desired activity. The radioisotopes were added to vials and the benzene allowed to evaporate completely before addi- tion of sediment and anaerobic tubing as described above. Anoxic ASW (1.0 ml) was added to each sample to mix sediment and radioisotopes. Radiolabelled benzene and toluene were dissolved in anoxic ASW and added (1.0 ml) after anoxic tubing of sediment. When indicated, sam- ples were incubated in darkness by wrapping with aluminum foil or electricians tape. In one experiment samples contained in anaerobi- cally sealed tubes were incubated within an anaerobic chamber (Gaspak) with a H2 14CO 2 atmosphef4. During long term incubations gaseous meta- bolities ( CO and/or CH4) were quantified in samples of the gas headspace as Ascribed below. After incubation was completed, samples were poisoned by addition of 0.5 ml 10% formalin. Carbon dioxide was reabsorbed by addition of 2 ml 2N NaOH. The sediment was extracted four times by vortex mixing with 6 ml methylene chloride: methanol (9:1) followed by centrifugation to break the emulsion. Solvent frac- tions were removed from beneath the aqueous phase and pooled together with three 6 ml rinses of the original sample vial. Anhydrous sodium sulfate was added to dry the sample. The extract was concentrated to 0.1 ml by evaporation and the volume increased by addition of 0.7 ml of hexane. This sample was separated into saturate (f 1), aromatic (f 2) 163 and methanolic (f ) fractions by silica gel chromatography as des cribed for oil samples gelow. Each solvent fraction was concentrated by ro- tary evaporation to 4 ml and radioactivity of a portion of the extract was determined in Aquaso as described above. Using these methods [1-14 Cl-hexadecane and [1 `14 Cl-heptadeff,ne standards were recovered in the fraction, wh 74 eas [1(4, 5, 8)- Cl-naphthalene was recovered in the f2 fraction. CO was transferred from the extracted sediment after acidification anY distillation to a CO absorbant trap (Carbo- sorb, Packard). This trap was combined with Rquasol and radioactivity determined as described above. Gas Measurements Gas subsamples (0.2 to 1.0 ml) were removed from the headspace of incubation vials using a helium flushed 1 ml glass syringe (Glaj&ak) fittq@ with a Mininert pressure-lock syringe valve (Supelco). CH 4 and @02 were measured by gas chromatography- gas proportional count- ing using the method of Nelson and Zeikus (1974) as modified by Ward and Olson (1980). This method ensured the specific detection of these gaseous metabolites. All numerical results were' based on amounts clearly above detection limi Y4 and quantifiable by integration using a Spectra-Physics Minigrator. CO2 values were corrected for CO solu- bility and bicarbonate equilibrium as described by Stainton @1973). Methane concentrations were quantified on a Varian 3700 flame ioniza- tion gas chromotograph as described by Ward and Olson (1980). All values for gas analyses are reported on a per vial basis. Chemical Analysis of Oils Similar methods were used for the analysis of sediment hydrocar- bons (ERCO) or oils (MSU) although the specific details differed. Sediment hydrocarbon samples were solvent extracted and fractionated according to an analytical scheme patterned after that of Brown et al (1980). Hydrocarbons were separated from methanol-dried samples by high-energy shaking with methylene chloride: methanol (9:1), fraction- ated into saturate, aroma tic/unsaturate and methanolic fractions by alumina/silica gel column chromatography, and analysed by gas chroma- tography, mass spectrometry and/or mass fragmentography as described by Boehm, et al (1981). AMOCO CADIZ mousse, fresh or evaporated crude oil, and extracts from 14C -hydrocarbon experiments (see above) were analyzed as follows. Each sample (a 25 pl aliquot of each oil sample diluted in 0.5 ml of hexane, or extracts described above) was loaded onto a glass column (1 cm ID x 20 cm long) packed with 40-140 mesh silica gel (Baker Chemical Co.). Saturate components (f 1) were removed from the column by eluting with 200 ml of hexane. The aromatic fraction (f 2) was then eluted with 200 ml of a solution of hexane and methylene chloride (70/30 V/V). The methanolic fraction (f 3) was then eluted with 150 ml of methanol. Each fraction was concentrated to 3-4 ml by flash evaporation and further 164 concentrated to less than 1 ml under a stream of nitrogen. The volume of each fraction was adjusted to one ml with the appropriate solvent. A 0.2 pl subsample was injected into a Varian 3700 flame ionization gas chromatograph equipped with a 30 m glass WCOT column packed with SE-54 (Supelco). Operating conditions were ' as follows: column temperature programmed f rom 60 C to 260 C at 3 C/min with a 30 min hold at 260 C; injection temperature: 250 C; detector temperature 250 C; carrier gas: 1 ml helium/min with a helium make-up gas of 29 ml/min. Results were recorded on a Spectra-Physics model 4100 recording integrator. Hydro- carbons labelled in Figure 9 were identified by comparison of retention times to those of pure hydrocarbon standards. Sediment Chemistry Eh Duplicate sediment cores for Eh measurements were collected using a plexiglass tube (30 cm x 25 mm ID) which had been split lengthwise and taped together. Upon returning to the laboratory, one half of the core liner was removed, and fresh sediment exposed 1 cm at a time by slicing the core lengthwise with a spatula. Eh measurements were taken by pressing a combination platinum electrode (Orion) into the,freshly exposed sediment surface. All Eh values are reported relative to the normal H2electrode. pH pH was determined using a VWR pH Master pH meter and glass com- bination electrode. Interstitial Water Sediment porewater was obtained using the porewater squeezer of Kalil (1974). After appropriate dilutions were made, sulfate was meas- ured by the turbidometric method of Tabatabai (1974) and chloride was measured by silver nitrate titration (Am. Pub. Health Assoc. , 1976). Methane Dissolved methane was quantified by killing a 2 ml subcore in a sealed vial by the addition of 0.5 ml of formalin and mixing on a vor- tex mixer to strip the dissolved methane into the headspace. A gas subsample was then analyzed by flame ionization gas chromatography as described above. Radioisotopes, Chemicals, and Oils The following radio*ive chemicals were used (radiochemicall ur- ity in parentheses): Na so 738,4mCi/mmole (on 3/5/79), Na-[2- Cl- 2 4' acetate, 44 mCi/mmole, and [rJ ng-1- CI-toluene, 3.4-5.2 mCi/mmole (97- 165 99%) from New Englj@d Nuclear; n-(l- 14 C]-hexadecane, 54 mCi/mT41e (97- 99%), [1(4, 5 , 8)- Cl-naphthalene, 5 j@i/mmole (97-98%), [U- Cl-ben- zene, 101 mCi/mmole (98-p@%), [7, 10- C]-benzo[alpyrene, 60.7 mCi/m mole, (99%), and [Whyl- C]-toluene, 30 mCi/mmole (96-99%) from er- sham Corp.; n-[l- Cl-heptadecane, 16 mCi/mmole (>99%), and [1 "Cl- heptadecene, 18.5 mCi/mmole (97%) from ICN. AMOCO CADIZ mousse was obtained from the NOAA National Analytical Facility and was collected at Ploumanach on April 30, 1978 (44 days after the spill). Light Arabian crude oil (SX#0308) was obtained from Exxon Corp., Baytown, Texas. The crude oil was weathered by evapora- tion at 25 C for 8 and 48 h. All' oil samples were stored at 4 C until used. RESULTS Physical-Chemical Comparison of Sites Beach cores consisted of medium grained sand, while estuary and marsh cores consisted of fine grained silt and clay. It was possible to determine Eh profile for muddy sediments (Fig. 3). In all sediments Eh (mv) -100 0 +100 +200 +300 -100 0 +100 +200 4300 0 1 . ____ 9 .1 1 1 0, @a I A k1G0 0 AW E 4- IG 0 W 6- 0 SALT MARSH MUDFLATS ABER MUDFLATS Figure 3. Eh profiles in muddy sediments. Bars indicate the range of measurements on duplicate cores. pH increased with depth from 7.5 to 8.2 in Ile Grande, from 6.7 to 7.2 in Aber Ildut, from 5.9 to 7.5 in Aber Wrac'h and from 7.0 to 8.1 in the Ile Grande oiled site. conditions became more reducing with depth. The steepest Eh Orofile was observed in the Ile Grande oil site where a brown layer approxi- mately 2 mm thick covered black sediment. Marsh mudflat sediments 166 showed steeper Eh profiles than estuarine sediments and oiled sediments were more reducing and showed steeper Eh profiles than unoiled sedi- ments of the same type. The large Eh change with depth (300mv over 1 cm (Ile Grande oiled site) or 2 cm (Aber Wrac'h) suggested that sedi- ments below these depths are likely anoxic. Chloride (data not presented) was relatively constant at all depths in all sites and was near seawater chloride levels (20,000 mg/liter). The concentrations of dissolved sulfate and methane with depth in each site are reported in Table 1. No major differences in sulfate concen- tration with depth or between cores were observed,and levels were similar to seawater values (approximately 800 mgSO 4-S/l ).. Methane TABLE 1. Sediment Chemistry a SULFATE CHEMISTRY b DEPTH AMC 4 TREZ-HIR ABER WRAC'H ABER ILDUT ILE GRANDE ILE GRANDE (cm) (oiled) (control) 0-5 720 720 860 840 840 760 5-10 850 770 900 830 790 800 10-15 820 820 870 970 750 760 15-20 1070 1190 820 810 740 790 20-25 860 920 800 710 METHANE CHEMISTRY C 0-5 0.17 0.16 0.74 1.57 3.36 0.24 5-10 0.52 0.84 0.80 0.89 2.89 1.09 10-15 0.64 0.67 0.79 0.66 3.34 0.26 15-20 0.99 0.93 4.26 0.35 20-25 0.50 0.81 4.36 0.36 aSediments collected in March 1979 bResults expressed in pg so 4 -S/ml porewater cResults expressed in pmoles CH 4/1 sediment 167 concentrations were extremely low (less than 5 pmoles/1) and relatively constant with depth. Methane concentrations were significantly higher at the Ile Grande oiled site (p < .001). Sediment Hydrocarbons Sediments collected during December 1978 and March 1979 were ana- lyzed for hydrocarbon content (Table 2) and type (Figs. 4 and 5). Sur- face sediments (0-5 cm) at all sites oiled with AMOCO CADIZ oil exhi- bited a composition indicative of highly weathered oil residues. The saturate fractions were comprised of a degraded hydrocarbon assemblage with greater degradation in estuary and marsh mudflat samples than in the beach,sample as evidenced by the relative dominance of the branched isoprenoid hydrocarbons (Fig. 4). Residual alkylated phenanthrenes, and dibenzothiophenes in the aromatic/unsaturate fractions (Fig. 5) also indicated the presence of weathered petroleum. All samples known to be impacted by AMOCO CADIZ oil exhibited a characteristic unresolved complex mixture (UCM) in both saturate ahd aromatic/unsaturate frac- tions indicative of weathered petroleum (Farrington and Meyers, 1975). Qualitative and quantitative differences existed between oiled and unoiled control sediments in the 0-5 cm depth interval. Hydrocarbon content was always higher in oiled sediments (Table 2). The control estuary sediment exhibited a small UCM and hydrocarbons indicative of biogenic origin in the saturate (odd chain alkanes n-C g3 to n-C 11 ) and aromatic/ saturate (polyolefinic material) fractions. T e contro marsh sediment exhibited a mixture of hydrocarbons of biogenic (odd-chain al- kanes n-C 25 to n-C 31 ) and petroleum (UCM) origin, with low concentra- tions of residual aromatic/ unsaturate hydrocarbons. The control beach sediment exhibited a n-alkane series (n-C 15 to n-C ) and UCM in the saturate fraction, and polynuclear aromatic componend originating from combustion of fossil fuel (eg. , nonalkylated 3-5 ringed polynuclear aromatics) (Youngblood and Blumer, 1975). The 'types and amounts of hydrocarbons were consistent with the known degree of impact from the AMOCO CADIZ oil spill. It is clear that in control beach and marsh sediments impact by hydrocarbons of petroleum or other anthropogenic sources had occurred. Evidence for degraded Amoco Cadiz oil at various sediment depths is summarized in Table 2. At the beach station AMC-4, AMOCO CADIZ oil was evident in the hydrocarbon assemblage down to the 10-15 cm interval in a sample collected in December 1978, and to the 15-20 cm interval in a sample collected in March 1979. At Aber Wrac'h there was evidence of AMOCO CADIZ oil to the 10-15 cm interval at both collection dates. The., amount of oil decreas 'ed with depth as evidenced by the total hydrocar- bon concentration and the increasing contribution of native sediment hydrocarbons (e.g. , plant derived saturate and aromatic/ saturate com- pounds) which dominated in the deepest layers as in the entire Aber Ildut core (see Figs. 4, 5). Similar results were found at the Ile Grande oiled site where Amoco CADIZ oil was detected in the 5-10 cm layer on both sampling dates and biogenic compounds dominated deeper layers. 168 TABLE 2. Preliminary Results of Total Hydrocarbon Levels in Brittany Sediments (AC=AMOCO CADIZ Oil Indicated by GC-MS Data) Sediment Type Depth Total Hydrocarbons (pg/g) Interval Oiled Control (cm) 12/78 3/79 3/79 Beaches AMC-4 Trez-Hir 0-5 295 AC 217 AC 110 5-10 158 AC 181 AC 46 10-15 244 AC 162 AC 130 15-20 72 128 AC 20-22 123 Abers Aber Wrac'h Aber Ildut 0-5 977 AC 1095 AC 690 5-10 590 AC 630 AC 530 10-15 47 AC 307 AC 305 15-20 80 118 204 20-25 33 103 115 25-30 25 346 30-35 45 Salt Marshes Ile Grande Ile Grande 0-5 1137 AC 863 AC 465 5-10 144 AC 439 AC 365 10-15 28 134 217 15-20 32 220 154 20-25 74 54 Evidence for Weathering of Sediment Hydrocarbons It was evident that oil was present at depths where extremely re- ducing conditions indicated the lack of oxygen (Aber Wrac'h and the Ile Grande oiled site), as well as in surface sediments which were more likely exposed to oxygen. This provided an opportunity to compare weathering patterns in sediments with markedly different exposure to oxygen. Since the actual amount of oil could vary betweensites, evi- dence of weathering was sought by comparing the relative amounts of hydrocarbons extracted from single samples known to be polluted with AMOCO CADIZ oil. Because of the rapid and extensive biodegradation which apparently followed the AMOCO CADIZ spill (Atlas, e@ al, 1981; .Ward et al, 1980) the comparison of n-alkanes to the more recalcitrant isoprenoid alkanes of similar volatility was not possible as an index of biodegradation. By the first sampling date (December 1978), n-C17/ 169 (A) Oiled Estuary Muctflat MIS 1, (B) Control Estuary Mudilat EQ ucm A I iC) Oiled Salt Marsh Mudflat (D) Control Salt fOarih Mudflat UCI., ucki (El Oiled Beach (Fi Control Beach Z I I I P Jcm " I k. )- , " I FIGURE 4. Gas chromatograms of saturate fraction of hydrocarbons ex- tracted from Brittany sediments collected in March 1979 (0-5 cm depth interval). (A) Oiled Estuary Mudflat (B@ Control Estuary Mudflat Pt,-n.h,n,, & D A. (Cl Oiled Salt Marsh Mudfldt (D) Control Salt Marsh Mudflat A (E) Oiled Beach (F) Control Beach A & cm 1A, F0,1,0 Estuary I LA 1@ FIGURE 5. Gas chromatograms of aromatic/unsaturate fraction of hydrocar- bons extracted from Brittany sediments collected in March 1979 (0-5 cm depth interval). 170 pristane and n-C18/phytane which were 3.3 and 2.8 in the reference mousse, were 0.07-0.58 and 0.024-0.36, respectively. Becas 'e of the relative persistence of aromatic hydrocarbons (Atlas, et al, 1991; Ward et al, 1980), alkylated naphthalenes, phenanthrenes and dibenzothio- phenes remained in Brittany sediments at levels sufficient to study weathering patterns in a long term 'time/depth series. The loss of these compounds relative to the more persistant C 3-alkylated deriva- tives of dibenzothiophene (C 3-DBT) is thought to be an index of weath- ering (Grahl-Nielsen, et al, 1978). There is no published evidence that changes in aromatic compounds relative to C 3 -DBT are due to bio- degradation, but these compounds are subject to microbial attack (Atlas, 1981). The ratios of aromatic compounds to C 3-DBT showed different changes with time depending on type of sediment and sediment depth (Table 3). In the AMC-4 0-5 cm interval there was a systematic de- crease in nearly all ratios from December 1978 to March 1979 to No- vember 1979. In muddy sediments decreases were less rapid and exten- sive. By March 1979, C - and C 4- naphthalenes C - and C4 - phenan- threnes and C 1-DBT showd relative decreases in the Aber Wrac'h 0-5 cm interval. An Aber Wrac'h core collected in November 1979 was sub- divided at one centimeter intervals at the depths of greatest change in Eh. In the 1-2, 2-3, and 3-4 cm intervals naphthalenes were'almost entirely depleted. Relative decreases in all except C 3- and C 4- phen- anthrenes were noted compared to earlier samples, with nogreat differ- ences among these depth intervals. Phenanthrenes were enriched rela- tive to C 3-DBT by May 1980 in the 0-3 cm interval. Ratios of aromatic hydrocarbons to C 3-DBT were generally greater in deeper layers (5-10, 10-15 cm) than in surface layers on any given date. Nevertheless, the data suggest that decreases in naphthalenes and DBT's occurred in the 5-10 cm interval in the one year between December 1978 and November 1979, despite the relative constancy of C 3-DBT over the time period. Results for Ile Grande were similar to those for Aber Wrac'h but the relative amounts of most components were greater. Between December 1978 and March 1979, there was a relative enrichment of all components except C - naphthalene. In March 1979, ratios were higher at 5-10 cm than at 40 -5 cm. Finer resolution around the depths of greatest Eh change was attempted on samples collected in May 1980. There was evi- dence of decreases in naphthalenes compared to earlier samples at all depths, even though the concentration of C 3-DBT was very high in all samples. Relative decreases in C -phenanthrenes and DBT's were noted in the 0-2 mm and 2-10 mm intervals. Except for C 4 -phenanthrenes, ra- tios were always higher in the 3-4 cm interval. Hydrocarbon Biodegradation In initial experiments, various 14 C-labelled aliphatic or aromatic hydrocarbons were incubated with surface sediments under aerobic condi- tions, or with subsurface sediments under conditions designed to pre- vent the exposure of obligately anaerobic bacteria to oxygen. e suc- cess of anaerobic methods was evidenced by the reduction of 1P so to 4 171 TABLE 3. Aromatic Hydrocarbons in Brittany Sediments Oiled by the Spill STATION RATIOS TO C 3-DIBENZOTHIOPHENES ng/g DATE DEPTH NAPTHTHALENES PHENANTHRENES DBT'S C3-DBT C2 C3 C4 C I C 2 C 3 C 4 C I C2 AMC-4 12/78 0-5cm .17 .52 .67 .28 .29 .48 .36 .35 1.08 785 3/79 0-5cm .02 .21 .38 .04 .23 .30 .17 .32 .97 735 11/79 0-5cm .01 .01 .01 .10 .20 .28 .08 .01 .38 400 ABER WRAC'H - 12/78 0-5cm .03 .19 .29 .05 .16 .35 .18 .16 .84 3,976 5-10cm .02 .51 .54 .16 .34 .39 .24 .35 1.14 '811 10-15cm .05 .17 .20 .28 .42 .39 .36 .19 .90 275 3/79 0-5cm .03 .18 .18 .13 .18 .32 .08 .01 .92 1,598 5-10cm .06 .22 .51 .22 .33 .42 .12 .16 1.64 831 .04 .18 .32 .21 .34 .31 .24 .19 1.01 492. 11/79 1-2cm 0 0 0 .10 .08 .33 .28 .01 .63 2,400 2-3cm 0 0 .01 .05 .10 .32 .23 .01 .59 2,200 3-4cm 0 0 0 .02 .05 .05 .26 0 .44 1,800 5-10cm 0 .02 .09 .14 .23 .59 .53 .06 0.60 860 5/80 0-3cm 0 0 .04 .13 .20 .76 1.00 .04 .42 1,700 ILE GRANDE OILED 12/78 0-5cm .04 .22 .32 .13 .22 .28 .22 .16 .91 2,809 3/79 0-5cm .10 .36 .37 .24 .38 .32 .22 .54 1.37 2,666 5-10cm .34 .88 .76 .46 .42 .46 .27 .63 1.44 745 5/80 0-2mm 0 .01 0 .05 .24 .48 .56, .09 .63 16,000 2-10mm 0 0 .05 .06 .19 .42 .45 .07 .62 12,000 3-4cm .007 .07 .18 .09 .28 .52 .35 .15 .82 82,000 172 H235S) and in some cases the generation of 14 CH4within two hours of Acubat '014 (Winfrey and Ward, submitted) in long term incubations no CO or CH was dete(iV in fumalin-killed controls. In most cases 2 4 significant amounts of CO 2 + CH4 were detected in vials incubated aerobically (Table 4). Anaerobic incubation severely reduced the amount of radiolabelled gases formed. However, small amounts of these metabolites were formed from n-hexadecane, n-heptadecane, heptadecene, ring or methyl labelled toluene, and benzene after lengthy anaerobic incubation. The amount of gaseous metabolites formed did not exceed 5% of the added radiolabel and reproducibility was poor. Repeated ef- forts by two individuals experienced in cultivation of methanogenic bacteria led to the same observations. Additions of FeCl or KNO (I ml of 0.5% (w/v) solutions in anoxic ASW replacedl@@e I mP afeitio'n of anoxic ASW) did not stimulate the formation of CO and CH from .2 4 n-hexadecane or heptadecene in Aber Wrac'h 5-10- cm sediment The possibility of initial accidental exposure to oxjjrn during tubing of samples was investigated by late addition of C-toluene which was soluble in water and could be added as an anoxic solution well after any oxygen initially present should have been consumed dur- ing dark incubation. Revsbech, et al, (1980b) have shown that oxygen consumption in intertidal sediments occurs in a matter of minutes fol- lowin f4 darkening to eliminate [email protected] As shown in Figure 6, ring- C-toluene was readily metabolized to CO when added either at the time of anaerobic tubing or 38 hours after 3W. anaerobic incuba- tion began. Similar results were found for [methyl- Cl-toluene. It was conceivable that the radiolabelled gases might have been produced from contaminants rather than from the hydacarbons them- selves. When an attempt was made to,4xecover the added C '@4 long-term radiolabelling experiments with (1- Cj-heptadec ff e and 14- C]-hepta- decene, it was noted that the total amount of CO2 + CH 4 produced during anaerRic incubations was similar to the level of impurities measured in C-labelled hydrocarbons recovered from formalin controls or from unpurified stocks of added radioisotopes (Table 5). Stock solutions were chromatographically separated into fl, f 2 Itnd f3 com- ponents which were then tested separately as sources of C-gases in dark anaerobic incubations with anaerobic sediments. The results of such experim'V. s are presented in Wure 7. The repurified f frac- tions of [I- C) hexadecane and 11- 14 Cj-heptadecane were clearly Sig- nificant sources for production of CO during dark anaerobic incuba- i4ons with a slurry of Ile Grande oile 3-6 cm sediment. Increases in CO2 with time following a lag of 5-15 days also suggested that oxi- dation did not result from any oxygen which might have been introduced accidentally _q@ring tubing. Similar results were, observed with repuri- fied f1of [I Cl-heptadecene. A final control was run to test the possibility that slow diffu- sion of oxygen through the vessels containing incubating samples could account for the obswed metabolism. Dark 14@ naerobic incubations of repurified f of [1- Cj-hexadecane and [1- Cj-heptadecane were car- ried out wit@ a slurry of mud from the 3-6 cm interval of Aber Wrac'h sediment. The individual vials were incubated inside an anaerobic 173 TABLE 4. 14 002 + 14 CH4 produced on 223 Day Incubation of Oiled Sediments 14 with C-Hydrocarbons (12/78). % OF ADDED 1 4ca ABER WRAC'H ILE GRANDE AMC-4 HYDROCARBON O-5cM 5-10cm 10-15cm 0-2cm 2-7cm 7-12cm 0-5cm 9-13cm 13-18cm AEROBIC ANOXIC ANOXIC AEROBIC ANOXIC ANOXIC AEROBIC ANOXIC ANOXIC 1- 14 -Hexadecane 33 3 ?b 45 ? ? ? ? 1- 14 C-Heptadecane 25 1 0 34 0 ? 20 0.5 0.6 1- 14 C-Heptadecene 28 0 ? 34 0 0 72 0.4 1.4 Ring- 14 C-Toluene 27 2 1.4 34 2.3 2.9 28 0 ? Methyl- 14 C-Toluene 18 ? ? 29 5 1.4 23 ? ? 1-(4,5,8)- 14 C- Naphthalene 72 0 0 78 0 0 U- 14 C-Benzene 26 0 2 45 0 0 25 0 0 7,10,- 14 C- Benzo(a)pyrene@ 3 0 0 0 0 0 0 0 a Initial levels ranged from 0.22-2.2 x 100 DPM/vial b Indicates that 14 CO2was apparently present but was not quantifiable because levels were near detection limits 40000- 30000- E Toluene W 0 20000- CL Toluene (late addition) 10000 10 20 30 40 Days FIGURE 6. 14 C02 production from [ring- 14 C]-toluene during dark anaerobic Toluene T luee 0 (late add, incubation with sediments from the Aber Wrac'h 5-10 cm interval. Radiolabel was added at the beginning of the incubation period or 38 hours after incubation began (late addition). Bars indi- cate one standard deviation. 174 TABLE 5. Recovery of 14 C following long term incubation with sediment collected at Aber Wrac'h, March 1979 COMPOUND/DEPTH % RECOVERED IN Co2 CH 4 CO2 +CH4 F I F 2 F 3 14 I- C-Heptadecane Mean of Controls 98.6 0.5 0.8 0-5cm Aerobic 6.0 3.0 (9-0) 89.4 0.5 1.2 5-10cm Anoxic 3.1 0.3 (3.4) 95.7 o.4 0.5 10-15cm Anoxic 3.9 (3-9) 94.9 0.5 0.6 14 I- C-Heptadecene Mean of Controls 91.5 2.9 5.6 0-5cm Aerobic 3.5 6.2 (9.7) 84.0 2.9 3.5 5-10cm Anoxic 3.9 1.2 (5-1) 89.0 2.2 3.6 Unpurified Radioisotope 87.8 5.6 6.6 Repurified Radioisotope 99.4 0.06 0.5 chamber sealed under a H 2 + CO 2 atmosphere. As shown in Table 6 these incubatiOP4 conditions did not prevent the slow, albeit variable, gener- ation of CO 2' Effects of Oiling on Anaerobic Process Evidence for effects due to AMOCO CADIZ oiling was first sought by comparing anaerobic processes at sites oiled or not oiled by this spill. Sulfate reduction dominated methane production at all sites (the ratio of sulfate reduction rate to the sum of sulfate reduction and methane production rate ranged from 0 '.951-0.998, even though me- thane production may have been overestimated). Thus, major changes in the type of terminal process were not evident during the period of our experimentation. The potential use of acetate by either of these two 175 IGO 3-6 cm slurry IGO 3-6cm slurry 1_14C Heptodecone j_ 14C -Hexadecone 3 - 6 CL E a 2 4 cli 0 L) 1 E CL 2 f3 and f Formalin f2, f3 and Controls __ZFormolln Controls 0 15 30 45 0 15 30 45 DAYS 14 Figure 7. CO2 production during dark anaerobic incubations of repuri- fied fit f 2 and f 3 fractions of radiolabelled hydrocarbons with a slurry made from the 3-6 cm interval of Ile Grande oiled sediment. Bars indicate one standard deviation. TABLE 6. Recovery of 14 C following dark incubation with a 3-6 cm Aber Wrac'h sediment slurry in double anaerobic incubator ISD) COMPOUND % RECOVERED IN CO 2 F I F 2 F3 1- 14 C-Hexadecane 3.94�3.48 96.1�3.48 0 0 1- 14 C-Heptadecane 16.4�23.6 83.4�23.6 0 0.2 176 groups makes [2- 14 Cl-acetate metabolism a useful means of differentiat- ing the 1@aportance of bacterial processes which accompli h its oxida- tion to co 2 (sulfate reduc Wn) or its conversion to CH14' (me Wne production, see Fig. 1) . [Z- Cl-acetate was MetaboliZed only to C02 at all sites except in t % surface layer of the oiled Ile Grande site where small amounts of CH4 were detected on one occasion (March, 1979). This observation, however, was not repeated at later sampling dates. Rates of sulfate reduction were highest in the surface layer and decreased with depth in all sites (Fig. 8). Rates in the 0-3 cm intervals were higher in oiled compared to control Aber mudflat sedi- ments (p = .004). In the beach and salt marsh mudf lat gediments, rates were higher in control sites than in oiled sites (p < .001). It is not possible, however, to attribute differences to the presence of AMOCO CADIZ oil since other differences between sites (e.g., amount of or- ganic loading) could also explain differences in sulfate reduction. ABER ILDUT TREZ-HIR ILE GRANDE 0. 0- 0. _J 0 OfX Z L I o lo- to- 0 w 20 201 - - 1.0 2.0 3.0 1.0 2.0 3.0 1.0 2.0 3.0 ABER WRAC'H AIVIC4 ILE GRANDE OILED 0 0. 0. W :r 10 to- to 0 a. W 20 20 20[ 1.0 2*0 3.0 1!0 a 3,0 1.0 2@O .3.0 SULFATE REDUCTION RATE (Pmoles S04@ .M1 -day-') Figure 8. Depth profiles for rates of sulfate reduction in oiled and control sediments. Bars indicate one standard deviation. To more directly observe the effect of oiling on microbial activi- ties in sediments, AMOCO CADIZ mousse was added to Ile Grande sedi- ments, and activities compared between mousse-treated and untreated sediments. Gas chromatograms of the saturate and aromatic fractions of this mousse are shown in Figures 9 and 10, respectively. These tracings can be compared to fresh light Arabian Crude in Figures 9 and 10. In the saturate fraction of the mousse, no hydrocarbons below C-12 were -_ze- - @-7 detected. In the aromatic fraction all predominant compounds were gone and only a UCM remained. Thus, extensive weathering of the low mole- cular weight compounds in the mousse had occured before collection. Extensive biodegradation of aliphatic compounds was not suggested as evidenced by the dominance of normal compared to isoprenoid alkanes. 177 FRESH OIL CIO C12 C C 4 C16 C18 C20 8 h WEATHERING 11AA LL@ 48 h WEATHERING 44 d AMOCO -CADIZ MOUSSE Figure 9. Gas chromatograms of saturate fraction of light Arabian crude oils and Amoco Cadiz mousse. Table 7 shows the ef f ect of the addition of 5% and 25% mousse (v/v) on sulfate reduction in sediments from the oiled and unoiled site at LL L dLLJ@', Ile Grande. A slight inhibition of sulfate reduction rate was observed at the control site, while a slight stimulation was observed at the oiled site. These rates, however, were not significantly different from the unoiled controls (p 2@ 0.08). No significant differences in methane production between control sediments and sediments treated with 178 FRESH OIL 8 h WEATHERING 48 h WEATHERING 44 d AMOCO-CADIZ MOUSSE Figure 10. Gas chromotograms of aromatic fraction of light Arabian crude oils and Amoco Cadiz mousse. mousse were observed at either site- U ata not shown). Table q4 shows the IV fects of mousse additions on [2 C]-acetate oxidation to co2 No CH4 was detected in any of the samples. At the ui@iled control site, the addition of 5% mousse decreased the amount of CO produced in 2h by 70% (p = 0.02), while an 86% inhibition was observei with the additions of 25% mousse (p = 0.01). A@4 the oiled site at Ile Grj4de, mousse additions appeared to inhibit COI production from [2- Cj- acetate although these differences were no significantly different from the control (p @_ 0.25). The effect of fresh and slightly weathered light Arabian crude oil on microbial activities in sediments was also examined because of the probable chemical differences between the highly weathered mousse and the oil which impacted the sites studied. The effect of 0.05% benzene and toluene was also examined. They are highly volatile aromatic com- pounds with known inhibitory effects (Robertson et al, 1973). Figure 9 shows chromatograms of the saturate fraction of the fresh and weathered light Arabian crude. The fresh oil contained large amounts of low 179 TABLE 7. Effect of AMOCO CADIZ Mousse on Sulfate Reduction in Marsh Se iments a T Grande Control Pe Grande Oiled Addition Rate % of Control p Rate % of Control p Control 4.99 100% - 1.64 100% 5% Mousse 3.16 63 .08 2.36 144 .15 25% Mousse 3.53 71 14 1.67 102 .95 aSediment samples were collected in November, 1979. bRates are the mean of 3 replicates and expressed as pmoles SO S/ml/d. 4 TABLE 8. Effect of Mousse on [2- 14 CI-Acetate Metabolism to 14 CO in Marsh Sediments 2 a 14 Ils Grande Control 14 T Grande Oiled Addition co2 % of Control p CO2 % of Control P Control 513 100% - 184 100% 5% Mousse 156 30% .02 115 63% .41 25% Mousse 72 14% .01 78 42% .25 aSediment samples were collected in November, 1979. bMean of 3 replicates expressed as DPM x 10- 3. molecular weight compounds which decreased relative to less volatil;p n-alkanes (e.g. , C-24) in the 8h and 48h weathered oil. Octane and other compounds of similar volatility were nearly depleted after 48h of evaporative weathering. Figure 10 shows the aromatic fractions of each of the oils used. After 8h of evaporative weathering, toluene and sev- eral other volatile aromatic compounds were significantly reduced, and after 48h most of the predominant volatile aromatics were evaporated. 180 Table 9 reports the effect of these oil and aromatic additions on sulfate reduction rates in Ile Grande surface sediments. At the unoiled control site, all oil additions decreased the rate of sulfate reduction although not significantly below the control (p @_! 0.16). The greatest inhibition was observed for toluene and benzene additions (p = 0.10 and 0.09 respectively). At the oiled site, rates under all conditions were not significantly different from the control rate. The effect of the oil, benzene, and toluene additions on methanogenesis (results not shown) were variable. None of the additions resulted in. a rate of methanogenesis that was significantly different than the rate without additions. TABLE 9. Effect of Hydrocarbons on Sulfate Reduction in Marsh Sedi- ments a Ip Grande Control Ije Grande Oiled Addition Rate % of Control p Rate % of Control p Control 2.27 100% - 1.01 100% - 10.% Fresh Oil 1.67 73 0.95 0.85 84 0.08 10% 8 h Weathered Oil 1.57 71 0.16 0.87 86 0.11 10% 48 h Weathered 0ii 1.82 82 0.97 1.07 106 0.48 .05% Toluene 1.42 64 0.10 1.11 110 0.23 .05% Benzene 1.32 59 0.09 1.13 112 0.15 aSediment samples were collected in April, 1980. bRates are the mean of 3 replicates and expressed as pmoles SO S/ml/d. 4 The effe ci@ of the oil and aromatic additions on [2- 14 Cl-acetate oxidation to CO in Ile Grande is ikown in Table 10. At both the oiled and unoilei site, the amount of CO 2produced in 2 h was signi- ficantly reduced (p !_S 0.01) with all of the additions. In general, the magnitude of inhibition was greater at the control site (76-97%) than at the oiled site (51-93%). In both sites inhibition was greatest with the unweathered oil and decreased with the extent of weathering of the oils. 181 TABLE 10. V@ect of Hydrocarbons on. [2- 14 CI-Acetate Metabolism to a CO2 in Sediments 14 116 Grande Control 14 Ije Grande Oiled Addition CO2 % of Control p CO2 % of Control p Control 358 100% 244 100% Fresh Oil 10 3 0.00 17 7 <0.001 8 h Weathered Oil 44 12 0.01 66 27 <0.001 48 h Weathered Oil 87 24 0.01 119 49 <0.001 Toluene 29 8 0.002 86 35 <0.001 Benzene 31 9 0.002 65 27 <0.001 aSediment samples collected in April, 1980. bMean of 3 replicates expressed as DPM x 10- 3. DISCUSSION The major objectives of this study were to address the fate and effects of hydrocarbons from the AMOCO CADIZ spill in anaerobic sedi- ments. It is first necessary to consider the chemistry of the various intertidal sediments with respect to exposure to oxygen. A great deal has been learned recently due to the application of microelectrodes to the study of oxygen distribution and dynamics in marine subtidal and submerged intertidal sediments (Revsbech, et al, 1980a, b). A variety of sediments (including medium-grained sands) exhibited very narrow oxygenated intervals ranging from 2-10 mm, below the sediment water in- terface. Oxygen depletion has been measured to occur above the verti- cal discontinuity of Eh in coastal sediments (Revsbech, et al, 1980a). Thus the Eh profile of a sediment may serve as a conservative estimate for the aerobic/anoxic boundary. Eh profiles then indicate that anoxic conditions were likely below 1 cm in the Ile Grande oiled site and be- low 2-3 cm in Aber Wrac'h sediment. The real depth of oxygen penetra- tion at any given time is likely to be less. Changes in color from brown to black at about 2 mm in the Ile Grande oiled sediment may in- dicate an extremely narrow aerobic zone. Similar color changes in Aber Wrac'h at about 2-3 cm may indicate that net oxygen penetration in this sediment is deeper, but oxidants other than oxygen could keep Eh high- er. The Eh profiles were measured on undisturbed sediments and pro- 182 bably reflect the chemistry of sediments under relatively calm condi- tions. Mixing which occurs as a result of tidal or storm driven wave action might alter the depth to which oxygen can penetrate sediments. This should vary with the nature of the sediment so that muds should be less affected than sandy sediments on high-energy beaches. Over sea- sonal time intervals, it is likely that the aerobic/anoxic boundaries predicted by Eh profiles are preserved in the muddy sediments sampled. However, it is likely at the oiled beach AMC-4 that erosion and deposi- tion created considerable instability in the depth of oxygen penetra- tion and could even have caused vertical redistribution of sediments (Gundlach and Hayes, 1978). It is also possible that oxygen could be -introduced to depths below the lower boundary of its diffusion by sedi- ment infauna which can burrow into anaerobic sediments. The vertical distribution of the dominant anaerobic process, sul- fate reduction, indicated maximum activity in the surface 0-3 cm inter- val at all stations. The obligately anaerobic sulf ate- reducing and methane -p roduc ing bacteria were also present in maximum number in the 0-3 cm depth interval (Winfrey and Ward, submitted). These observa- tions suggest that at least portions of the 0-3 cm interval at all sites were sufficiently anoxic to allow survival and activity of obli- gately anaerobic microorganisms. A survey of the various sediments sampled confirmed the presence of AMOCO CADIZ pollutants in oiled sites. Although control sites were not polluted by the AMOCO CADIZ spill, each contained some hydrocarbons of anthropogenic origin. The extent of oiling was greatest at the sedi- ment surface where anaerobic processes were greatest. Oiling decreased with depth, but there was clear evidence of AMOCO CADIZ hydrocarbons in sediments likely to be free from exposure to oxygen. Sediments below the aerobic/anoxic boundary and above the deepest level of penetration of AMOCO CADIZ pollutants provided an environment suitable for the en- richment of anaerobic hydrocarbon- degrading microorganisms, and appro- priate for comparison to aerobic surface sediments to study the differ- ences in weathering in situ due to different exposures to oxygen. Because of the rapid biodegradation of aliphatic components and relative enrichment of aromatic components of the spilled oil (Atlas, et al, 1981; Ward, et al, 1980), ratios of naphthalenes, phenanthrenes and dibenzothiophenes to the more persistent C 3-DBT were used as an index of weathering. This index should be independent of absolute amounts of oil within sediment samples which could vary due to patchy distribution of oil. Changes in sediment aromatic hydrocarbons oc- curred in all sediments and at all sediment depths where comparisons were made for one year or longer. The greatest and most rapid changes were noted in surface sediments of the beach station AMC-4 where most compounds had decreased by one year after the spill and extensive losses had occurred by about 20 months after the spill. This seems consistent with the expected mixing and oxygenation of this high energy beach sediment. In muddysediments, slower changes in relative amounts of aromatic compounds were noted. Extensive losses were observed main- ly among the naphthalenes and DBT's. This may have been related to the low energy nature of these sediments and/or the corresponding lack of oxygenation indicated by reducing conditions. Decreases in these com- 183 pounds were, however, noted at depths below the minimum Eh 20-26 months after the oil spill. It is difficult to rule out exposure to oxygen in these subsurface muddy-sediment environments. The degree of storm driven mixing and/or irrigation by sediment fauna are unknown and might be significant over a 12-18 month time course. More than two years after theAMOCO CADIZ spill many well resolved aromatic compounds per- sisted (e.g., phenanthrenes); however, the slow disappearance of some aromatic hydrocarbons (e.g., naphthalenes) from subsurface muds may indicate that these resolved compounds will not persist indefinitely. The potential biodegradation of hydrocarbons measured using 14 C_ labelled hydrocarbons was much lower under,4naerobic than under aerobic conj@tions. For example, 33% of added [1- Cj-hexadecane was converted to C-gases aerobically, whereas only 3% was converted anaerobically after 233 days in Aber Wrac'h sediment (Table 4). Due to the expected high oxygen demand of the surface sediments, itis likely that 0 deple- tion occurred rapidly in vials incubated aerol%@cally. Thus, ii is t 144 140 surprising that complete conversion of added C-hydrocarbons to C_ gases did not occur in long term aerobic incubations. Since the added radiolabelled hexadecane (4.2. pg/sample) exceeded the indigenous hexa- decane measured in these sediments (16-174 ng/sample), potential rates of aerobic and anaerobic metabolism can be calculated by multiplying the percentage conversion by the amount of hexadecane added and divid- ing by the number 24 days incubation and dry weight of the sample. The maximum level of C-gases produced under aerobic conditions was de- tected at the earliest analysis time (66 days). This leads to a cal- culated rate of 13.8 ngm/gm-dry weight/day. The true potential rate is probably greater due to incomplete exposure of the entire sample to oxygen and to oxygen depletion. Calculations using rates of oxygen consumption for European coastal sediments suggest that oxygen should have been completely consumed within the first ten days of incubation. The corresponding rate would be 91 ngm/gm-dry weight/day, approximately e-fifth of the potential rate reported by Atlas and Bronner (1981). C-gases increased with time during anaerobic incubation. The corre- sponding potential anaerobic rate of hexadecane metabolism after 233 days incubation of 0.3 ngm/gm-dry weight/day is 46 times slower than the measured potential aerobic rate and over 300 times slower than the aerobic rate calculated from reasonable assumptions about the condi- tions under which aerobic controls were run. These results are similar to other reports which demonstrate the severe limitations on hydrocar- bon metabolism imposed by reduced amounts or, la@k of oxygen (Ward and Brock, 1978; Hambrick, et al, 1980; DeLaune, et al, 1981). It was in- teresting that no evidence was obtained for anaerobic naphthalene oxi- dation. Obvious problems of naphthalene volatility may have decreased the amount actually added to vials thereby lowering the sensitivity for detecting its oxidation. As in earlier studies (Ward and Brock, 1978), it was not possible to eliminate metabolism of hydrocarbons under stringent anaerobic con- ditions. It was important to investigate the possibility that slow anaerobic oxidation might occur in order to predict whether or not hy- drocarbons buried in permanently anoxic sediments persist indefinitely. Controls were run against the possibility of initial accidental inclu- sion of oxygen, photosynthetic oxygen production, oxygen leakage into 184 experimental vials during incubation, and against the possibility that 14 mpounds contaminating radiolabelled hydrocarbons were the sources of C-gases. The results of these experiments were consistent with the conclusion that slow anaerobic oxidation of some petroleum hydrocarbons may be occurring in anoxic sediments polluted with AMOCO CADIZ oil. A second major objective of this work was to study the effects of AMOCO CADIZ oil on the dominant anaerobic processes within sediments. As mentioned above, oiling was heaviest in the surface sediments where anaerobic processes occurred at maximum rates. We were unable to moni- tor any immediate effects of the AMOCO CADIZ spill because our first sampling trip was in December 1978, 9 months after the spill. We were also limited by the lack of any data on the microbial activities in our sampling sites prior to the oil spill. However, by examining unoiled sites, we were able to see if any major alterations in chemistry and activities had occurred in the oiled sediments. Comparisons between oiled and unoiled sites revealed lower rates of sulfate reduction in oiled beach and marsh sediments. It is not possible, however, to attribute these differences to the presence of AMOCO CADIZ oil because 1) inhibition was not observed at all sites where oiling occurred, 2) the magnitude of differences observed was small (largest difference was 51% of the control), and 3) other dif- ferences between sites which could influence sulfate reduction rate (e.g. , organic loading) were unknown. It is also difficult to inter- pret whether rates measured in control sites were typical of unpolluted sediments, as hydrocarbon analyses revealed a previous history of oil- ing in control sediments. No differences in methane production or ace- tate metabolism wer noted. Although prelimir@Vy results demonstrated 74 small amounts of CH production from [2- Cl-acetate at the Ile Grande oiled site (Winfrey and Ward, 1981), this observation was not confirmed in subsequent work. Sulfate reduction alwai@ dominated me- thane production and acetate was metabolized only to CO . The lack 'iormed 9-18 of profound differences in comparative experiments per months following the spill, suggest that no long-term effects on an- aerobic processes occurred in these sediments. I We examined the effect of an unweathered oil on anaerobic pro- cesses in order to determine whether any short-term effects of oiling might have occurred. A fresh light Arabian crude oil did not signi- ficantly effect sulfate reduction or methane production, but inhibited acetate oxidation 93-97% in oiled and control sediments from Ile Grande. The ability of the crude oil to inhibit acetate oxidation was reduced in oil samples which had been evaporated to increasing degrees. Highly volatile molecules such as toluene and benzene caused 91-92% inhibition in control sediments and 65-73% inhibition in oiled sedi- ments. The data suggest that volatile components of the oil may be re- sponsible for inhibitions to acetate oxidation. However, mousse col- lected near Ile Grande 44 days after the AMOCO CADIZ spill also inhi- bited acetate oxidation at the Ile Grande control site. This mousse sample had lost most of its more volatile components, but was appar- ently not extensively altered by biodegradation. Inhibitions were al- ways greater at the control site than at the oiled site. Thus, the. oiled area of the marsh appears to have become less sensitive to the 185 effects of additional heavy oiling. This may be a result of an adapta- tion of microorganisms to the presence of oil, or to a selection of oil resistant populations. It is not surprising that hydrocarbons did not directly inhibit sulfate reduction since this process appears to be active in oil forma- tion waters (Bailey, et al., 1973; ZoBell, 1958). However, it was sur- prising that compounds which inhibited acetate oxidation did not dir- ectly inhibit sulfate reduction, as other investigations have pointed to the importance of acetate as a substrate for sulfate reduction (Win- frey and Ward, submitted; Sorensen et al, 1981; Banat and Nedwell, per- sonal communication). The chemistry of hydrocarbons present in the various sediments one year after the spill indicated the presence of oil highly altered by evaporation and biodegradation. The levels observed in the environment were also lower (0.1-1 mg/g) than the levels added in our experiments to simulate heavy oiling (50-250 mg/g). It is possible that a tempor- ary inhibition of acetate oxidation could have resulted from very heavy oiling of relatively fresh oil. Such conditions could have existed at all polluted sites immediately following the AMOCO CALDIZ spill, al- though rapid loss of volatile compounds probably occurred between spillage and beaching of oil (Dowty, et al, 1981; Ward, et al, 1980). Any inhibitory effect would then have been reduced as cleanup or trans- port of hydrocarbons out of the sediments decreased hydrocarbon amount, and as evaporation, dissolution and biodegradation altered the remain- ing sediment hydrocarbons. By the time site comparison experiments could be performed, recovery from any negative effects which might have occurred had apparently taken place. The inhibitory effects on acetate oxidation we observed may be significant in extremely cold regions where slow rates of evaporation would occur. ACKNOWLEDGEMENTS We are indebted to the Centre Oc6anologique de Bretagne at Brest and the Station Biologique at Roscoff for providing laboratory space and assistance. We also thank Bob Clark of the NOAA National Analyti- cal Facility for supplying AMOCO CADIZ mousse, George Ward of Exxon Corp. for supplying the light Arabian crude oil, and Dale Meland and Melinda Tussler for technical assistance. This study was part of a joint effort undertaken by the Centre Na- tional pour 1'Exploitation des Ockans (CNEXO) of the French Ministry of Industry and the NOAA of the U.S. Department of Commerce to study the ecological consequences of the AMOCO CADIZ oil spill. It was financed by funds given to NOAA (contract NA 79RAC00013) by the Amoco Transport Company and by the NOAA Outer Continental Shelf Environmental Assess- ment Program, through an interagency agreement with the Bureau of Land Management. 186 REFERENCES American Public Health Association. 1976. Standard methods for the examination of water and wastewater, 14th ed. American Public Health Association, Inc. New York. Atlas, R. M. 1981. Microbial degradation of petroleum hydrocarbons: an environmental perspective. Microbiological Rev. 45:180-209. Atlas, R.M.- and A. Bronner. 1981. Microbial hydrocarbon degradation within intertidal zones impacted by the AMOCO CADIZ oil spillage. pp. 251-256 in AMOCO CADIZ: Fates and Effects of the Oil Spill. Proc. of the Int. Symp. Centre Nationale pour L'Exploitation des Ockans, Paris. Atlas, R. M. , P. D. Boehm and J. A. Calder. 1981. Chemical and biolo- gical weathering of oil, from the AMOCO CADIZ spillage, within the littoral zone. Est. Coastal Mar. Sci. 12:589-608. Bailey, N. J. L. , H. H. Krouse, C. R. Evans, and M. A. Rodgers. 1973. Alteration of crude oil by waters and bacteria - evidence from geochemical and isotope studies. Am. Assoc. Pet. Geol. Bull. , 57:1276-1290. Boehm, P. D. , D. L. Fiest, and A. Elskus. 1981. Comparative weather- ing patterns of hydrocarbons from the AMOCO CADIZ oil spill ob- served at a variety of coastal environments. pp. 159-173 In AMOCO CADIZ: Fates and Effects of the Oil Spill, Proc. of tii-e- Int. Symp. Centre National pour L'Exploitation des Oc6ans, Paris. Brown, D. W.) L. S. Ramos, M. Y. Vyeda, A. J. Friedman and W. D. Macloed. 1980. Ambient temperature extraction of hydrocarbons from marine sediments: comparison with boiling-solvent extrac- tions. In L. Petrakis and F. Weiss (eds), Petroleum in the Marine Environment, Advances in Chemistry series no. 185, American Chemi- cal Society, Washington, D.C. Bryant, M. P. 1976. The microbiology of anaerobic degradation and methanogenesis with special reference to sewage. pp. 107-118 In Microbial energy conversion, H. G. Schlegel, G. Gottschalk, and Pfennig (eds.) E. Goltze, K. G. Gottingen. Burkholder, D. 1963. Some nutritional relationships among microbes of the sea sediments and waters. pp. 1133-1150 in Symposium on ma- rine microbiology, C. H. Oppenheimer (ed.). Thomas Springfield, Ill. Calder, J. A., J. Lake, and J. Laseter. 1978. Chemical composition of .selected environmental and petroleum samples from the AMOCO CADIZ oil spill. pp. 21-84 in The AMOCO CADIZ Oil Spill, a preliminary scientific report. W. N. Hess (ed. ). NOAA/EPA Special Report, U. S. Govt. Printing Of f ice. Washington, D. C. 187 Centre Nationale pour I'Exploitation des Ockans, Institut Francais du Petrole, and Institut Geographique National. 1979. Remote sens- ing of hydrocarbon pollution, preliminary report. Centre Nation- ale pour l'Exploitation des Oc6ans, 55 Avenue d'i6na, 75016, Paris. DeLaune, R. D., W. H. Patrick, Jr. , and M. E. Casselman. 1981. Effect of sediment pH and redox conditions on degradation of benzo(a)- pyrene. Mar. Pollut. Bull. 12:251-253. Dowty, B. J. , J. W. Brown, F. N. Stone, J. Lake and J. L. Laseter. 1981. GC-MS analysis of volatile organics from atmospheres im- pacted by the AMOCO CADIZ oil spill. pp. 13-22 in AMOCO CADIZ: Fate and Effects of the Oil Spill. Proc. Int. Symp. on the AMOCO CADIZ Centre Nationale pour L'Exploitation des Oc6ans, Paris. Farrington, J. W. and P. A. Meyers. 1975. Hydrocarbons in the marine environment. In Environmental chemistry, Vol. 1, G. Eglinton (ed. ) The ChemiEa-1 Society Special Report No. 35. London. Fenchel, T. M. and B. B. J@rgensen. 1977. Detritus food chains of a- quatic ecosystems: The role of bacteria. Adv. Microbial Ecol. , 1:1-58. Grahl-Nielsen, 0. , J. T. Staveland and S. Wilhelmsen. 1978. Aromatic hydrocarbons in benthic organisms from coastal areas polluted by Iranian crude oil. Fish. Res. Bd., Can. 35:615-623. Griffiths, R. P. , T. M. McNamara, B. A. Caldwell, and R. Y. Morita. 1981. Field observations on the acute effect of crude oil on glu- cose and glutamate uptake in samples collected from arctic and subarctic waters. Appl. Environ. Microbiol. 41:1400-1406. Griffiths, R. P., B. A. Caldwell, W. A. Broich and R. Y. Morita. 1981. The long-term effects of crude oil on uptake and respiration of glucose and glutamate in arctic and sub-arctic marine sediments. Appl. Environ. Microbiol. (in press). Gundlach, E. , and M. Hayes. 1978. Investigations of beach processes. pp. 85-196 in The AMOCO CADIZ Oil Spill, a preliminary scientific report. W. N. Hess (ed. ) NOAA/EPA Special Report. U. S. Govt. Printing Office. Washington, D.C. Hambrick, G. A., III, R. D. DeLaune and W. H. Patrick, Jr., 1980. Ef - fect of estuarine sediment pH and oxidation-reduction potential on microbial hydrocarbon degradation. Appl. Environ. Microbiol. 40: 365-369. Hann, R. W. Jr., L. Rice, M. Trujillo, and H. N. Young. 1978. Oil spill clean up activities. pp. 229-275 in The AMOCO CADIZ Oil Spill, a preliminary scientific report. W. N. Hess (ed.) NOW EPA Special Report. U.S. Govt. Printing Office. Washington, D.C. 188 Hungate, R. E. 1969. A roll tube method for cultivation of strict an- aerobes, pp. 117-137 in Methods in Microbiology, Vol. 3B, J. R. Norris and D. W. Ribbons (eds.) Academic Press, New York. Ivanov, M. W. 1964. Microbiological processes in the formation of sulfur deposits. Israel Program for Scientific Translation, Ltd., Jerusalem. (translated from Russian for the U.S. Department of Agriculture and the National Science Foundation in 1968). Jorgensen, B. B. 1978. A comparison of methods for the quantification of bacterial sulfate reduction in coastal marine sediments. I. Measurement with radiotracer techniques. Geomicrobiol. J. 1: 11-28. Kalil, E. K. 1974. Rapid pore water analysis for sediments adjacent to reactor discharges. pp. 269-283 in Environmental surveillance around nuclear installations, vol. 1. International Atomic Energy Agency, Vienna. Knowles, R. and C. Wishart. 1977. Nitrogen fixation in arctic marine sediments: effect of oil and hydrocarbon fractions. Environ. Pollut.) 13:133-149. Lund, R. E. 1979. A user's guide to MSUSTAT - an interactive statis- tical analysis package, 1979. Montana State University, Bozeman. Mah, R. A., D. M. Ward, L. Baresi, and T. L. Glass. 1977. Biogenesis of methane. Ann. Rev. Microbiol. 31:309-341. Nelson, D. R. , and J. G. Zeikus. 1974. Rapid method for the radioiso- topic analysis of gaseous end products of anaerobic metabolism. Appl. Hicrobiol., 28:258-261. Revsbeck, N. P. , B. B. Jorgensen and T. H. Blackburn. 1980a. Oxygen in the sea bottom measured with a microelectrode. Science 207: 1355-1356. Revsbeck, N. P. , J. Sorensen, T. H. Blackburn and J. P. Lomholt. 1980b. Distribution of oxygen in marine sediments measured with microelectrodes. Limnol. Oceanogr. 25:403-411. Rheeburgh, W. S. and D. T. Heggie. 1977. Microbial methane consump- tion reactions and their effect on methane distributions in fresh- water and marine environments. Limnol. Oceanogr. 22:1-9. Robertson, B., S. Arhelger, P. J. Kinney, and D. K. Button. 1973. Hydrocarbon biodegradation in Alaskan waters. pp. 171-184 in The Microbial Degradation of Oil Pollutants, D. G. Ahearn and S. P. Meyers (eds. ) Publication No. LSU-SG-73-01. Baton Rouge, La. Sorensen, J., B. B. Jorgensen, and N. P. Revsbeck. 1979. A comparison of oxygen, nitrate, and sulfate respiration in coastal marine sed- iments. Microbial Ecology. 5:105-115. 189 Sorensen, J " D. Christensen, and B.B. Jorgensen. 1981. Volatile fatty acids and hydrogen as substrates for sulfate-reducing bac- teria in anaerobic marine sediment. Appl. Environ. Microbiol. 42:5-11. Stainton, M. P. 1973. A syringe gas-stripping procedure for gas chro- matographic determination of inorganic and organic carbon in fresh water and carbonates in sediments. J. Fish Res. Bd. Can. 30: 1441-1445. Tabatabai, M. A. 1974. Determination of sulfate in water samples. Sulphur Inst. J. , 10: 11-13. Walker, J. D. , P. A. Seesman, and R. R. Colwell. 1975. Effect of South Louisiana crude oil and no. 2 fuel oil on growth of hetero- trophic microorganisms including proteolytic, lipolytic, chitino- lytic and cellulolytic bacteria. Environ. Pollut. 9:13-33. Ward, D. M. and T. D. Brock. 1978. Anaerobic metabolism of hexadecane in sediments. Geomicrobiol. J. ) 1: 1-9. Ward, D. M. and G. J. Olson. 1980. Terminal processes in the anaero- bic degradation of an algal-bacterial mat in a high-sulfate hot- spring. Appl. Environ. Microbiol., 40:67-74. Ward, D. M. , R. M. Atlas, P. D. Boehm, and J. A. Calder. 1980. Micro- bial biodegradation and chemical evolution of oil from the Amoco spill. Ambio 9:219-225. Winfrey, M. R. , and D. M. Ward. 1981. Effect of the AMOCO CADIZ oil spill on predominant anaerobic microbial processes in intertidal sediments. p. 257-267. In AMOCO CADIZ: Fates and Effects of the Oil Spill, Proc. of the Int. Symp. Centre National pour L'Exploi- tatibn des Oc6ans, Paris. Winfrey, M. R. , D. G. Marty, A. J. M. Bianchi and D. M. Ward. 1981. Vertical distribution of sulfate reduction, methane production, and bacteria in marine sediments. Geomicrobiol. J. (in press). Youngblood, W. W. and M. Blumer. 1975. Polycyclic aromatic hydrocar- bons in the environment: homologous series in soils and recent marine sediments. Geochim. Cosmochim. Acta 39:1303-1314. ZoBell, C. E. 1958. Ecology of sulfate reducing bacteria. Producers Monthly, 22:12-29. 190 PART II Biological Studies After the AMOCO CADIZ Oil Spill Edited by M. Marchand Centre Oceanologique de Bretagne Brest, France REPONSES DES PEUPLEMENTS SUBTIDAUX A LA PERTURBATION CREEE PAR LIAMOCO CADIZ DANS LES ABERS BENOIT ET WRAC'H par Michel GLEMAREC et Eric HUSSENOT Laboratoire d'Oce'anographie biologique, Institut d'Etudes Marines, Faculte' des Sciences et Techniques, 29283 Brest Ce'dex ABSTRACT During three years after the Amoco-Cadilz oil-spill, the succes- sion in time of different ecological groups with regard to excess of organic matter, allow to define chronological process. First,total disappearance of sensible and tolerant species by toxicity. When pollution is stabilized, we describe appearance, development and regression of an opportunist fauna, finally the excessive develop- ment of tolerant species before return to a new equilibrium. This temporal succession is studied along two different gradient of de- creasing hydrodynamism, the abers, where the chemical decontamina- tion and the biological process are not synchronized. Near three years after the oil-spill most communities are still perturbated and unbalanced. Patterns of temporal evolution and succession is discussed. INTRODUCTION Les deux Abers, Benoit et Wrac1h, sont situe's a proximite' de 116chouage de l'Amoco-Cadiz (Fig. 1). Clest Vaire qui a 6t6 la plus affect6e par la mare'e noire. Depuis, de nombreuses e'tudes ont con- cerne ces abers et l1analyse des successions e'cologiques au sein des communaute's de l1endofaune s'est ave'r6e tre's int6ressante dans le cas d'6tudes A moyen et long terme. Sont apparues des fluctuations temporelles non saisonnie'res mais e'volutives, qu'il est possible de synth6tiser trois ans apr@s 116chouage. Apr@s la destruction quasi-comple'te des communaute's d1origine, la recolonisation passe par le de'veloppement transitoire d'une faune spe'ciale, caracte'risant un exc@s de matie're organique sur les fonds se'dimentaires. Notre approche consiste d'abord a reconnaltre les groupes taxonomiques, constitue's par les espkes de polluosensi- bilit6 6quivalente en face d'un exce's de mati@re organique. Leur apparitibn successive dans le temps, leur disparition, constituent les param@tres obligatoires de cette approche dynamique. Ce type d'analyse montre comment une perturbation biologique peut persister longtemps dans 1'e'cosyste'me, alors que les facteurs e'cologiques semblent normaux. Le long du m6me gradient e6cologique spatial, 116tude compar6e des deux Abers peut montrer des diff6rences dans les vitesses de retour a un nouvel e'quilibre. 191 ac N qSqg U, 4qR I 'IN q4 Sq" AbOf Wrach__,2qFqi_-'_----1 qS qSqf S al Ell qI4qN8qO I A.18ei FIGURE 1. Localisation de 11q6pave de 11,q@MOCO-CADIZ (A-C) par rapport I aux deux abers, ou' sont represente'es les diffq6rentes unite's biose'dimentaires. qMETHO2qDES La macrofaune qui vit dans les aires sq6dimentaires est e'tudiq6e le long des chenaux suqbtidaux de ces deux Aqbers. Cinq prq6lq@vements sont re'alise's a chaque station avec une qbenne "Aberdeen". Ces q6chan- tillons sont rq6pe'te's trois fois dans ql'anne'e (hiver, printemps, q6tq6). Les stations -plus de quinze dans chaque Aber- sont reprq6sentatives des diffq&rents peuplements. Leur distribution spatiale de 0qVaval vers ql1amont est la suivante (Gentil et Cabioch, 1979 ; Gle'marec et Hussenot, 1981) : sables grossiers (SG), sables dunaires fins et moyens (DU), sables fins et envase's (FV), sables hq6tq6rogq@nes envasq6s (SHV) et vases sableuses (VS). Cette distribution (Fig. 1) correspond a un hydrodynamisme de'- croissant.et, inversement, A une accumulation croissante d'hydrocar- bures dans les se'diments (Marchand et Caprais, 1981). Dans l'Aber Benoit, la re'partition est en partie diffq6rente parce que les sables dunaires sont trq@s dq6veloppq6s. Il y a une vaste accumulation en aval (DUl) sous la pointe de Corn ar Gazel et, au milieu du chenal, deux unitq6s quelque peu diffq6rentes, DU2 et DU3, la derniq@re est en con- tact avec les sables he'tq6rogq@nes envase's (SHV). Les sables fins ntexistent pas dans le chenal, mais sont localise's en aval dans une aire protegee entre des plateaux rocheux, collectan4qt les particules fines qui peuvent sortir dans 28qVaber. RESULTATS Tout d'abord -phase premi6q6re de Marchand (1981)q- la toxicit2q6 des hydrocarbures a induit de lourdes mortalit2q6s au sein de toutes les communauteq's (Chass4q6, 1978, Cabioch et a0qZ., 1979). Dq'autres es- p0q6ces, plus opportunistes, slinstallent dans une deuxieq'me phase (Gleq'marec et Hussenot, 1981 ; Le Moal et Quillien-Monot, 1981). 192 Les nouvelles populations caractrisent un exc@s de matie're organque, dans le cas d'effluent urqbain arrivant en mer par exemple. Aprq@s dif- fe'rents travaux sur ces problmes (Pearson et Rosenberg, 1978 ; Bellan et aZ., 1980 ; Gle'marec et Hily, 1981) il est possible de regrouper ces especes en fonction de leur polluosensiqbilitq6. 2q@qr2qjipqfq@_I : espe'ces sensihles, larqgement dominanteqs qen conditions nor- males. Elles diffq@rent selon chaque type de peuplement. Dans les sables dunaires fins et moyens par exemple, les Amphipodes sont nom- qbreux (Bathyporeia spp., AmpeZisca sqpp.) ; ils meurent rapidement dans tous les cas de mare'e noire (Chasse' et aqZ., 1967 ; Cabioch et aqZ., 1980 Sanders, 1978 ; Pfister, 1980). Grouqpe_II espe'ces tole'rantes, toujours en petites quantitq6s. Elles ne fluctuent pas significativement dans le temps. La majoritq6 dlentre elles sont des prdateurs : Neq@q)htys hombergii, Morphysa beZZii, GqZycera spp., PZatynereis dumeriZii ... qgqrq2qLiqpq@q@_III : espq6ces sensibles*qui disparaissent tout d'abord puis I reapparaissent en e'largissant leur rq6partition q&cologique par rap- port aux conditions normales : Spio spp., Notomastus qlatericeus, Phy6qNodoce spp., Nereis diversicoZor ... qgqL-oupq2_IV : espe'ces opportunistes, essentiellement des Polyche'tes qCqirritulidq6s et Capitellidq4s (Chaetozone setosa, Heterocirrus spp., PoqZydora spp., Cirratuqlus cirratuqZus, Andouinia tentacuZata, Capito- mastus minimum ...). A l'intrieur de ce groupe une succession q6co- logique a pu qiq@tre prq6ce'demment de'finie (Gle'marec et Hussenot, 1981 Le Moal, 1981). Gqrouqpq@_V : espq6ces opporqtunistes, trq6s peu nombreuses (2 ou 3) qui restent seules mais sont prq6sentes en densitq6s conside'rables lqa oq@l la pollution est maximale : CapiteqZZa capitata, CapiteqlZides giardi, ScoqleZepis fquZiginosa, Oligochq@tes ... Ces diffe'rents groupes peuvent coexister et le long du gradient de pollution organique, six q6tapes peuvent q@q@tre de'finies (Fig. 2a) q@qjLaqpe_l : groupes I, II et III sont en plus faibles densitq6s qquIen cond7it7lons normales, mais iql n'y a aucun changement qualitatif. q@qjjape_2 : 11cosyste'me est dq6sq6quilibrq6 et le groupe III est domi- nant. Etaqpqtq@_q4_q@t_6 : elles sont dq6finies respectivement par la prolifq6- ration des groupes IV et V. Etaqp_esq-3q-etq-5 : elles correspondent a des e6cotones ; le groupe II' est seul car la comp2q6tition entre groupes est affaibliqe. Ce diagramme (adapt8q6 de G18q6mqarec et Hily, 1981) est statique. Il illustre par exemple la disposition dq'aureq'oles concentriques de pollution d crqoissante en face dq'qun effluent urbain arrivant en mer. La perturbation creee par 1q1AMOCOq-CADIZ apporte une nouvelle dimen- sion temporelle a ce diagrqamme. Nqous proposons de d8q6crire 1q1instal- lation progressive des diff8q6rents groupes (Fig. 2b) et la r8q6gression de cette faune speq'ciale de substitution (Fig. 2c). la toxicit2q6 des hydrocarbures. 193 65 F-3 2 1 0 8qV6qA 14q1 IV X X ... .......... ... q6 4-2 4 b qfrq2 q2-6 2-4 2 q1 0 .......... ... ...... .. .. ...... ..... . X ... ......... X X FIGURE 2a. Importance relative des diffq6rents groupes et definitions des e0qf0qtapes 0 'a 6 le long du gradient d'excq6s en matie're organique. 2b. Installation de la faune opportuniste (groupes IV et V) apres stabilisatioD de la pollution. 2c. Evolution regressive de la faune opportuniste et de'velop- pement du groupe III. Trois mois aprq6s la perturbation (t3), la population est large- ment stabilisq6e, c'est le dq6but de la deuxie'me phase de Marchand. A tq8, tous les peuplements sont de0qf0qtruits (Fig. 3 et 4). On notera que dans l'Aber Wrac1h, quelqques espe'ces tole'rantes ou du groupe IV meurent ulte'rieurement (t2l). Dans l'Aber Wrac1h, les sables gros- siers d1aval ne sont pas affecte's (e'tape 0, cf. TaqBleau 1), le peu- plement montre seulement quelques fluctuations saisonnie'res. Partout ailleurs les hydrocarbures sont stockq6s dans les sq6diments en fonc- tion de 1q1h6qydqrodynamisme deq'croissant de 2qlq'aval vers 6qlq1amont, et il est possible de deq'qcrire 1q1installation des diffeq'rents groupes dqles- peces le long du mqgme gradient. Dans les sables dunaires (DUD de lq'Aber Benoit (Tableau II), la teneur dq1hydrocarbures est partout infeq'rieure 'a 50 ppm et la com- munaut6q6 montre une leq'ge're baisse des densiteq's (eq'tape 1) et une eq'tape 2 de deq'seq'quilibre appara8qlt, de fa8qgon fugace. Pour tous les autres seq'di- 0qb0qm ments des deux abers, l6qa o6qa la teneur en hydrocarbures est toujours sup2q6rieure a 1 000 ppm (Marchand et Caprais, 1981), leurs peuplements sont tr6q6s appauvris. 194 TABLEAU I. Levolution des peuplements dans 1ABER WRAC'H est illus- tree par llutilisation des e'qtapes dq6finies sur la Figure 2. t 8 13 17 21 25 29 33 VS - - 6 5 4 2 2-6 SHV 4 4 4-2 4-2 4 - 2-4 FV 4 4 4-2 3 4 3 3 SF 4 4 4-2 2-4 4-2 2-4 2-4 DU 4 4 4-2 2-4 4-2 q1 2 SG q0 0 0 q0 q0 - - TABLEAU II.L'volution des peuplements dans 11ABER BENOIT est illus- trq6e par ilutilisation des q6tapes dq6finies sur la Figure 2. t 8 13 17 21 25 29 33 vs 6 6 6-2 2-6 6-2 6-2 6-2 SHV 4-2 4-2 2-4 2-4 2 6 2-4 DU3 4-2 4-2 6-2 2-6 2 2 2 SF - 4 4 2 2-4 4-2 2-4 DU2 4 1 1 1 1 q1 2 DUI I I 1 2 0 0 0 A partir de t8 le groupe IV est prdominant sur les groupes I et III (e'tape 4). Dans les aires envasq6es d'amont, les peuplements sont tre's sinistrq6s et le groupe V prolifq@re (e'tape 6). Nq6anmoins, dans l'Aqher Benoit, les peuplements de la partie moyenne (DU3 et SHV) montrent dq6jqa la prq6dominance du groupe III sur le groupe I, ce qui de'finit une tape intermq&diaire 4-2 qui apparaqlt 'a t8 dans l'Aber Benoit mais seuqlement a qt17 dans qI'Aber Wrac1h. Le long du gradient spatial dlexcq@s en matie're organiue de l1aval vers ql1amont, les etapes successives 6 et 4 illustrent l1apparition successive des groupes V et IV, les q6tapes intermdiaires 6-2 et 4-2 celle du groupe III toujours domine' par les groupes V ou IV. Les conditions hivernales, par leurs tempq&tes frq6quentes, brassent eqt oxygq@nent les sq6diments. Aprq@s le premier hiver q(t12q), Marchand et Caprais notent la de'contamination chimique dans l'Aber Benoit au niveau des sables dunaires DU1 et DU2, puisque la teneur en hydrqocarbures est inf2q6rieure a 100 ppm. Cependantq, dans la majo- rite' des autres se'diments, cette teneur, a t12q, est toujouqrs qSqU0qP2q6q- rieure 'a 1 000 ppm et peut atteindre plus de 10 000 ppm dans les vases et dans les sables h2q6t2q6rog2q@nes envas2q6s. Cq'est seulemen4qt durant le deuxi8q@me hiver (t2l) que la d2q6contamination peut 4q&tre prouv2q6e par des donn6q6es 0qbiologiques (Tableau I) dans 1'ensemble de 1 q@q'Aber Wracq1h. La Figure 2c iilustre cet0qte r2q4gressioqn des groupes opportunistes V et IV, 1q'extension puis finalement la disparition du groupe III avec les diff8q6rentes 8q6tapes etape 2-6 le groupe III domine les groupes V et IV. 8q6tape 2-4 le groupe III domine les groupes IV et V. 195 Sg Du Sf FV Shv VIS w 8 13 17 SV 21 25 IV 29 T 33 FIGURE 3. Evolution temporelle de la densite' des peuplements et de l'importance relative des diffe'rents groupes de I 'a V, ceci dans les diffe'rentes unites blos6dimentaires de l'Aber Wrac'h, de l1aval A gauche vers l1amont a droite. 196 Dui DU2 Sf DU3 Shv VS V IV .............. ..... ..... ... ....... ..... ..... FIGURE 4. Evolution temporelle de la densite' des peuplements et de 11importance relative des diffe'rents groupes de I 'a V, ceci dans les diff6rentes unit6s, biose'dimentaires de l'Aber Benoit, de l'aval 'a gauche vers l1amont 'a droite. 197 6tape 2 le groupe III domine les groupes I et IV. I 6tape I le groupe I domine le groupe III, le groupe IV est encore present, les densite's totales sont encore plus faibles qu'en conditions normales. Ces quatre 6tapes prouvent que 11q6cosystq6me est encore de'sq6qui- libre'. Le dq6but de 11q6volution rq6gressive de cette faune de substi- tution semble simultane' dans l'Aber Wrac1h. Dans l'Aber Benoit, cette q6volution apparaqit diffe'remment au sein des peuplements, plus rapidement lqa oqa les se'diments sont bien oxyge'ne's, apres t8 pour les sable5 dunqairqeqs (DqU2), aprqe'q5 t1q3 pour leqs sables he'te'rogqe'nes qenvaseqs (SHV), aprq@@s t17 pour les sables fins, les DU3 et les vases sableuses (VS). Deux ans apre's la perturbation (t25), les sables dunaires (DUqI) presentent un peuplement qui semble normal (e'tape 0), tous les autqres peuplements sont en dq6sq6quilibre (q6tapes 1, 2 et 2-4) et, dans les aires envasq6es, la de'contamination commence a peine (q6tape 6-2). Ensuite les processus dynamiques sont plus lents et, au cours du troisiq@me hiver (t33), les sables dunaires (DU2 et DU3) atteignent 11q4tape 2, les sables fins et les sables hq6te'rogenes envasq6s 1'q6tape 2-4. Les dragages mq6caniques qui ont q6tq6 pre'conise's pour rq6sorber les poches de vase et d1hydrocarbures ont eu lieu en avril 1980, a proximitq6 des sables hq6tq6rogeNnes envase's. Leur peuplement te'moigne d'un accroissement passager de la perturbation, q6tape 6 a' t29. Il faut moins de deux mois en effet pour que prolifq@re une nouvelle gq6nq6ration de CapiteqlZa capitata. Cette q6tape 6 est apparue comme '1q6vation dans 1q"volution logique de 11e'cosyste'me ; elle est une e e provoquee par une intervention anthropique. Il est important de noter que les conditions hydrodynamiques ne sont pas aussi efficaces dans l'Aber Wrac1h que dans l'Aber Benoit, ce qui se traduit par une decontamination qui nlest pas si- multanq6e dans les abers. L1q6volution rq6gressive de la faune oppor- tuniste nlest pas aussi rapide dans l'Aber Wrac1h et nous pouvons observer qu'au mq&me moment (t25 par exemple), les m@q@mes peuplements restent plus perturbe's que dans l'Aber Benoit : DqU2 SHV SF qVqS Abers Benoit q1 2 2-4 6-2 Wrac'h 4-2 4 4-2 4 A t33 cette diffq6rence disparaqlt : DU2 SHV SF qVqs Abers Benoit 2 2-4 q@ q_-4 q@ q_-q2 Wracq1h 2 2-4 2-4 2-6 DISCUSSION Ind2q6pendemment des fluctuations cycliques annuelles, 1q'6q6voluq- tion temporelle des diff2q6rents groupes le long des gradients spa- tiaux que constituent les a2qberqs, montre une 4q6volution qacyclique et des processus chronologiques tout a fait similaires de ceux deq'crits par Le Moal (1981) dans la zone intertidale de ces abers. Ils sont r8q6sum8q6s sur la Figure 5a. Il y a d'a6qbord req'gressqion quasi-totale des 198 populations initiales durant la premi@e p6riode tot8. Le groupe II ne fluctuant pas significativement ou e'tant seulement dominant lors- que les autres groupes disparaissent (q6tapes 3 et 5), nlest pas repre'- sente' sur les figures. Entre t8 et t13, clest le commencement de la seconde phase et la faune opportuniste slinstalle. Le groupe IV est abondant partout A partir de t8 dans l'Aber Wrac1h, ses densite's sont maximales au printemps, clest-a'-dire q! t13 et t25. Son importance est de'croissante dans les peuplements de l'Aber Benoit apre@s t8, t1q3 Ou t17 selon 11hydrodynamisme. Le grqoupe V appqaraqlt de faq@on significative seulement dans les vases sableuses de l'Aber Benoit, mais, sur un plan ge'ne'ral, le de/- veloppement de la faune opportuniste IV et V est maximal entre tqe et t2q0. Le groupe III re'apparaqlt A t13 et son de'veloppement est trq@s important partout durant le deuxie'me hiver. De faqqon simultanee, le groupe I re'apparaqlt, un an aprq@s la marq6e noire, mais aprq@@s deux annees son importance est encore limitq6e par le de'veloppement anormal du groupe III qui semqble entrer en comp6tition. Avec le dq6veloppement des groupes I ou III, essentiellement aprq@s t2q0, commence donc la phase de reconstitution. A cq5tq6 de ce schq6ma gq6nq6ral d'q6volution, nous avons regroupq6 les diffq6rents scq6- narios d1q6volution temporelle : celui des vases saqbleuses de l'Aber Benoqtt est e'voquq6 plus haut (Fig. 5qb), celui des vases sableuses de l'Aber Wrac1h montre d'abord la prq6dominance du groupe V, remplacq6 ensuite par celui du groupe IV (Fig. 5c). Pour les autres sq6diments de l'Aber Wrac'h, il y a deux pics successifs du groupe IV sq6parq6s par un maximum du groupe III a t2l. Le cas aberrant des sables he'tq6- rogq6nes envasq6s de l'Aqber Benoit est illustrq6 par la recrudescence du groupe V, apre's celui du groupe IV. Pour l1ensemqble des se'diments dunaires des deux abers, les Fi- gures 5d, e et f, illustrent les diffq6rentes possiqbilite's o@i la com- pq6tition entre les groupes I et III apparaqlt de faqqon tout 'a fait evidente. Ces scq6narios sont 0qkablis sur les densitq6s relatives des diffe'- rents groupes, mais la communaute sera juge'e en e'tat d'q6quilibre lorsque les caracte'ristiques essentielles (A = aqbondance relative des esp6qkes ; S = nombre d'esp0qkes ; B = biomasse) Irestent relati- vement inchangq6es, hormis les fluctuations saisonnieres. Qualitati- vement, 8qVensemqble de ces peuplements est encore en dq6sq6quiliqbre et 2qlq'analyse simultan2q6e des trois param2q6tres S, A, B, suggqeq're des faits compleq'mentaires qui me'ritent dql6q&tre suivis dans le temps. On notera que clest dans le cas des s8q6diments d'aval de lq'Aber Benoit que lqlon est encore le plus eq'loigneq' dq1une certaine stabilisation de ces trois facteurs. Au contraireq, crest dans le cas des s2q6diments envas2q&s que 1q12q60qquilibre semble atteint le plus vite. Nous avons de'ja' expose' (Gleq'marec et a6qZ., 1981) comment le mod0q@le q'thodologique, mis au point dans le cas des effluents urbainqs arri- vant en mer, pouvait 4qi8q@tre utiliseq' dans le cas de catastrophes peq'tro- li8q@res, et 1q1e'chelle de temps des pheq'nomqe'nes observ2q6s semble avoir I et6q6 la qm6q@me dans le cas du Tanio (Aelion et Le Moal, 1981). 199 Mod6le gdn6ral Ontertidal par ex.) a -to 12 20 20 3,3 VSB ........ b .................. ............... ........... 11T VS W Y SHVW C S F W FVW -III SHVi3 M: SF DU W DUqB nz DU3B 00, N% e f DU, B 3 12 2b 29 33 FIGURE 5. Principaux mod@les de succession temporelle des diff6rents groupes I 'a V au sein des peuplements des Ahers Wrac'h (W) et Benoit (B). Le mode'le ge'ne'ral (Fig. 5a) est synthe'tique il illustre les trois phases apr@s une telle perturbation mortalite', substitution, reconstitution. 200 VS8 a VS W ............... --,,SHVW FVW SFW SHVqE3 ........................... -- - - - - - - - - - - - - - - ................ ........... ............. DUI B c DUW -------------------------------- SFB DU2 B qDqU3B d qt qL qL a q13 17 21 q25 25 33 ---- ------------ FIGURE 6. Evolution des parame'tres synthe'tiques, nombre dq'qesp2q6ces S en trait plein, A abondance des individus en pointilleq', B biomasse en tireteq'. La stabilisation simultaneq'e dans lqe temps de ces trois parame'tres n'est pas acquise dans le cas des peuplements d'aval de 1'Aber Benoit. 201 L'accident de l'Amoco-Cadiz slest re've'le' pour nous une expe'- rience d'cologie exprimentale tout a fait exceptionnelle. Elle permet d'apporter des 616ments de re'flexion quant aux mq6canismes qui mettent en place de telles se'quences, probl@me posq6 rq6cemment par Connell et Slatyer (1977). Dans la success ion dq6crite, les premiers stades correspondent a des espq6ces a vie courte, de type opportuniste, capaqbles de sup- porter une proportion de matiq6re organique encore importante dans les se'diments. Cette premiere phase de recolonisation par substitu- tion est mieux expliquq6e par les caracqtq6ristiques qbiologiques des espq6ces en cause que par queique propriq6tq6 e'mergente de la commu- taute' toute entie're (Sousa, 1980). Si ces premieres espe'ces ne faci- litent pas le retour des espq@ces caracte'ristiques des stocks ulte'- rieurs (modq@le de facilitation), il leur est difficile d'inhiqber la reapparition des espe'ces 'a strate'gie diffq6rente qui slinstallent plus lentement, mais de faqqon plus durable. Les phq6nomq@nes de compq6tition existant, peu a peu les premiq&res especes disparaissent. Ce type de succession secondaire peut correspondre au modle de tolrance de Connell et Slatyer, dont il n1existe jusqu'ici que peu d1exemples connus. Les premiq6res 6tapes ont donc q6te' relativement rapides, pour les suivantes c'est plus long. La reconstitution, si elle est quali- tative, doit aussi q@tre quantitative, e'nerge'tique. L'approche que nous avons de'veloppe'e semble plus efficace que 11q6tude dynamique de certaines populations. Elle s'est inspire'e de la recherche des indicateurs biologiques bien connus en milieu ter- restre et dleau douce. Mime si les paramq@tres q6cologiques aqbiotiques semqblent normaux, ces bioindicateurs peuvent re'vq6ler des perturba- tions dans les e'cosyste'mes, ulil est impossible de dq6tecter par une a,nalyse des paramq6tres physiques. REFERENCES CITEES Aelion, M. et Y'. Le Moal, 1981, Impact q6cologique de la marq6e noire du "Tanioll sur les plages de Trq6gastel (Bretagne nord-occiden- tale) : Rapport Contrat CNEXO, n' 80/6295 Bellan, G. , Bellan-Santini D. et J. Picard, 1980, Mise en q6vidence des mode'les e'cobiologiques dans des zones soumises a perturba- tions par matie'res organiques : Acta Oecologica, Oecol. Applic. vol. 3 (3), pp. 383-390 Cabioch, L., Dauvin J.C., Mora-Bermudez J. et C. Rodriguez-Baqbio, 1980, Effets de la mareq'e noire de lq'qI'Amoco-Cadizq" sur le ben- thos sublittoral du nord de la Bretagne :Helgoi2qEnder wiss. 0qMeeresunters., vol. 33 (1-4), pp. 192q-208 Chasse', C., 1978, Impact 2q6cologique dans la zone c2q6ti2q@qre concern6q6e par la mar2q6e noire de 2q1q1q"Amoco-Cadiqzqiql : Mar. Poll. Bull., qvol. 11, pp. 298q-301 Chasse', C., L'Hardy-Halos M44qJ. et Y. Perrot, 1967, Esquisse dq'un bi- lan des pertes biologiques provoues par le mazout du "Torrey- Canyon" : Penn ar Bed, vol. 6, pp. 107-li2 202 Connell, J.H. et R.O. Slatyer, 1977, Mechanisms of succession in na- tural communities and their role in community stability and or- ganisation : The Amer. Naturalist., vol. 3 (982), pp. 1119-1144 Gentil, F. et L. Cabioch, 1979, Premi6res donn6es sur le benthos de l'Aber Wrac1h (Nord-Bretagne) et sur 11impact des hydrocarbures de l'I'Amoco-Cadiz" : J. Rech. Oce'anogr., vol. IV (1), pp. 35 Gl6marec, M. et C. Hily, 1981, Perturbations apport6es a la macro- faune benthique de la baie de Concarneau par les effluents ur- hains et portuaires : Acta Oecologica, Oecol. Applic., vol. 3, pp. 139-150 Gle'marec, M., C. Hily, E. Hussenot, C. Le Gall et Y. Le Moal, 1981, Recherches sur les indicateurs biologiques en milieu se'dimen- taire marin : Colloque "Recherches sur les indicateurs biolo- giques", A.F.I.E., Grenoble Gle'marec, M. et E. Hussenot, 1981, De'finition d'une succession 6co- logique en milieu meuble anormalement enrichi en mati@re orga- nique a la suite de la catastrophe de 1'"Amoco-Cadiz" :In "Amoco-Cadiz, Conse'quences d'une pollution accid6ntelle par les hydrocarbures", CNEXO Ed., pp. 499-512 Le Moal, Y., 1981, Ecologie dynamique des plages touche'es par la mare'e noire'de 11"Amoca-Cadizil :The'se 3@ cycle, Universite' de Bretagne Occidentale, 131 pp. Le Moal, Y. et M. Quillien-Monot, 1981, Etude des populations de la . I macrofaune et de leurs Juveniles sur les plages des Abers Benoit et Wrac'h :In "Amoco-Cadiz, Cons6quences d1une pollution acci- dentelle par les hydrocarbures", CNEXO Ed., pp. 311-326 Marchand, M., 1981, Bilan du Colloque sur les cons6quences d'une pol- lution accidentelle par les hydrocarbures : In Rapport Sc,ient. et Techn., CNEXO, 44, pp. 1-86 Marchand, M. et M.R. Caprais, 1981, Suivi de la pollution de 1"Amoco- Cadiz" dans lleau de mer et les se'diments marins : In "Amoco- Cadiz, Cons6quences d'une pollution accidentelle par les hydro- carbures", CNEXO Ed., pp. 23-54 Pearson, T.H. et R. Rosenberg, 1978, Macrobenthic succession in rela- tion to organic enrichment and pollution of the marine environ- ment : Oceanogr. Mar. Biol. Ann. Rev., vol. 16, pp. 229-311 Sousa, W.P., 1980, The responses of a community to disturbance the importance of successional age and species' life histories Oecologia (Berl.), vol. 45, pp. 72-81 Ce travail a 6t6 r6alis6 avec ltaide financi6re de la N.O.A.A. (Contrat C.N.E.X.O. 79/6180). Il a fait partiellement l'objet d'une communication pr6sentge au 166me European Marine Biology Symposium de Texel, Septembre 1981. 203 LES EFFETS DES HYDROCARBURES DE L'AMOCO-CADIZ SUR LES PEUPLEMENTS BENTHIQUES DES BAIES DE MORLAIX ET DE LANNION D'AVRIL 1978 A MARS 1981 par Louis CABIOCH Jean-Claude DAUVIN 2, 2 Christian RETIERE Vincent RIVAIN et Diane ARCHAMBAULT3, 1) Station Biologique de RoScoff, 29211, ROSCoff, France 2) Laboratoire Maritime de Dinard, 35801, Dinard, France 3) Universit6 de Laval, Qu6bec. 1) INTRODUCTION Les premi6res nappes d'hydrocarbures de I'Amoco-Cadiz atteignent le littoral de la r6gion de Roscoff et les c6tes orientales de la baie de Morlaix le 21 mars, quatre jours apr6s 1'6chouage du p6trolier sur les roches de Portsall. Alors que les masses d'eau charg6es en hydro- carbures plus toxiques transitent rapidement sur 1'ensemble de la r6- gion, des particules ol6os6dimentaires contaminent les fonds sublitto- raux. Elles se d6posent pr6f6rentiellement dans les zones calmes et forment localement des poches de mazout r6siduel; en baie de Morlaix, on note leur pr6sence le 3 avril dans les sables fins peu envas6s de la Pierre Noire par 20 m6tres de profondeur et le 13 avril dans les chenaux des rivi6res de Morlaix et de Penz6 (CABIOCH et al., 1978). Durant la m@me p6riode ce ph6nom6ne est observ6 par MARCHAND etCAPRAIS (1981) en baie de Lannion. La connaissance de la composition et de la distribution des com- munaut6s benthiques de cette r6gion depuis 1968 (CABIOCH) et celle de la dynamique du peuplement des sables fins de la Pierre Noire engagee depuis 1977 (DAUVIN, 1979) permettent une meilleure 6valuation des cons6quences de la pollution sur le macrobenthos subtidal. Ces 6tudes, financ6es par la NOAA (contrat 78/5830, 80/6145), entreprises imm6diatement apr&s la catastrophe, ont d6jA fait l'objpt de publications : DAUVIN (1979 a, b et sous presse), CABIOCH et al. (1980, 1981 et sous presse). 2) LES PEUPLEMENTS ETUDIES En Manche les s6quences bio-s6dimentaires sont principalement sous le contr6le de l'intensit6 des courants de mar6e; en cons6quence del.'en- tr6e vers le fond des baies, les peuplements des s6diments grossiers sont progressivement relay6s par des peuplements de s6diments fins plus ou moins envas6s. Certains termes de cette s6quence ont une distribution spatiale discontinue; tel est le cas des peuplements des sables fins s4par4s par de vastes 6tendues de nature diff6rente. 205 Les unites cenotiques correspondant aux principaux maillons de Cette succession bio-sdimentaire ont t6 rgulirement q6chantillon- nq6es : - les sables grossiers q6 Venus fasciata - Aff8qphioxus lance- Oqlatus dqe la baie de Morlaix (au large de la Pointe de Primel; carte 1, P.P.), peu classq6s avec pour mode la classe 1000 a 5000 microns (55 a 70% du sq6diment total) : relevq6s trimestriels d'aoqat 1977 a aoqat 1980; _ le maerl envasq6 de Trq6beurden en baie de Lannion (carte 1, L6), trq6s peu classq6, aqVec pour mode la classe 250 a 500 microns (7 qL 58% du sq6diment total) : relevq6s trimestriels d'avril 1978 a mai 1979; _ les sables fins faiblement envasq6s a Abra aqZba - Hyaqlino- ecia biqlineata des baies de Morlaix et de Lannion (carte 1, P.N., L7, L6qN, bien classq@s avec pour mode la classe 125 q@ 250 microns (42 q@ 62% du sq6diment total) : relevq6s mensuels d'avril 1977 q& mars 1981 (P.N.) et trimestriels d'avril 1978 a fq6vrier 1981 (L7 et L8); - les sables trq6s fins peu vaseux q@L TeZqZina fq@buqta - Abra aqZba en baie de Lannion (sous Saint-Efflam; carte 1 : Ll, L2, L3, L4 et L5), trq&s bien classq6s avec pour mode la classe 63 a 125 microns (70 a 78% du sq6diment total) : relevq6s trimestriels d'avril 1978 q6 fq6vrier 1981 sauf L4 d'avril 1978 a mai 1979; - les vases sableuses q6 Abra aqZba - Meqlinna paqZmata de la riviq&re de Morlaix (carte 1 : R.M.) o domine la classe des parti- cules infq6rieures q6 63 microns (47 q& 74% du sq6diment total) accompa- gnq6e d'une importante fraction de sables trq6s fins et fins relevq6s trimestriels d'aoqat 1977 a fq6vrier 1981. Les stations de la baie de Morlaix ont q6tq6 q6tudiq6es par qJ.C. DAUVIN depuis 1977; Ch. RETIq@RE et V. RIVAIN ont observq6 les stations de la baie de Lannion avec la contribution de D. ARCHAMBAULT pour ql' q6tude de la fin du 3iq&me cycle annuel. qL'q6chantillonnage a q6tq6 effec- tuq6 parallq6lement au moyen d'une benne Smith Mc Intyre et d'une benne Hamon (relevq6s : 10 prq6lq6vements a la benne Smith Mc Intyre et 4 ou 5 q@ la benne Hamon; surface q6chantillonnq6e IqM2 ou plusq),qle tamisage a q6tq6 rq6aqlisq6 sur une maille circulaire de lmm. 3) RE'SULTATS Nous avons suivi 11q6volution des paramq6tres q6q6ologiques richesse spq6cifique, densitq6 et biomasse. Les richesse specifique et densitq6 sont actuellement connueqs pour 6qla totalit2q6 des sites et jusquq, au prin- temps de 1981. Il en est de m8q6me pour les biomasseqs relatives aux stations de Primel et de la rivi2q6re de Morlaix; par contre 2q@ la Pierre Noire elles ont 6t8q6 mesur8q6es entre avril 1977 et novembre 1978, puis calcul8q6es pour les autres re6qleqv4q6s ; elles n'qont pas encore 2q6t8q6 quantiq- fi2q6es pour les stations de la baie de Lannion. 3.1) Caract2q6re du stress Les effets du stress Wont pu 0q@tre 0q6valu6q6s a6qWec pr0q6cision que sur 206 4sW 50 4 35 LE CRAPAUD' 2p 45' :j*j:::::j:::::: L i .......... .......... ...... + + ..... .... . .......... ME . ....... ........... .. . . .... ..... ............. A G 2 r T..... c I ScEM., r,* 40' 0 D 2qX E K F 4qr 4Sirqw 4W Carte 1 Rq6partition des stations d'q6chantillonnages en baie de Morlaix et en baie de Lannion. A fonds rocheux B - C communautq6 des sdimenqts grossiers q@ Amphioxus - Venus fasciaqta, relativement indpendante de 1'tagement (C : facis d'qifaune @q): SabeZqZaria spinuqZosa). D - E communautq6 du maq6rl (D : faciq6s q6 Lithothamnium coraZqZio-qEdes var. coraqZqZioides; E : faciq@Rs q6 L. coraZqlioqF0qdes var. minima). F communautq6 des sables dunaires fins q6 Abra prismatica- GZycymeris gqZyqcymeris. G qc6q@ I qcommunaut6q6 des s6q6diments fins 6q6 Abra a2qZba (G : faci0q6s sablqeux 0q6 HyaZinoecia bi0qZineata; H : fqaci8q6s qenvas4q6 q@0q) MeZinna pa6qZmata; I : faci8q6s h4q6t4q6rog4q6ne envas4q6 8q6 Pista cristata). 0qj communaut8q6 des cailloutis et graviers pr4q6littoraux c0q6tiers faci6qLq-s 2q@ Ophiothrix fragi2qlis et 6q6 Bryozoaires dominants dans 24qPencro8qOtement. K communaut0q6 des cailloutis et graviers pr0q66qlittorqaux du large faci4q@s dq'ensablement 8q6 PorqeZ0qZa concinna. 207 les peuplements de la baie de Morlaix pour lesquels nous disposions d'observations juste avant la pollution. Richesse specifique. (fig. 1) -*-Cycle normal IOCycle q-1p2pppp20CYcle -q*--q4q-3)Cycle-1P- N.e pe'ces A.C 100- PN 4 L7 --M L8 qIff "4q? 50- 'A' 2qT '0q4 'F' 0' 'E9 'F' 'A' 'qJ' 'A' '0' 'DI 'qr 'A' 'J '0' '8q0 0qT 1977 1978 1979 q1980 Figure 1 Peuplement des sables fins q@ Abra aZba - qIq@qyaqZinoecia biqZineata q6volution de la richesse spq6cifique des relevq6s (4 prq6qlq@vements q@ la benneHamon) d'avril 1977 q@t fq6vrier 1981 (A.C. : dq6but de la pollution par les hydro- carbures de l'"Amoco Cadiz"). A la Pierre Noire la richesse spq6cifique totale passe de q9q4 es- pq6ces le ler mars a 61 le 3 avril; cette brusque diminution est due pour une grande part a une forte rq6duction du nombre d'espq6ces d'am- phipodes (20 le 18 janvier, 24 le ler mars, 10 le 3 avril et 7 le 25 avril). Par contre dans les peuplements de Primel et de la riviq6re de Morlaix la richesse spq6cifique totale se maintient, la chute du nombre d'espq&ces d'amphipodes 6tant compensq6e par 4qVaccroissement concomitant de celui des polychq&tes. Les effets du stress sur le groupe des amphipodes sont d'autant plus intenses que les espq6ces avaient, avant la pollution un haut de- grq6 de constance dans le peuplement; ils concernent les espq&ces sui- vantes : Pierre Noire Rivi8qke de Morlaix Primel Aqn32qpe6qlisqca armoricana A8qmpe0qlisca armoriqcana AqnqTpeZisca armoricana Aqnqype2qZisca brevico4qrnis Ampe2qZisca brevico4qrnis A6qmpe6qlisca spinipes Aqr32qTeZisca 0qspinipes Aqn24qVeZisca tenuico4qrnis A4qmpeZisca tenuicornis Aqn36qpe0qZisca typica Cheirocratus intermedius Cheirocratus intermedius 20qPhotis 0qZongicaudata 12qPari8qambuqs typicuqa Corophium crassic6qmq-ne MeZita qobtusata Me0qZita gZad8q4qcqrqdq. MegaZuropus agiZis q- Ampe6qZiqsca spinimana 208 Des 1972, les travaux de SANDERS et al. ont mis en 6vidence le caract6re desp6ces "tests" que pouvaient pr6senter les amphipodes vis-q&-vis de ce type de pollution. Dans le mq6me esprit les q6tudes ex- pq6rimentales de LINDEN (1976), LEE, WELCH et NICOL (1977), LEE et NICOL (1978a et b) ont montrq6 1'extrq@me sensibilitq6 des amphipodes aux hy7 drocarbures. Cette sensibilitq6 est confirmq6e par les observations in- situ de SANDERS et al. (1980), DEN HARTOG et JACOBS (1980), ELKAIM (1980) et ELMGREN et al. (1980) relatives aux marq6es noires de West Falmouth en 1969 aux Etats-Unis, de l'Amoco-Cadiz sur les cq6tes de Bre- tagne et de la Tsesis en mer Baltique. Cependant des recherches de CABIOCH et al. en baie de Morlaix (1981) il ressort que certaines espq6ces appartenant a d'autres groupes zoologiques rq6agissent plus ou moins intensq6ment auqx effets du stress; il en est ainsi des polychq&tes PhyqZqlodoce kosteriensis, TerebeqZqZides stroemi et de nombreuses espq6ces de micro-mollusques. Les fortes morta- litq6s subies, en baie de Lannion, par les populations de mollusques, surtout celle de Pharus qZegumen et de 0qVoursin Echinocardiuqm cordatuqm, viennent a ql'appui de ces observations. Il convient de noter, qu'en rq@-gle gq6nq6rale, ce sont les annq6lides polychq6tes qui constituent le noyaqu d'espq6ces le plus rq6sistant (tabl.qi ). L'amplitude du stress est donc fonction de la composition origi- nelle des peuplements et varie fortement d1une unit cnotique a ql'au- tre. Densitq6s et biomasses Les travaux de DAUVIN montrent qu'qa cet q6gard les effets du stress sont particuliq6rement dq6vastateurs sur les peuplements de sables fins de la Pierre Noire (1979). Les densit6s passent de 8707 q6 1808 individus par m2 et la biomasse de 5.0 q6 2.5 g. par m2 soit une rq6duction respec- tive de 80 et 50 % des valeurs initiales; elle est essentiellement due q@ 11q61imination des ampeliscidq6s An8qVeqZisca armoricana et ArapeqZisca te- nuicoqrnis et a la diminution des effectifs des populations d'Aqmqpeqlisca' sarsi. Ces trois espqL-ces reprq6sentaient a elles seules 90 a 99 % des densitq6 et biomasse moyennes annuelles du peuplement.q(qfiq5.q2,q) Sur les peuplements des sables grossiers de Primel et des vases sableuses de la riviq6re de Morlaix, les effets du stress,- en terme de densitq6 et de biomasse, sont peu marquq6s. En conclusion, il apparait clairement que les effets du stress sont sq6lectifs au niveau : - des espq6ces et groupes zoologiques : les amphipodes et plus spq6cialement les ampeliscidq6s se rq6vq&lent extrq6mement sensibles; - des peuplements : la communaut8q6 infralittorale des sables fins a Abra a2qlba Hya2qZinoecia bi6qlineata est la plus intens8q6ment pertur- b4q6e. 3.2) Evolution a long terme des peuplements. Les peuplements des s8q6diments fins sabloq-vaseux subissent les per- turbations les plus; qsqLgnificatives; dans cet expos2q6-q, 1q'2q6tude de leur dy- namique fera donc lq'objet dq'une attention toute partiqa0qA0q@8q@i2q6re. 209 Pierre Noire Aricideu minuta Tharyx marioni Aricidec cerrutii Abra prii;matica % Aricidpa fraaiZis Nucula hunZpy@ % Aiediomactus fragilis Venus ovatc % % Fteonp.Zonga CulteZIus P,'Iucidus % Exoqorc, habe.: Bathyporeia tenuipes tiotomactus latericeus Urot6Y pulcheZIc % Spio fizicornis PhascoZion strombi L % L 8 % 7 % %% % % % % % % Sphiophanes bombyx I Owenia fusiformis %% Minna palwt, CZumene oerstedii Ampharete ambei Chaetozone setosa % Nematonereis unicornis % DipZocirrus glaucus 1 Glycera convolutc Praxi'llura sp. % Marphysa beZZii 0 HyaZinoecia bilinecta % I Nephtys hombergii Acrocnida brachiata NucuZa turaida % Thyasira @Zexuosa Paradoneis armato Ophiura albida Abra oZba PeriocuZodes Zongimanus I %% %% J %%Cirratulidae ind. I %Stylarjoides pZumos a' % % %% % % % % Tableau I Espkes dont la constance est sup6rieure ou 6gale A 90% sur la dur6e totale des observations d'avril 1978 A f6-vrier 1981 aux stations Pierre Noire, L 7 et L 8* -*-Cycle normal-4-10 Cycle -)--4-2Cycle @-4-3Cycle - N.m2 40 000- A. C A. a A. s 30000- A.t 20 000- 10000- -T - - rqj-r-2qgr-,6qWr 6qW I0qX 8qWq( 6qM 8qT 4qW 8qW IS 'q?qd qWq( 0qT 2qW 4qW qI0qX 'qIW 1977 197q6 1979 1980 Figure 2 Peuplement des sables fins q@t Abra aqZba - HyaqZinoecia biqZineata de la Pierre Noire : q6volution de la densitq6 totale d'avril q1977 -q@ fq6vrier 1981 avec mise en q6vidence de la part des trois espqeces d'0qk2qTeqZll@sca trq@s dominantes (A.a. AmpeqZisca armoricana; A.s. : AqmpeqZisca sarsi; A.t. : An8qpeZisca tenuicornis (A.C. : dq6but de la pollution par les hydrocarbuqres de 1q"'Amoco Cadiz"). 3.2.1) Peuplement des sq6diments grossiers q@ Venus fasciata - Aqmphioqxus qlanceoqZatus. Pollution (tabl. 2) Dq6s le 27 avril les sq6diments sont contaminq6s par les hydrocarbu- res; aux fortes teneurs enregistrq6es jusqu'en novembre (231 ppmq) suc- cq6de, en fin d'hiver, une phase de qdq6pollution rapide. Richesse spq6cifique (fig. 3) Le flq6chissement des valeurs de la richesse spq6cifique au cours du premier cycle annuel suivant la pollution semble surtout liq6 Cq@ la di0qsparition de crustacq6s et de polychq6tes; cependant, de 1978 2q@ 1980, on note globalement, abstraction faite des fluctuations s0qaisonni2q6resq, un accroissement graduel auquel contribuent largement les a2qm0qphipodes (48 taxons en ao8qat 1978 contre 80 eqn ao8qCqit 1980; 4 esp8q6ces dq'amphipodes en ao2q@4qit 1978 contre 19 en aqo,6qat 1980). Densit8q6s et biomasses Les densit8q6s tr2q6s faibles varient de 100q'2q6 290 individus, par 8qM2q; maximales en ao6qat, minimales en f6q6vrier, les valeurs sont du m0q6me or- dre de grandeur d'une ann8qe,4qa 1q1autre. 211 --------------------------------- ----------------------------------------------------- 7------------------------------------------- avri 1 1978 :ao5t 1978 :nov. 1978 f6v. 1979 mai 1979 ao6t 1979 nov. 1979 f@v. 1980 mai 1980 aolit 1980 nov. 1980 f6v. 198 1 -- ------------------ L----------- 4---------- 4--------- ---------- - --------- --------------- -------------- - --------- - ----------- PRIMEL 250 81 231 < 10 < 10 < 10 < 10 16 28 28 33 < 10 ------------------- L------------ ------- I---------- --------------------------------------------- ---------------------- ----------- ------------ :juin 1979 18 57 < 10 463 14 < 10 58 < 10 44 < 10 52 PIERKE NOIRE 34 -------------------- L----------- I-----------I--------- --------- --------- -------------------------------- ---------r --------- ---------- ------------ KIVIERE DE 10 3152 136 36 48 60 12 20 ? ?IORLAIX 297 333 44 ------------------- L-------------------------------- --------------------- ---------- I----------- ---------- I----------r---------- T----------- ------------ L 18 22 10 37 22 < 10 16 16 < 10 12 41 20 ------------------- L----------- 4---------- 4--------- 4---------- --------------------- --------------------- -------------------- T-- ------------------- L 58 102 91 < 10 66 58 10 12 12 20 2 ------------------- L --------- ----------- ------------------------------------------ --------- -------------------- ------------ L3 02 18 254 269 < 10 83 49 28 12 12 20 ------------------- L----------- ---------- --------- 4-----------4---------- --------------------- ---------- --------- ---------- --------- ------------ L 22 13 < 10 26 14 4 ------------ - ------------------- L------------L---------- 4---------- ---------- 4----------I--------- - --------- -------------------- ----------------- L5 69 < 1.0 132 318 < 10 1 165 < 10 < 10 12 28 < 10 ------------------ - -r -----------T----------- --------- ---------- j------------L----------L---------- L---------- L-------------------- ----------- ------------ L 26 1 14 162 63 12 28 < 10 12 28 28 6 ------------------- ------------ L---------- k--------- j---------- j---------- --------- - --------- ---------- - -------- - --------- ---------- ------------ L 31 49 74 138 < 10 165 20 < 10 12 28 20 7 ------------------- ------------ L---------- L--------- I---------- I--------- - --------- --------- --------- ---- - --------- I---------- ------------ L 16 < 10 Q 12 <10 1015 44 34 120 83 91 ------------------- ------------ L---------- L---------- I---------- j-------------------- ------------------------------------- ------------- Tableau 2 - Evolution des teneurs en hydrocarbures mesur6es en spectrophotom6trie aux infra-rouges selon la m6thodologie suivie par BESLIER et aZ. (1980) exprim6es en milligrarme par kilogramme de s6diment sec ou ppm (don- n6es BESLIER, 1981). -.*--avant 1 Cyc I e 2 Cyc I e -@-3Cyc I e N.i spi?ces n. ini lividus N 100- *---4 n -1000 A-C q50- ----4 -500 A' '2qd DqI 6qY A: 4qT 4qW '0' 'DqI Fe 'A: T 'A' 0 '0qd 'F 8qV-7 0qW 0 1977 1978 1979 q1980 Figure 3 Peuplement des sables grossiers Venus fasciata - Anrphioxus qZanceoqZatus q6volution de la richesse spq6cifique des relevq6s et du nombre d'individus (10 prq6lq6vements q@ la benne Hamon) d'aoqat 1977 q@ aoqEt 1980 (A.C. : dq6but de la pollution par les hydrocarbures de qI"'Amoco Cadiz"). Les biomasses des diffq6rentes composantes, faunistiques sont trq&s inq6gales; par exemple, le lamellibranche GqZycyme2-is gqlycymeris reprq6- sente 75 % de la biomasse totale du peuplement. Aprq6s avoir dq6passq6 25 g. par m2 en aoq3t et novembre 1977, les valeurs n'oscillent ensuite qu'entre 8 et 12 grammes. Toutefois en novembre 1979, q6 la suite d'une rq6colte plus importante de Gqlycymeris gqlycymeris, elle culmine 'a pres de 24 g. par m2 rejoignant les valeurs donnq6es par HOLME (1953) et RETIERE (1979) pour des peuplements analogues de la Manche occidentale. 3.2.2) Peuplement de maerl envasq6 Pollution La texture hq6tq6rogq6ne des sq6diments a favorisq6 le piq6geage et la rq6tention des particules olq6osq6dimentaires pendant-une longue pq6riode. Richesse sp8q6cifique Cette bioc2q6nose d6finie par CABIOCH (1968) comme un maerl 2q@qL Lithothaqr2qmium cora2qZ2qlio2qEdes a 8q6vqolu8q6 vers un nouveau faci8q6s, vraisemblaq- blement sous 1q'effet dq'un ensab4qlement li2q6 a des extractions industrielq- les; elle est actuellement tr2q6s comparable 8q@qL celle du maerl de Ricard, en baie de Morlaix : maerl 8q@ Lithothamqnium coraZ2qZioq-6qrdes var. minima. Le caract6q6re c6q6notique dominant est son extreq-me appauvrissement; aucune esp4q6ce d'8q6pifaune sessile nq'est en effet pr8q6sqente dans nos 8q6chantillons pr2q6lev8q6s apr2q6s la pollution, ce qui atteste sans doute unqe profonde perturbation. 213 Nous n'avons recolte dans ces fonds quun petit nombre desp6ces de 1'endofaune. On passe toutefois de 23 especes en avril 1978 q6 34 en fq6vrier 1979 et parmi les plus abondantes il convient de citer trois annq6lides polychq6tes : Goniada macuZata, Staurocephaqlus qke2qfersteini, Heteromastus fiqZiformis et deux pq6racarides : Nototropis swaqmqmerdani et PeriocuqZodes qLongimanus. De plus il faut noter que les deux especes q6qlectives du faciq6s hq6tq6rogq6ne envasq6, StheneZais boa et Pista cristata prq6sentes dans les q6chantillons d'avril 1978, ont ensuite disparu. Densitq6s L'q6volution de la densitq6 globale du peuplement reflq&te principa- lement celle de quelques populations d'nnq6lides polychq6tes, en par- ticulier Goniada emerita et StaurocephaqZus qkefersteini dont les re- crutements surmaille de 1 mm s'observent de faqgon synchrone en au- tomne. La premiq&re espq6ce bq6n6ficierait de la proximitq6 de biotopes servant de "rq6servoirs" q@ partir desquels se'rq6aliserait la disper- qsion des larves pq6lagiques. En conclusion il est important de rappeler que depuis 1968, le peuplement macrobenthique de ces fonds a 6voluq6, vraisemblaqblement sous 0qVeffet d'un ensablement 6qHq6 q@i 6qVextraction du maerl. Ne dis- posant d'aucun q6tat de rq6f6rence juste avant la catastrophe il est extrq&qmement difficile de connaitre la part qui revienqtqa la pollution par les hydrocarbures, dans la modification de cette communautq6. Pour cette raison son suivi q6cologique a q6tq6 rapidement abandonnq6. 3.2.3) Peuplements des sq6diments fins q6 Abra aqZba 3.2.3.1) Peuplement des sables fins q@ Abra aqZba - HyaqZinoecia biqZineata Pollution Aprq6s une phase initiale de forte pollution, que nous navons pu mesurer, les teneurs en hydrocarbures (mesurq6es en spectrophotomq6trie aux infra-rouges, exprimq6es en mg par kilogramme de sq6diment sec ou ppm) atteignent 1000 ppm a la station L8 au cours de 1'q6tq6 1978 alors qu'elles ne dq6pasqsent pas 50 ppm aux deux autres stations (L7 et P.N.), puis elles diminuent fortement au cours de ql'automne dans ql' ensemble des stations. Ensuite une remobilisation, lors des tempq@tes automnales et hivernales, des stocks d'hydrocarbures pie'ge's en zone intertidale entraine une augmentation des teneurs en hydrocarbures au printemps 1979 sur 1'ensemble des stations, (ces teneurs dq6passent 200 ppm a la station de la Pierre Noire); elles d8q6croissent rapidement ensuite, sauf en L8 o4q@qi elles se maintiennent a des valeurs voisines de 100 ppm jquqsq- quq'2qa 2qlq'automne 1979. Enfin, dans 16qVensemble des stations, apres une phase de recontamination au cours de 1q'hiver 1979q-1980, les teneurs de- viennent inf6q6rieures a 50 ppm a partir de mai 1980. Richesse sp2q6cifiquqe Apr8q6s le stress on note dans les trois stations un enrichissement progressif en esp2q6ces qui aboutit en 1980, 8q& la Pierre Noire, 8q& des va- 214 leurs du meme ordre qu'avant la pollution. Cette remont&e provient d'une part de la rq6apparition d'espq6ces q61iminq6es lors du stress, d'au- tre part de 4qVaugmentation de la frq6quence de quelques espq6ces de poly- chq6tes et de mollusques. Les mqC-mes phq6nome"nes s'observent en baie de Lannion 4qM et L8), mais avec des valeurs infq6rieures. La comparaison des prq6sences d'espq&ces dans les trois stations apporte a cet q6gard desinformations significatives dans le cas des amphipodes, groupe le plus affectq6 par le stress initial. Certaines espq&ces, qui ont survq6cu en baie de Morlaix, sont absentes en baie de Lannion immq6diatement aprq&s la marq6e noire, mais elles y apparaissent au cours du premier cycle annuel suivant le stress (Leucothoe incisa, Urothoe grimaqZdii-, Tryphosites Lon- gipes., Bathyporeia tenuipes), du deuxiq6me cycle (Am8qpeqZisca sarsi) ou du troisqiq&me (Bathyporeia eqZegans, SyncheqZidiqwn macuqlatum).Cet- te constatation tq6moigne de leur prq6sence probable en baie de Lan- nion avant la pollution et renforce 1'hypothq6se d'une plus grande intensitq6 de l'impact initial dans cette baie par rapport a la baie de Morlaix (CABIOCH et aqZ., 1981). Des espq6ces q61iminq6es temporairement par le stress en bqaie de Morlaix et qui s'y rq6installent au cours du premier cycle pertur- bq6 (MegaqZuropus agiqlis, An8qVeqZisca spinipeq@s) ou de deuxiq6me (Chei- rocratus intermedius) (DAUVIN, 1981) ne sont rencontrq6es dans les sables fins plus lonquement poqlluq6s de la baiede Lannionqqulau cours du troisiq6me cycle. Ces introductions comme les prq6cq6dentes,inter- viennent gq6nq6ralementplus tq6t en L7 qu'en L8, station la plus pol- luq6e des deux. Les espq&ces dont on est certain de 1'"insularitq6" figurent parmi celles qui se rq6installent plus tardivement dans qI'une ou 2qVautre des baies (Ampeqlisca tenuicornis, A. bre-vicornis, A. ar- moricana) ou qui n'ont pas encore rq6apparu (Photis qlongicaudata). A 6qVopposq4, An0qpeqZisca spinipes, espq6ce non insulaire, commune dans les sables grossiers avoisinant la zone poqlluq6e, repeuple les sables fins de la Pierre Noire dq&s le premier cycle. En conclusion, les effets q61iminateurs du stress, plus ou moqi0qns accusq6s selon les localitq6s d'un mq6me type de peuplement sem- bleqnt aqqcompagnq6s de rq6implantations d'autant plus tardives que la pollution rq6siduelle du sq6diment est durable et intense. La con- jugaison de l'insularitq6 gq6ographique des populations de certaines espq&ces et de leur mode direct de reproduction a aussi un effet re- tardateur. Densit8q6s et biomasses En baie de Morlaix, qc4q@ la station de la Pierre Noire la densi- t6q6 moyenne du peuplement passe de 19450 individus par qM2 lors du cycle normal a 2135 au cours du premier cycle apr6q6s la pollution. Cette tr8q&s 6qgrande diff8q6rence est le fait de la disparition et de 2qla forte r2q6duction dq'effectifs de trois esp2q&ces d'Aqfqf32qpe6qZisca largement dominantes avant la pollution. Durant les deuxi2q6me et troisi2q6me cy- cles annuels lesdensit8q6s moyennes atteignent les valeurs respectives de 3650 2q6t4110 individus par m2q. 215 Parallelement 6 la forte d6croissance des densit6s,la biomasse subit une r6duction de pr6s de 50 % pendant le premier cycle annuel aprq6s la pollution (4.4 g. par m2 contre 8.1). Dq6s le second cycle elle retrouve des valeurs comparables q@ celles du cycle normal (8.0 g. par m2) et les d'passe mq6me au cours du troisiq6me cycle an- nuel (8.7 g. par m2q). Ainsi, trois ans aprq6s le stress, bien que les densitq6s soient encore infq6qiieures, d'environ 80 %, a celles observq6es avant la pollution, les valeurs de biomasse sont du mq6me ordre de grandeur que celles du cycle normal. En effet les espq&ces subsistant aprq6s le stress, dont les densitq6s sont pass6es de 2100 q@t 3860 individus par m2 du premier au troisiq6me cycle annuel aprq@s la pollution, ont des poids individuels moyens trq6s supq6rieurs a celui des ampeliscidq6s. En baie de Lannion 1'q6volution de la densitq6 globale du peuple- ment est fortement marquq6e par les variations d'abondance, d'ailleurs difficiles q6 interpr6ter, d'une seule espq6ce, Paradoneis armata. La densitq6 de celle-ci diminue progressivement a la station L8 pour at- teindre 500 individus par m2 en janvier 1981 et se maintient a un ni- veau compris entre 500 et 600 individus par m2 q@ la station L7 pendant la pq6riode d'observation. Il faut toutefois noter qu'au cours des trois cycles annuels la densitq6 correspondant q6 1'ensemble des autres espq&ces tend q6 s'accroqltre. Les rq6introductions d'espq6ces temporairement q61iminq6es appor- tent trq&s qpeu q@ cette croissance de la densitq6 q6 moyen terme aprq6s le le stress. Elle rq6sulte principalement de quelques cas de recolonisa- q@ion par des espq6ces rq6duites en effectifs pendant le premier cycle perturbq6 et de q2qE2q1ifq6ration d'especes non affectq6es par la pollution. Les autres especes poursuivent des cycles annuels peu diffq6rents du cycle normal. Les recolonisations significatives sont le fait de trois espq6- ces : An4qpeZisca sarsi, An6qpharete acutifrons-, Nephtys han4qbergii. Alors que la communautq6 des sables fins de la Pierre Noire a pu hq6berger jusqu'q6 40.000 ArnpeZisca par m2, 6qVespq6ce subsistante, A.sarsi qne re- colonise cette vaste niche q6cologique vacante qu'q& une cadence res- treinte (fig. 4) de par la conjonction de sa distribution "insulaire" et de ses caractq@res biologiques (reproduction directe printaniq6re et estivale, femelles porteuses de 8 a 20 embryons seulement et ne se reproduisant qu'une fois, vie brq6ve ne dq6passant qguq@re un an (DAUVIN, 1979)). Limitq6e par l'insularitq6 au seul potentiel reproducteur de sa population rq6siduelle, 1'espq6ce multiplie cependant son effectif ma- ximum annuel par un facteur de 5 a 9 d'une annq6e sur ql'autre ce qui tq6moigne du succq@s de la reproduction directe. An4qpharete acut 'ifrons, (fig. 5) insulaire, de dur6e de vie infq6rieure q@ deux ans, q@ larve presque immq6diatement benthique, suit un schq6ma de recolonisation len- te du mq6me type. Par contre, le repeuplement de Nephtys hombergii (vie longue, non insularitq6, larves pq6lagiques pendant plus d'un mois) sq'effectue rapidement (fig. 6); d4q6s 1980, les effectifs estivaux, qui ne d0q6passaient pas 30 individus par m2 en 1978, attei0qgnent 170 par qw6q2q, qvaleur sup6q6rieure 0q@ celle observ0q6e avant pollution (90 par m20q). En ce qui concerne les prolif4q6rations, la br0q&ve pouss4q6e dq'Hete- rocirruqs a8qZatus 0q@qi 8qlq'autqomne de 1978 (fig. 7) est suivie, au cours du deuxi0q&me et du tqrois6qi4q6me cycle par des accroissements importants de Chaetozone setoqsaq, (Fig.q-0q70q), Spio fi0qlicorniqsq. Sco0qZop4qZoqs armiger, Thyaqsira f2qlexuosa, Abra a8qZbqa. Ainsi se dessine probablement un pre- mier 4q618q6ment d'une s8q6rie de "successions" (PEARSON & ROSENBERG, 1978), ph6q6nom0q&ne moins imm0q6diat et moins accus8q6 ici que sur les fonds sublittoraux bien plus poll0qO6q6s des Abers (GLEMAREC & HUSSENOT, 1981; GLEMAREC & HUSSENOT, sous presse). 216 N/m 2 --g-Cycle normal -,)0-4-110Cycle 20Cycle-----0--`K-30Cycle -,jo- AX. 10 10. -JA 07 @DF @A. F' 'A A 0 D' 'F- 'A' -J' -A 0 D F 1977 1978 1979 1980 1981 Figure 4 Peuplement des sables fins Abra aZba - Hyalinoecia bilineata de la Pierre Noire : 6volution de la densit6 d'Ampetisca sarsi d'avril 1977 f6vrier 1981 (A.C. : d6but de la pollution par les hydrocarbures de 1"'Amoco Cadiz"). CYCLE NORMAL - V CYCLE - - - - - - - - - 2*cYCLE - - - - - - - - 31cYCLE - 420 200 A.C. @L7 100 - 0 La \-.*PN r-r-T-T-1 ii. V M. j I ISO f-541 IMI 11,1 lil T I I MI I M7 I -T- -T- T M M M i S I SO NI W i ST NT 1977 1978 1970 1980 Figure 5 Peuplement des sables fins @t Abra aZba - HyaZinoecia biZineata 6vo- lution de la densit6 d'Ampharete grubei d'avril 1977 A f6vrier 1981 (A.C. : d6but de la pollution par les hydrocarbures de 1"'Amoco Cadiz"). 217 -4-Cycle normal-0-4- laCycle 20 Cycle 30Cycle-lo- N-m 2 150- 100- A. C Z, "I \N, h L7 so- PN LS 4, 'J' 'S7 V V V M T W IJ V M T 'S' 'N' V M V T 'S' @N @J@ 1977 1918 1979 1980 Figure 6 Peuplement des sables fins @ Abra aZba - Hyalinoecia biZineata 6volu- tion de la densit6 de Nephtys hombergii d'avril 1977 A f6vrier 1981 (A.C. : d6but de la pollution par les hydrocarbures de l"'Amoco Cadiz"). --9 Cycle normal-*-K- 1 0 Cycle -)P@- 20Cycle 30C yc le ---4- N.Mi- A-C // ". \ S-f 1000- S. f C.S- T f a r" I . - Ile V, a a 100- ts -.S. f A A/ a T. f A. a IIII` T. -H.a M' M' T V W I X M M 'J 'S' rN-TjT-rW- M T 'S' W IS V W( W S W IT 1977 1978 1979 1980 Figure 7 Peuplement des sables fins Abra aZba HyaZinoecia biZineata 6volu- tion sch6matique de la densit6 d'Abra aZba (A.a.), de Chaetozone setosa (C.s.), d'Heterocirrus aZatus (H.a.), de ScoZopZos armiger (S.a.), de Spio fiZicornis (S. f.) et de Ayasira fZexuosa (T.f.). (A.C. : d6but de la pollution par les hydrocarbures de 1"'Amoco Cadiz"). 218 3.2.3.2) Peuplement des sables tr6s fins @ Tellina fabula Abra alba Pollution Les teneurs en hydrocarbures aprq6s un pic passager en aoq3t 1978 (premiq6res mesures) redeviennent fortes de fq6vrier q@ mars 1979 (valeurs comprises entre 80 et 300 ppqm). A la fqaveur d'une recontamination du sq6diment au cours de ql'automne 1979, les teneurs dq6passent de nouveau 50 ppm en novembre (70 q@t 165 ppm) elles se maintiennent A ce niveau en L2 et L3 au cours de 1'hiver 1980, puis redeviennent infq6rieures a par- tir de mars 1980 dans 1'ensemble des trois stations. Richesse sp6cifique et densitq6s* On sait indirectement que ce peuplement a subi une perturbation considq6rable lors du stress; les immenses quantitq6s d'Echinocardium cordatum et de mollusques de diverses especes, rejetq6s morts sur la grq6ve de St-Efflam en mars-avril 1978 en tq6moignent (CHASSE & GUENO- LE-BOUDER, 1981.1. Les phq6nomq6nes observq6s a la suite du stress prq6- sentent de grandes analogies avec ceux que nous venons de dq6crire : croissance q& moyen terme de la richesse spq6cifique (fig. 8) et de la densitq6 totale, phq6nomq6nes de recolonisation. --CYCLE NORMAL -q@ - - - VCYCLE 2'CYCLE - - - - - - - - 3'CYCLE - N esp6ces 50. A.C. L2 --------- q:L3 5 2 5 I I 1,,p PN- qi,r w w i a VT T T T F I I I I I I I I I T I I I r-r-T-r-r-4q78q7 j . M j N 'I M j j a I'l j , N 1977 Isis 1979 1980 Figure 8 Peuplement des sables tr2q6s fins 4q@ Te6ql6qtina fabu6qta - Abra a6qlbqa 2q6volution de la richesse sp8q6cifique des relev2q6s (4 pr2q6l2q6vements 8q@ la benne Hamon) dq'avril 1978 8q@ f2q6vrier 1981 (Aq.Cq. : d2q6but de la pollution par les hydro- carbures de 12q"'Amoco Cadiz"). *Les r8q6sultats du suivi de la station Ll dont le peuplement pr2q6sente un caract2q6re nettement intertidal ne sont pas int2q6gr2q6s dans ce travail qui a pour objet 1q'8q6tude des communaut8q6s subtidales. De m8q6me le suivi de la station L4 a 2q6t8q6 abandonn8q6 en mai 1979, le d8q6pouillement des donn0q6es nq'apportait pas dq'informations compl6mentaires de celles recueillies aux stations L3 et L5. 219 La croissance de la densit est principalement liee a la proli fq6ration, depuis la perturbation, du Capitellidae Mediomastus fragiZis (fig. 9) dont les effectifs q6 la station L3 passent de 100 individus par m2 en avril 1978 q6 plus de 7000 par m2 en mai 1980. N.m -2 VCYCLE - - - 2'CYCLE - - - - - - 3'CYCLE 41 -------- 103 --.A L2 -L5 IqV 2_ AC- 10 j IM.M 'S. U IMI IMI IJI ISI ONS q1j, IMI IM I IJ, Is, "I qii, IMI 1978 1979 1980 Figure 9 Peuplement des sables trq@-s fins q@ TeZZina fabuta - Abra aZba : q6volution de la densit de 6qMediomastus fragiqZis d'avril 1978 A fq6vrier 1981 (A.C. dq6but de la pollution par les hydrocarbures de 1q"'Amoco Cadiz"). Les recolonisations significatives sont ici le fait de trois es- pq6ces : Nephtys hombergii, GZycera convoqluta et TeqZqZina fabuqla; mais alors que les deux premiq6res voient leurs effectifs augmenter progres- sivement d'ann6e en annq6e (respectivement de 63 et 80 individus par M2 en 1978 a 162 et 360 en 1981) q1'abondance de TeqlZina fabuqla qui n1a pas dq6passq6 250 individus par m2 en 1978 et 1979, s'q61q@!ve brusquement 6 plus de 1000 individus par m2 pendant le troisiq&me cycle. 3.2.3.3) Peuplement des vases sableuses 8q@ Abra aZba - Me6qlinna paZmata Pollution Les teneurs en hydrocarbures sont tr4q6s 4q62q1evq6es jusquq'en juillet 1979 (elles dq6passent toujours 100 ppm sauf en f8q6vrier 1979 et elles atteignent m2qCqeme 3000 ppm en mars 1979); ensuite on observe une d8q6polq- lution graduelle. 220 Richesse specifique (fig. 10) D'avril 1978 a avril 1980 la richesse specifique est stable; elle s'accrolt considerablement au cours de 1'ete 1980, diminue ensuite et se maintient durant Phiver suivant a un niveau plus q6qlevq6 que celui des hivers prq6cq6dents. L'augmentation de la richesse spq6cifique provient a la fois de la r6introduction d'espces d'amphipodes q61iminq6es lors du stress et de 0qVaccroissemenqt du nombre qdlespe'ces de polycheiqteqs. --CYCLE NOMMAI - I'CYCLE ----2*CYCLI - - - - - - - - 3'cYCLE N especes )Do A C. 5 0 ......................... ........... .. ............... ...................... ............. X.. .......... ...... ...................... ...... . ................. ................. ........... ....... ....... ......... M IIA I IiI ISI 1". qlil IMIIM' 8qV IS' INI I Jul I'll lil IS"I INI lie I'll I'Al -i ISI IN) 1977 1978 199 1980 Figure 10 Peuplement des vases sableuses e Abra aZba - MeZinna paqZmata q6volution de la richesse spq6cifique des relevq6s (10 prq6lq6vements -q@ la benne Smith Mc Intyre) d'aoqat 1977 q@ mars 1981. Les rq6apparitions d'amphipodes se rq6alisent de maniq6re graduelle: les premiers exemplaires de Cheirocratus intermedius et Am2qpeqlisca brevicornis sont rq6colt6s plus d'un an apr6s leuqt disparition, ceux d'ArnpeZisca tenuicornis seulement au coursu second cycle; les espq6ces A2qpeqlisca armoricana et ArripeqZisca spin-qimana n'ont pas encore q6tq6 retrou- vq6es. En ce qui concerne les polychq6tes on observe conjointement la cap- ture plus frq6quente dq'esp0q6ces sporadiques avant la pollution et lq'intru- sion dq'esp2q6ces constantes, pour la plupart, dans le peuplement des sa- bles fins a Abra a6qZba - Hya6qZinoecia bi6qZineata de la Pierre Noire. Densit2q6s qet biomasses La densitq6 qui West pas modifiq6e lors du stress croqlqlt au cours du premier cycle annuel perturb2q6 : 4467 individus par m2 en moyenne contrqe 2855 durant le cycle normal avant la pollution. Cette diff2q6rence est due essentiellement aux fluctuations d'effectifs dqlespeces pr4q6sentes, avant la pollution, en densit2q6s faibles (16qMediomastuqs fragi6qlis, fig. 12 et Tharyx m6qmq-ioni_, fig. 13) ou fortes (Chaetozone setosaq,, fig.8q11). 221 Cycle normal Cycle>4- 2 Cycle ----30 Cycle- N m2 qC. S 6000- A.C q5q0q0q0- 4000- 3000 2000 A 14 1000 - S N J 5 N J 4qV 4qT 0qV 'N' IS 8qM 4qV 0qW 'S' 2qW IS 8qM 1Vf J S 4qq qIqi, 1977 1978 1979 1980 Figure 11 Peuplement des vases sableuses q@ Abra aqZba - MeqZinna paqtmata q6vlution de la densitq6 totale d'aoqat 1977 q@ mars 1981 avec mise en q4vidence de la part de Chaetozone setosa (qC.s.) (A.C. : dq6but de la pollution par les hydrocarbures de 1q"'Amoco Cadiz"). Au cours du second cycle, la densitq6 redevient voisine de celle du cycle normal, puis augmente A nouveau durant le troisiq6qme cycle : 3724 individus par m2; cette derniq6re valeur s'explique par le haut niveau d'abondance de Mediomastus fragiqlis et Tharyx marioni, par une q61q6vation de la densit6 d'autres especes, de polychq6tes notamment Lanice cnchiqZega et MeqZinna paqZmata et enfin par la recolonisation de Nephtys hombergii et A8qVeZisca tenuicornis. --CYCLE NORMAL - ICYCE ---2'cYCLE - - - - 3*YCLf - N.m-2 50 0 A C 0qZ 2 q50 IMqI Fqj I S qr I T qM qM qM j q5 1q"1 0qIJI qM qM S N 0qJi IMqI IqMI 1q4q& ISI I., I,, 1977 1978 1979 1980 Figure 12 - Peuplement des vases sableuses a Abra a6qZba - Me6qZinna paqtmata q6volution de la densit8q6 de Mediomaqstus fraqgiqZiqs d'aoq6t 1977 8q@ fq4vrier 1981 (A.C. dq6but de la pollution par les hydrocarbures de 6qIq"'Amoco Cadiz"). 222 - - CYCLE NORMAL - - - CYCLE - - - - - - - -2*cYCLE - - - - q----3CYCLE N.m2 300 250- 200- C. 150- 100- 50 M i r., N.J -IAI I MI Ij I I IS I go il IMI SMI Iii [IF INF q1j, Im. am. j. 131 FPO$i-r-I 1117 1970 1979 1980 Figure 13 Peuplement des vases sableuses a Abra aqZba - Meqlinna pqqqZmata : q6volution de la densitq6 de Tharyx maImoni d'aoiit 1977 q@ fq6vrier 1981 (A.C. dq6but de la pollution par les hydrocarbures de I"'Amoco Cadiz"). L'accroissement des biomasses moyennes correspondant aux quatre cycles annuels d'observations est qli6 surtout q& 11installation progres- sive de Lanice conchiqZega dans le peuplement. De 1977 q@L 1980 les valeurs relatives aux observations effectuq6es entre les mois d'aoqat et avril sont respectivement 9.0, 13.0 et 12.4 g. par m2; pour la pq6riode comprise entre aoqat 1980 et mars 1981 la biomasse moyenne atteint 16.8 g. par m2. Les valeurs moyennes de densitq6 (3425 individus par m2) et de bio- masse (12.9 g. par m2) s'accordent avec celles donnq6es par les auteurs travaillant sur des peuplements analogues (RETIERE, 1979). Au terme de 1q'8q6tude des peuplements de sables fins vaseux iq.1 im- porte de souligner que les modalit4q6qs quantitatives des divers ph8q6no- 223 menes qui se succedent dans le peuplement A Abra alba MeqZinna paqZ- mata de la rade de Morlaix diff6rent profondq6ment de celles observq6es dans le peuplement -q@ HyaqZinoecia biqZineata de la Pierre Noire. Dans le ureml*er, qies espq6ces sensibles aux hydrocarbures sont en effet peu reprq6sentq6es avant la pollution, si bien que les mortalit6s initiales sont faibles et n'altq6rent que 1q6gq6rement la structure du peuplement, contrairement q@t la modification structurale considq6rable intervenue a la Pierre Noire. Lq6volution ultq6rieure des deux peuplements q6 moyen et long terme offre un contraste d'une autre nature. La communautq6 subsistante de la Pierre Noire poursuit une dynamique de reconstitu- tion, dans un milieu benthique rapidement dq6contaminq6, et naturelle- ment oligotrophe. Au contraire, le peuplement de la rade de Morlaix vit dans un milieu benthique naturellement eutrophe, plus durablement chargq6 d'hydrocarbures et de mati6res organiques. on assiste alors ,q@ un qdq6veloppement plus important des populations de qd6tritivores, d6- jqa abondants en conditions naturelles (Chaetozone setosa) et aussi q@ de vq6ritables prolifq6rations d'espq6ces r6eqllement opportunistes (6qMedio- mastus fiLifrmis et Tharyx ma0qrioni sp.), pr6sentes seulement a 1q6tat latent au cours du cycle normal pr6cq6dant la pollution. 4) CONCLUSIONS GENERALES Le recul qu'appor tent troiannq6es d'observations perme t de dis tin- guer parmi les ph6nomq6nes qui ont affectq6 les peuplements subtidaux de la r6gion de Roscoff ceux liq6s au stress proprement dit de ceux qui se sont succ6dq6s au cours des cycles annuels suivants et d'en interprq6ter les diff6rentes modalit6s : - 1q61imination des esp6ces lors du stress est s6lective; elle est fonction a la fois de 1'6co-q6thologie des esp&ces et des con- centrations du milieu en hydrocarbures toxiques dissouts. Cette phase de mortalitq6 est relativement limit6e dans le temps (quelques semaines); - l'intensit6 des perturbations dues au stress varie d'une communutq6 a qPautre le long du gradient q6daphique et @ l'intq6rieur de la mq6me unit6 de peuplement. Alors que les peuplements des s6di- ments grossiers et des sables vaseux ont q6t6 peu modif iq6s qualitative- ment et quantitativement par le stress, ceux des sables fins et tr6s fins ont 6t6 intensq6ment perturb6s. Ces divers degrq6s daltq6ration sont fonction du nombre et de qVabondance des espq&ces sensibles aux hydrocarbures; dans le cas extr@qme du peuplement des sables fins @ Abra aq4ba - HyaZinoecia biqZineata on assiste @ une rq6duction respecti- ve de 80 et 50 % des valeurs initiales de densitq6 et de biomasse. Tou- tefois il convient de noter que ce nq'est pas n2q6cessairement sur le peu- plement oq6 le stress a 6t2q6 le plus dq6vastateur que les qeffets secondai- res sont les plus marquq6s; - dans les biotopes o6q6 les effets du stress se sont faits sentir sq6v4q6rement les valeurs de la richesse spq6cifique, de la densit8q6 et de la biomasse restent faibles pendant le premier cycle annuel aprq6s la pollution; on n'enregistre pas, au cours de cette pq6riode, de morta- lit2q6s massives d'adultes mais 24qVabsence de recrutement chez un certain nombre d'esp0q6ces freine considq6rablement la recolonisation du milieu. Dans la plupart des cas, durant cette premiq6re phase, les effets sont 224 proportionnels aux quantites-d'hydrocarbures residuels; - le deuxiq&me cycle annuel marque globalement le dq6but de la r6introduction des espq6ces q61iminq6es et de la recolonisation des fonds par celles dont les populations ont q6tq6 affectq6es par le stress. Ces phq6nomq6nes s'accq6lq6rent et s'accentuent au cours du troisiq6qme cy- cle. La vitesse de rq6introduction et le taux de recolonisation de certaines d'entre elles sont d'ailleurs limitq6s par le caractq6re insu- laire de leur distribution et 2qVabsence de phase pq6lagique; les cycles de densitq6 de la plupart des espq6ces qui n'ont pas ete affectq6es par le stress se dq6roulent 'a peu pres normalement; par contre un petit nombre d'espq6ces regroupant surtout des qCapiteqlqLi- dae et Cirratuqlidae prolifqLrent soit au cours du premier cycle, soit plus tardivement, constituant probablement les amorces d'une succes- sion q6cologique; - trois ans aprqL-s la pollution par les hydrocarbures la ri- chesse spq6cifique du peuplement le plus perturbq6 , c'est-qa-dire celui des sables fins peu envasq6s, a retrouvq6 son niveau initial et bien que sa densitq6 soit toujours beaucoup plus faible qu'elle ne 1'q6tait aupa- vant les valeurs de biomasse sont a nouveau tout q6 fait comparables a celles de 1977. Il semble donc que ce peuplement q6volue vers un "nou- vel q6quilibre"; En outre, de cette q6tude se dq6gagent un certain nornbre d'enseigne- ments : - les cartes bio-sq6dimentaires, support des 6tudes dynami- ques, constituent un q6tat de rq6fq6rence dont 0qVint6rq6t est 6vident dans le cas de pollutions accidentelles du type "Amoco-Cadiz"; - sur 1'ensemble du secteur touchq6 par la pollution les premi&res investigations doivent tre engag6es tr6s rapidment; au sein des communautqes pr6sumq6es les plus sensibles il est indispensable de sq6lectionner plusieurs stations, le suivi de certaines dentre elles pouvant &tre abandonn a la lumiq6re des premiers r6sultats; le suivi q6cologique doit sq6taler sur une pq6riode suffi- samment longue pour quau dela du bruit de fond des fluctuations natu- relles a plus ou moins long terme on puisse percevoir les phq6nom6nes r elqlement dependants de la perturbation. 225 BIBLIOGRAPHIE BESLIER, A., J.L. BIRRIEN, L. CABIOCH, C. LARSONNEUR & L.LE BORGNE, 1980. La pollution des baies de Morlaix et de Lannion par les hydrocarbures. de 1'"Amoco Cadiz" : rq6partition sur les fonds st q6volution. Helgoqlq8nd. wiss. Meeresunters, Vol. 33, pp. 209-224. BESLIER, A., 1981. Les hydrocarbures de qI'Amoco Cadiz dans les sediments sublitto- raux au nord de la Bretagne. Distribution et q6volution. qThq6qse 3q6me cycle Gq6ologi, Universite de Caen, 2D4 pp. CABIOCH, L.1968. Contribution e la connaissance des peuplements benthiques de la Manche occidentale. Cah. Biol. mar., Vol. 9, pp. 488-720. CABIOCH, L., J.C. DAUVIN & F. GENTIL, 1978. Preliminary observations on pollution of the sea bed and disturbance oqf sub-littoral communities in northern Brittany by oil from the Amoco Cadiz. Mar. Pollut. qBulql.,Vol.9, pp.303-307. CABIOCH, L., J.C. DAUVIN, J. MORA BERMUDEZ, C. RODRIGUEZ BABIO, 1980. Effets de la marq6e noire de l'Amoco Cadiz sur le benthos sublittoral du nord de la Bretagne. - Helgolq6nd. wiss. Meeresunters, Vol. 33, pp. 192-208. CABIOCH, L., J.C. OAUVIN, F. GENTIL, C. RETICRE & V. RIVAIN, 1981. Perturbations qinduites dans la composition et le qfonctionnement des peuplemeqhts'ben- thiques sublittoraux sous 1'effet des hydrocarbures de l'Amoco Cadiz. In : Consq6quences d'une pollution accidentelle par les hydrocarbures. Centre National pour l'Exploitation des qOcq6ans, Paris, pp. 513-525. CABIOCH, L., J.C. DAUVIN, C. RETIq@RE, V. RIVAIN & D. ARCHAMBAULT, 1982. Evo lution rq) long terme (1978-1981) de peuplements benthiques des fonds sq6dimqentai- res de la rq6gion de Roscoff, perturbq6s par les hydrocarbures de qIAmoco Cadiz. Neth. J. Sea Research (sous presse). CHASSE, C. & A. GUENOLE-BOUDER, 1981. Comparaison quantitative des populations benthiques des plages-de St Efflam et St Michel-en-Grqbve avant et depuis le naufrage de qI'Amoco Cadiz. In : Conseguences dune pollution acciden- tellepar les hydrocarbures. Centre National pour 1'Exploitation des 0- cq6ans, Paris, pp. 513-524. DAUVIN, J.C., 1979a. Impact des hydrcarbures de l'Amoco Cadiz sur le peuplement infralittoral des sables fins d la Pierre Noire (Baie de Morlaix). J. Rech. oceanogr., Vol. 4 (1), pp. 28-29. DAUVIN, J.C., 1979b. Recherches quantitatives sur le peuplqementdes sables fins- de la Pierre Noire, Baie de Morlaix, et sur sa perturbation par les hy- drocarburqes de 1q"Amoco Cadiz". Thq6se 34q6me cycle 0qOc8q6anographie Biologique, Universitq6 P. & M. Curie, 251 pp. DAUVIN, J.C., 1981. Evolution 4q6 long terme des populations dq'Amphipodes des sa- bles fins de la Pierre Noire (Baie de Morlaix) apr4q6s 6ql'impact des hydro- carbures de lq'Amoco Cadiz. eqJ. Rech. oceanog., Vol. 6 (1), pp. 12q-13. OAUVIN, J.C., 1982. Impact of Amoco Cadiz oil spill on the muddy fine sand Abr2qa a6qlba and Me6qZinna pa6qZmata community from the Bay of Morlaix. Estua. coast. Shelf Science (in press). 226 Den HARTOG, C. & R.E.W.H. JACOBS, 1980. Effects of the Amoco Cadiz oil spill on an eelgrass community at Roscoff (France) with special reference to the mobile benthic fauna. - Helgol8nd. wiss. Meeresunters,Vol.33,PFO.182- 191. ELMGREN, R., S. HANSSON, U. CARSSON & B. SUNDELIN, 1980. Impact of oil on deep soft bottoms. In : J.J. KINEMAN, R. ELMGREN & S. HANSSON. The Tsesis oil spill U.S. Department of Commerce, NOAA, pp. 97-126. ELKAIM, B., 1981. Effets de la mar6e noire de l'Amoco Cadiz sur le peuplement sublittoral de 1'estuaire de la Penz6. In : Cons6quences d'une pollution accidentelle par les hydrocarbures. Centre National pour l'Exploitation des Oc6ans, Paris, pp. 527-537. GENTIL, F. & L. CABIOCH, 1979. Premibres donn6es sur le benthos de l'Aber Wrach (Nord Bretagne) et sur l'impact des hydrocarbures de l'Amoco Cadiz. J. Rech. oc6anogr., Vol. 4 (1), pp. 33-34. GLEMAREC, M. & E. HUSSENOT, 1980. D6finition d'une succession 6cologique en mi- lieu anormalement enrichi en mati6res organiques 6 la suite de la ca- tastrophe de l'Amoco Cadiz. In : Cons6quences d'une pollution acciden- telle oar les hydrocarbures. Centre National pour l'Exploitation des Oc6ans, Paris, pp. 499-512. GLEMAREC, M. & E. HUSSENOT, 1982. Ecological survey for the three years after Amoco Cadiz oil spill in Benoit and Wrac'h Abers. Neth. J. Sea Research (in press). HOLME, N.A., 1953. The biomass of the bottom fauna in the English Channel off Plymouth. J. mar. biol. Ass. U.K., Vol. 32, pp. 1-49. LEE, W.J. & J.A.C. NICOL, 1978. Individual and combined toxicity of some petro- leum aromatics to the marine Amphipod Elasmopus pectenicrus. Mar. Biol., Vol.. 48, pp. 215-222. LEE, W.H., M.F. WELCH & J.A.C. NICOL, 1977. Survival of two species of amphipods in aqueous extracts of petroleum oils. Mar. Pollut. Bull.,vol.8,pp.92-94. LINDEN, 0., 1976. Effects of oil on the amphipod Gamarus oceanicus. Environ. Pollut., vol. 10, pp. 239-250. MARCHAND, M. & M.P. CAPRAIS, 1981. Suivi de la pollution de l'Amoco Cadiz dans 1'eau de mer et les s6diments marins. In : Cons6quences d'une pollution accidentelle par les hydrocarbures. Centre National pour l'Exploitation des Oc6ans, Paris, pp. 23-54. PEARSON, T.H. & R. ROSENBERG, 1978. Macrobenthos succession in relation to orga- nic enrichment and pollution of the marine environment. Oceanogr. mar. Biol. Ann. Rev., Vol. 16, pp. 229-311. RETIERE, C., 1979. Contribution 6 1'6tude des peuplements benthiques du golfe normanno-breton. Th6se doctorat d'Etat, Sci. Nat., Univ. Rennes,,@4370 pp. 227 SANDERS, H.L., J.F. GRASSLE & G.R. HAMPSON, 1972. The West Falmouth oil spill. Woods Hole oceanogr. Instn. Tech. Rep. 1-72-20, 48 pp. SANDERS, H.L., J.F. GRASSLE, G.R. HAMPSON, L.S. MORSE, S. GARNER-PRICE & C. JONES, 1980. Anatomy of an oil spill : long-term effects from the grounding of the barge Florida off West Falmouth, Massachusetts. J. mar. Res., Vol. 38, 265-380. 228 ETUDE EXPERIMENTALE D'UNE POLLUTION PAR HYDROCARBURES DANS UN MICROECOSYSTEME SEDIMENTAIRE. I : qEFFET DE LA CONTAMINATION DU SEDIMENT SUR LA MEIOFAUNE par Boucher G., Chamroux S., Le Borgne L. et Mevel G. Station Biologique de Roscoff 29211* (FRANCE) RqIqESUME Les consq6quences de deux niveaux de contamination par hydrocarbu- res ont q6tq6 analysq6es,par rapport q@ un tq6moin.dans des microecosystq@- mes expq6rqimentaux en circuit clos contenant 100 litres de sable fin subqlittoral. 1,1q6volution des caractq6ristiques du peuplement de qmq6io- faune (Nq6qmatodes et Copq6podes) a q6tq6 choisie pour caractq6riser Vim- pact des hydrocarbures. Les densitq6s des Nq6matodes augmentent sensi- blement par rapport au tq6moin pendant les deux premiers mois de la pollution puis regressent lentement sans qu'il soit possible de dis- tinguer les effets d'une forte pollution de ceux d1une faible pollu- tion. Les densitq6s des Cope"podes qharpacticoqldes sont d'autant plus faibles que le sq6diment est plus contaminq4. Le rapport Nq6qmatodes/Co- pq6podes paraqlt qatre un indice significatif dq6 degrq6 de pollution. La composition faunistique des Nq6matodes est profondq6ment modi- fiq6e dans le module le plus poqlluq6 aprq6qi 3 mois d'expq4rience. Cette dq6gradation se manifeste par une chute brutale de la biomasse et de la diversitq6 spq6cifique. Des petites espq6ces opportunistes connues pour leur association avec 6qVenrichissement en matiq@re organique, de- viennent dominantes. Le module faiblement qpoqlluq6 ne prq6sente aucune dq6gradation interprq6table$par rapport au tq6moin,des param@tres du peuplement. -------------------------------------------------------------------- * Ce travail a 4q4t4q6 r4q6a6qlis8q6 avec 1q1aide d'un contrat n' 80/6189 pass4q6 entre : le Centre National de la Recherche Scientifique, le Centre National pour 1q'Exploitation des 6qOc8q6ans et la National Oceanographic and Atmospheric Agency (USA). Il a fait 4ql'objet dq'une prq4sentation au S2q6minaire Amoco CADIZ CNEXO-NOAA q: q"Bi2qZan des 6q6tudes bio6qlogiques de qla 28qPqOqI6qZution de 2qZqrAmoco Cadiz" organis2q6 au Centre qOc8q6anologique de Bretagne Brest (France) les 28 et 29 octobre 1981. 229 INTRODUCTION Les consequences des pollutions sur 6qVenvironnement sont souvent difficilement interprq6tables car leur dilution dans un milieu complexe peut provoquer des aqltq6rations 11 plus ou moins discernables des fluctua- tions naturelles, apparaissant immq4diatement aprq6s contamination ou diffq6rq4es dans le temps. L'expq6rimentation en laboratoire de la toxicitq6 des polluants sur des',organismes isol4s de leur environnement naturel a montre sou- vent ses limites car elle ne prend pas en compte les effets cumulatifs. Par contre, les simulations sur des micro4cosystq@mes complexes appelq6s microcosmes ou mesocosmes selon leur taille, permettent de mieux cer- ner les consq4quences des perturbations des q6cosystq6mes et souvent de compl6ter les observations rq4alisq6es dans le milieu naturel. 6qVutili-@- sation des microcosmes permet en outre d'intq4grer les interactions entre les niveaux trophiques d'organisation des q4cosyst@mes par exem- ple et de rq6aliser des manipulations et des replications. Quelques rares simulations au laboratoire de contaminations par hydrocarbures ont jusqu'ici q6tq6 rq6alisq6es soit pour envisager les vi- tesses de dq6gradation des hydrocarbures en milieu sq6dimentaire com- plexe (Johnston, 1970; Delaune et coll. 1980; Wade & Quinn 1980) soit pour comprendre les effets sur les organismes dans les diffq6rents ni- veaux d'organisation de 1'q6cosystq@qme (Lacaze 1979; Elmgren et coll. 1980; Grassle et coll. 1981; Elmgren & Frithsen, sous presse). A la suite de la pollution pq6troliq6re de lq"'Amoco Cadiz" sur les cq8tes de Bretagne Nord (Manche occidentale), nous nous somqmes attach6s parallq6lement @ 1'q6tude in situ des consq6quences de la contamination des sables fins sublittoraux (Boucher 1980 et 1981; Boucher, Chamroux et Riaux 1981), q@ rq6aliser une simulation du ph6nomq6ne dans des micro- q6cosystq@qmes en circuit clos. MATERIEL ET qMETHODES Trois modules exp4rimentaux en circuit clos dont le principe a q4t4 fourni dans Boucher et Chanroux (1976) ou M6vel (1979) sont uti- lisq4s (Figure n 1). Chaque bac comporte trois compartiments dont les niveaux sont rq4gul6s par contacteur q6lectrique (500 litres d'eau de mer du large). Le compartiment principal comporte 100 litres de sable rq6parti sur un double fond sur une surface de 0,41 ml et une hauteur de 25 cm environ,et perco12q4 par diff2q6rqence de niveau entre les compar- timents a une vitesse de filtration de 28qVordre de 15 I/M2/heure. Le sq6diment est pr2q6lev2q6 q@ la bennqe Smith-McIntyre en milieu sublittoral par - 19 mq6tres de profondeur (Station de la Pierre Noire). Sa 8qmq6dia- ne est de 136 � 5 4qp. La tempq4rqature pendant le durq6e de 1q'exp8q6rience a variq6 entre 10 et 15'C. Les 4qtrois mesocosmes ont q4t2q4 nourris tous les deux jours par des casaminoacides de DIFCO q@ des doses correspon- dant 8q@t 50 g d'Azote/an/M2. Lq'un des problq;q_@4qmes importants A r2q4soudre q6tait le mode dq'intro- duction des hydrocarbures dans les modules exp2q6rimentaux. Respective- meqnt 100, 10 et 0 g. dq'hydrocarbures Arabian lightqsq6t4q6t4q6s a 2400Cj 238q0 - PumP2 Urn cooler IF Water, 5 0 0 q1. amplifier pHq_rH an ... loo recorder pHq_rH 0,66XqO,98M=0,41 m 2 FIGURE 1. Schq6ma d'un bac expq6rimental ou mesocosme utilisq6 pour les simulations de pollution par hydrocarbures (Volume de sable 100 litres, Volume d'eau : 500 litres). (fournis par 1'qIFP) ont q6tq6 mq6langq6s a 1 kg de sable sec et homogq6nq6i- sq6s avec 100 ml de tetrachlorure de Carbone. Aprq6s q6vaporation totale, le sq6diment ainsi traitq6 a q6t6 ajoutq6 respectivement dans hacun des trois modules appelq6s : Bac fortement polluq6, Bac faiblemet polqluq6 et T6moin. Le coulage des hydrocarbures a 6tq6 ainsi quasi imm6diat et 6qVessentiel des particules contamin6es sest rq6parti a la surface du substrat. Une faible fraction est rest6e q@ la surface de leau conte- nue dans le compartiment q@ sable du bac le plus poqllu6 pendant quel- ques jours. Les pr6lq;@vements dans chacun des mesocosmes ont 6tq6 r6alis6s a 6qVaide de tubes de carottages en plexiglass de 5,72 cm'. Trois prises simultan6es ont q6t6 effectu6es pour les hydrocarbures et les compta- ges de mq6iofaune avec une frq6quence hebdomadaire puis mensuelle, pen- dant plus de six mois du 17 mars 1981 au 29 septembre 1981. Chaque carottage a 6tq6 fractionn6 en trois niveaux : 0-4; 4-8; 812 centimq6- tres pour analyse de'la r6partition verticale des paramq6tres. Les hydrocarbures ont q6t8q6 extraits au tetrachlorure de Carbone partir de 10 grammes de sq6diment s2q6chq6 8q@ 60*C 2q@ 1'q6tuve. Ap0qrqis pas- sage sur Fluorisil, destin2q6 a q61imqiner les fractions oxydq6es et les hydrocarbures endog2q6nes, 1'extrait a 2q6tq6 lu au spectrophotom6q6tre in- frarouge UNICAM SP 1100, q@ une longueur dq'onde comprise entre 2500 et 3000 cm-1, avec calage du pic caractq6ristique 6q@ 2925 cm-1. Les r2q6sul- tat2qs ont q6t8q6 exprimq6s en ppm (mg/kg sable PS) dq'apr6q;_@s I'abacle rq6ali- s2q6e sur l'Arabian light IFP. La filtration sur Fluorisil provoque une rq6tentiqon sur le fil- tre de 44qVordre de 60.4% du poids du produit d'origine. 231 Les organismes de la qmeiofaune ont 6t6 fix6s au formol 4%, colo- rq6s au rose bengal et triq6s aprq@s passage sur tamis de 40 qp et q6qlutria- tion. Des lots de 100 N6matodes ont e'tq6 mesurq6s et identifiq6s pour la dq6termination de la biomassqe et de la composition sp6cifique. RESULTATS Evolution des hydrocarqbures Les quantites d'hydrocarbures introduites dans les modules fai- blement et fortement polluq6s correspondent 'a des teneurs initiales thq6oriques de 223 et 2236 ppm puisque seuls les quatre premiers centi- mq6tres du sq6diment (soit 17.7 kg) sont contamings et que le passage de l1extrait tetrachlorure sur Fluorisil entra-ne une perte de 60.4% au dosage. En effet, ql'analyse de la rq6partition verqticale des hydrocarbu- res dans la colonne sq6dimentaire en utilisant ce type de dispositif expq6rimental rq6ve'le une pq6n6tration quasiment nulle du polluant sous la surface. Seule la tranche 0-4 centimq6tres cntient des hydrocarbu- res en quantitq6 notable et une pq6nq6tration limitq6e dans la tranche 4-8 centimq6tres napparaqlt q6pisodiquement qu'@ partir du 106q6me jour. Le tq6moin n'a jamais montr6 la moindre trace d'hydrocarbures. La figure n' 2 fournit I'6volution des teneurs au cours du temps dans les bacs faiblement et moyennement poqlluq6s L'q6volution des teneurs dans le module faqiblement poqlluq6 indique une dq6gradation extrq9mement faible au cours des six mois dexpq6rience avec une h6tq6rog6nq6it6 des teneurs mesurq6es trq6s 1q6gq6rement plus accen- tu6e en dq6but d'expq6rience. Par contre, 1'q6volution des teneurs dans le module fortement polluq6 met en q6vidence une trq6s forte hq6t6rogq6ne'it6 des concentrations jusqu'au 23q6me jour de prq6levement qui reflete la r6partition en aggrq6- gats des hydrocarbures q@ la surface du substrat ainsi qu'il a tq6 pos- sible de l'observer en plongq6e in situ sur les sables dorigine de la Pierre Noire. Cette hq6tq4rogq6n4itq6 tend q@ se rq6duire ensuite consid6ra- blement du fait de la bioturbation. Le pic d'abondance des hydrocarqbu- res observq6 au 23q6me jour reste compatible avec les incertitudes de qVintervalle de confiance q@ la moyenne. Il est liq& probablement aussi au dq6lai nq6cessaire au coulage de toutes les particules mazout6es ayant partiellement tendance q@ flotter q@ la surface de 2qVeau du compartiment principal en dq6but dq'exp8q6rience. La disparition des hydrocarbures entre 23 et 93 jours semble sa- tisfaisa4qnte puisquq'elle iqndique une 8q6volution de 2979 � 1672 pp8qm a 248 � 52 ppm soit 38.5 mg HC/kg sable/jour (24 mg HC/kg sable/jour si Von tient'compte dq'une valeur initiale th2q6orique de 2236 ppm. Il est cepen- dant impossible de considq6rer cette valeur comme un taux de dq6grada- tion r6q6aliste du fait de 8qlq'h2q6tq6rog0q6n6q6it6q6 des teneurs initiales mesur2q6es mais aussi d'une remont4q6e incompr4q6hensible des teneurs en hydrocarbu- res au 1060q6me et 124e'me jour. 232 00 1500 1000- V V q3 500- I to 48 so It IGO 120 140 100 180 Days FIGURE 2. Evolution des teneurs en hydrocarbures et de leur q6cart la moyenne, dosq6es par la qmq6thode infra-rouge aprq6s passage sur Fluorisil,- dans le sable des modules fortement poqlluq6s (100 g 6qHC : * - e) et faiblement poqlluq6s (10 g HC : o ----o). Les hydrocarbures sont presque toujours concentrq6s dans les quatre premiers centimq@tres du sq4diment. La difficultq6 d'interprq6tation des rq6sultats des dosages effec- tuq6s par infrarouge aprqi_@s passage de 2qVextrait au CC14 sur Fluorisil montre donc l1extrqame hq6tq6rogq6nq4itq6 de la rq6partition des hydrocarbu- res dans le sq6diment, mq;q&me q@ 1'q4chelle de quelqques dizaines de centi- m6tres. Il apparaqlt difficile de r4aliser des calculs de biodq4grada- tion dans des microcosmes oqtL Von ne pr4lq6ve qu'un faible aliquot du volume de sable contaminq4. Evolution des densitq6s de la Mq6iofaune Les densit6s initiales du Mq6iobenthos dans le mesocosme tq6moin et dans ceux contamin6s par les hydrocarbures q4taient comparables (non significativement diff6rentes au niveau 5% par le test de Kruskall Wal- lis). Les valeurs initiales trouv2q6es de 1218 � 13 Nq4matodes/10 CM2 et de 198 � 26 Cop8q6podes harpacticoides/10 cm:2, deux groupes qui consti- tuent la quasi totalit6q6 des organismes 4qm6q6iobenthiques recensq6s, peuvent 8q@_tre favorablement compar2q4es avec la densit8q4 relev2q6e dans le milieu na- turel 2q@ la meme date (1357 Nq6matodes et 105 Cop2q6podes/10 CMq2q).dans les douze premiers centim2q;q_@tres du s8q6diment en mars 1981. qV Bien que 40qVa2qnalyse de la rq6partition verticale montre qu.q'environ 70% des n8q6matodes et 44% des cop6q6podes sont concentr8q6s dans les quatre premiers centimq6tres du s2q6diment, les imp8q4ratifs de tempsde tri ont conduit 2q@ estimer les densitq6s seulement dans les quatre premiers cen- tim2q6tres. 233 M A M A qS I P left- n 500 q- 0 q20 40 60 80 10 120 140 150 180 Day FIGURE 3. Evolution des densitq4s (et de leur 6cart @ la.moyenne) des Nematodes dans les quatre premiers centim6tres du.sq6diment pendant une pq4riode de six mois.q4moin : 0 -- 0; Module faiblement poqluq6 : o -.- o; Module fortement pollu6 Quel que soit le module considq6rq6, les densit4s de n6matodes ont tendance A dq6croqltre en circuit clos (Fig. 3). Cependant, il est pos- sible de distinguer : - une pq4riode initiale de deux mois environ oq@ les valeurs trouvq6es dans les bacs poqllu6s sont gq4nq4ralement plus for-. tes que dans le t6moin; - une pq6riode ult6rieure o8q@qL les densit6s dans ces modules contamin2q6s ont tendance A 6qatre 1q6g6q6rement plus faibles par rapport au tq6moin. I8ql appara6qlt donc que la phase de pollution primaire serait ca- ract8q6risq6e par une prolif2q6ration des n6matodes (1,5 a 2 fois le niveau du t2q6moin) mais que rapidement appara8qltrait un dq6clin lent (0,5 a 0,8 fois la valeur du t6moin dans le bac le plus po4qllu2q6, 0,7 fois A une valeur comparable au t2q6moin dans le bac faiblement po4qllu6qQ. Ces obser- vations sont compatibles avec celles releves dans le milieu naturel (Elmgren 1980 a; Boucher et al. 1981). 234 L'evolution des Copepodes harpacticoldes (Fig. 4) montre au contraire un effet dq4pressif des hydrocarbures sur le niveau de den- sitq4 du peuplement. Les abondances observq6es dans le tq4moin restent toujours supq4rieures q@ celles des bacs poqlluq6s malgrq4 des fluctua- tions importantes des densitq6s au cours de 1'experience. Dans le bac le plus poqlluq6, les denqsitq4s de Copq6podes deviennent faibles aprq;_@s le 21q@mejour. 4qM A 4qM qi A 8qS ,:z 20C - 100- t 1 20 40 to $0 100 12 f40 is$ 100as FIGURE 4. Evolution des densitq4s (et de leur q6cart q@ la moyenne) des Copq6podes harpacticoqldes dans les quatre premiers centimq;_@- tres du sq6diment pendant une pq4riode de six mois. Tq6moin o ---- o; Module faiblement poqlluq6 : 0 -.- oq; Module forte- ment polluq6 : 0 - 0. Evolution du rapport Nq6matodes/Copq6podes Lutilisation de ce rapport dans les q6tudes de pollution vient I rq6cemment d'q@tre proposee par Raffaelli et al (1981). Cette proposi- tion sq6duisante d6rive de deux considq4rations : - dq'une part les Cop4q6podes sqont apparemment plus sensibles que les N6q6matodes au stress des pollutions; - d'autre part lq'utilisation de la m2q6iofaune dans les q6tudes dq'impact ne se justifie que si celleq-ci rq4pond avant que les effets ne deviennent visibles sur la macrofaune. Lq'un des obstacles majeurs 1q1interprq6tation de cet indice r2q6side dans le fait que celui-ci est corre1q6 n2q6gativement avec la 4qmq6diane granu6qlomq6trique du s8q6diment. L'ex- p2q6rimentation permet dq'q61iminer ce facteur qui obscurcit ilinterpr2q6ta- tion des rq6sultats. 235 A M 40 30 20 C 10 L 0 20 40 60 80 100 120 140 160 180 Days FIGURE 5. Evolution du rapport Nq6matodes/Copq6podes dans les quatre premiers centimq6tres du sq6diment pendant les six mois d'ex- pq6rience. Tq6moin : 0 --- 0; Module faiblement poqlluq4 : o---o; Module fortement poqlluq6 : e - e. La figure 5 montre 1'q6volution de ce rapport dans les trois mo- dules expq6rimentaux. Les valeurs sont d'autant plus fortes que le bac est plus pollu2qg par les hydrocarbures. Ainsi le t2q6moin pr2q6sente tou- jours des valeurs faibles comprises entre 1.07 et 12.40 (moyenne 5.04 �0.50 par 33 mesures en six mois). Le bac faiblement po4qllu2q6 prq6sente des va4qleurs 12q6g6q6rement plus fortes et plus variables comprises entre 1.64 et 20.07 (moyenne : 7.81 0.78 par 32 mesures). Enfin le module fortement poql6qlu4q6 pr4q4sente des valeurs nettement plus 2q64qlev2q6es mais fluctuantes. Deux pq6riodes corres- pondant a des fortes valeurs peuvent 2qZtre distingu2q6es entre 22 et 50 jourqs (N2qIC = 37 4q@ 51) et entre 124 et 188 jours (N/C 16 4q@qL 71) s4q6pa- r6q4es par une p2q6riode de faible valeur 2q@ 71 jourqs (N/C 5 2q@ 14). 236 La sensibilite de cet indice semble donc se confirmer puisqu'un accroissement sensible est discernable aprq@s 25 jours d'expq6rience dans le bac fortement poqlluq6 du fait de la quasi disparition des Copq6- podes, alliq6e @ une prolifq6ration des Nq4matodes. La gq6nq6ralisation de son utilisation demande cependant de prq6ciser les modalit6s de ces fluctuations dans diffq6rentes conditions expq6rimentales en tenant compte des changements de la composition spq6cifique aussi bien des Nq6matodes que des Copq6podes et de leur niveau de compq6tition pour la nourriture par exemple (Warwick 1981). Il est probable que le retour de ce rapport q@ des valeurs faibles dans le bac le plus poqlluq6 corres- pond @ un changement de la composition faunistique des Copepodes. Evolution des paramqi-0qAres Abondance, Biomasse et Diversit6. La plupart des q6tudes utilisant les modifications de la macro- faune pour mettre en q6vidence 2qViqmpact des pollutions pr6conisent 2qVemploi de paramq@tres simples tels le nombre d'espqkes, 6qVabondance et la biomasse (Pearson et al. 1978; Gl6marec et al.1981 entre autres). Cette approche pose pour 8qVinstant de sq6rieux problq@!mes m6thodoqlogi- ques en ce qui concerne la Mq6iofaune. Les progrq6s rq4alisq6s dans la systq6matique des groupes dominants des Nq6matodes et des Copq6podes permettent d'envisager raisonnablement la dq6termination de routine du nombre 6qVespq@ces dans un q6chantillon reprq6sentatif de la popula- tion (100 A 300 individus) malgr6 la grande diversitq6 de ces groupes. Il n'en est pas de m-eme en ce qui concerne 1'q6volution de la biomasse du mq6iobenthos dans un 4chantillon donn.q6. En effet, les mq6thodes ac- tuellement utilisq6es posent gq6n6ralement 6qVhypoth@se que la biomasse moyenne ne varie pas au cours du temps, ce qui n'est bien sur pas le cas. Elles consistent par consq6quent soit q@ q6valur les biovolumes q@ la chambre claire d'un microscope entiliant des formules dq4quiva- lence (Andrassy 1956; Wieser 1960; Juario 1975), soit peser q@ la microbalance de pr6cision un unique lot de quelques centaines a quel- ques milliers d'individus (De Bovq4e 1981; Guille et al. 1968). Les rq6sultats pr6sentq6s dans cet article doivent q!q@tre considq6- r6s comme la mise au point dune qmq6thode d'6valuation des indices vo- lumiques sur la'qmq6-iofaune qui permet d'analyser rapidemnt chaque prq6- 1q6vement et 6vite les incertitudes des qmq6thodes de pes6e. Chacun des lots de 100 specimens montq6s entre lame et lamelle pour la d6termina- tion a q6tqd grossi cent fos grace a un projecteur de profil. Les con- tours de chaque individu identifiq6 ont q6tq6 trace's puis analys6s q@ la table digitalisante d'un analyseur d'images. Le volume a 6tq6 calcul6 et exprimq6 en poids sec en utilisant une valeur de densitq6 de 1,13 (Wieser 1960) et un rapport Poids sec/Poids frais = 0,25. Les autres param2q6tres classiquement utilisq6s tels que nombre dq'esp2q@ces (S), Indice de diversitq6 de Fisher et al. (a), Indice de diversit8q6 de Shannon (H) et Equitqabilitq6 de Pielou 52qM ont 8q6tq4 aussi calcul8q6s au temps z8q6ro, 36 jours (avant la chute de densit6qO et 93 jours (apr4q@_-s la chute de densit60qO de 1'exp6q6rience dans chacun des trois modules sur des q6chantillons de 100 individus (Tableau I). Dans le bac t8q6moin, 1'8q6volution du poids sec individuel n'in- dique pas de fluctuations significatives puisque les limites des in- tervalles de confiance de la moyeqrine se recoupent (0.088 2q@ 0.207 jig PS/individu). 237 -------------------------------------------------------------------- ------------------------------ Modules !Indices TO 36 1 93 ----------- --------------- ---------------------- 4--------------------------- N 633 653 1 485 PS 0.125 � 0.015 (97) 0.150 � 0.057 (97) 0.101 � 0.012 (101) T6moin B 79.3 98.1 49.1 S 29 33 33 13.71 17.19 17.19 H 3.97 4.62 4.46 j 0.82 0.92 0.88 ------------ L-------- L------------------------- t ---------------------- IT --------------------------- N 1007 11 831 365 PS 0.130 �0.028 (100) 0.071 �0.0009 (90) 0.103 � 0.035 (101) B 131.2 59.1 37.7 Pollution S 29 34 34 faible ot 13.71 18.15 18.15 U-) H 4.46 4.35 4.15 00 0.92 0.86 0.82 ------------------------- ----------------------- --------------------------- r------------ N 11 1016 955 418 PS 0.107 � 0.015 (101) 0.125 � 0.029 (95) 0.019 0.003 (90) Pollution B 108.9 119.6 11 7.8 S 23 28 17 forte a 9.35 12.91 5.88 H 3.71 4.05 2.19 0.82 0.84 0.54 r------------- ------------------------------------------------------------------------------------ TABLEAU I. Evolution des densit6s (N), du poids sec moyen (PS), de la biomasse (B), du nombre d'esp@ces pour 100 individus (S) et des Indices de diversit6 du peuplement de N6ma- todes au cours du temps (z6ro, 36 et 93jours). ot = Indice de Fisher et al.; H = In- dice de Shannon; J = Equitabilit6 de Pielou. La biomasse est sensiblement plus forte au temps zero et sur- tout q@ 36 jours qu'.q@ 93 jours essentiellement du fait de densitq6s plus fortes. Le nombre despq6ces identifi6es est assez constant (29 q@ 33) ainsi que les divers indices de diversitq4. Dans le bac faiblement poqlluq6, le poids sec individuel aprq;_s 93 jours n'est pas significativement diffq6rent de celui de 1'q6tat initial. La dq6croissance de la biomasse 131,2 a 37,7 qpg PS) est surtout qliq6e q@ la chute des densit6s (1007 a 365). Le no6qmbre despq@ces recens6 a ten- dance q@ l4gq@rement augmenter (29 @L 34) ce qui provoque une augmenta- tion parall6le de qVindice de Fisher et al. (13,,71 q@ 18,15). La dimi- nution lente de 6qVindice de Shannon et de 1q6qquitabilit6 indique qlap- parition d' une hiq6rarchisation plus marqu6e au cours du temps. Daqns le bac fortement polluq6, 1q6volution des densit6s est si- milaire q@ celle du module faiblement poqlluq6. Apr6s 36 jours, les va- leurs des paramq@tres demeurent comparables h celles de 1q6tat initial. Aprq@s 93 jours, par contre, le poids sec moyen chute fortement (0.019 � 0.003 Pg PS) d'o@j une r6duction brutale de la biomasse (7.8 pg/10 CM2). Celle-ci est liq6e au remplacement du peuplement dorigine par quel- ques esp@qkes de petite taille caractristiques des milieux riches en matiq6re organique (Leptclaiqm tripaqvilqlatus Boucher 1977, qMonhqystera afqf. disjuncta Bastian i865; 4qMonhytera pui4qUa Boucher 1977) et mises en q4vidence dans des expriences pr6alables deutrophisation (Boucher 1979). DISCUSSION Ces simulations de pollutions en microq6cosy st6nes q6dimentai- res soulignent la difficult6 d'interprq6ter les ph6nom;qLes de dq6grada- tion des hydrocarqbures dans les s6diments. La disparition de ceux-ci niest pas d6tectable pendant la durq6e de 1'exp6rience dans le module faiblement poqllu6; elle est anarchique dans le bac fortement contami- nq6. Il ne semble doc pas possible de caract6riser aisq6ment une pol- lution par le niveau du polluant dans le milieu avec la mq6thode em- ploy6e. L'utilisation d'organismes sensibles au polluant (indicateurs biologiques) int6grant 4qVensemble des consequences du stress paraqlt plus fiabl pour caractq6riser un impact. Elle suppose, pour q@tre ef- ficace, que ceux-ci r6pondent avant que la perturbation devienne q6vi- dente. Si certains organismes de la macrofaue benthique rq6pondent de fagon nette au stress primaire de la pollution par hydrocarqbures (Dau- vin 1979 a et b et 1981) la dur2q6e des cycles (1 a 10 ans) rend problq6q- matique 6qVanalyse des effets diff2q6rq6s (Chass8q6 et al. 1981). Du fait de la rapiditq6 de reproduction, la m4q6iofaune, dont les N8q6matodes et les Cop2q6podes constituent 32qVessentiel des organismes, est I un matq4riel prometteur pour comprendre les mq6canismes regissant la destructuration et la restructuration dq'un 4q6cosystqiq_@me- Ces exp8q6riences de contamination brutale par hydrocarbures ne sugg2q@rent pas un effet tr6q@s important sur le niveau des densit2q6s des Nq6matodes. Leur augmentation entre 21 et 50 jours ne semble pas li6e 239 a un d6veloppement d'OpDortunistes n6crophages comme le sugg6re Chass6 (1978) puisque la compo;ition faunistique reste tr6s comparable a celle du t6moin. La chute des densit6s observ6e ult6rieurement dans le bac le plus pollu6 est conforme aux r6sultats obtenus exp6rimentalement in situ par Bakke et al. (1980) ou en mesocosmes par Elmgren et al. 0980 b). Elle s'accompagne d'un changement tr.@_,s perceptible de la com- position faunistique avec r6duction du nombre d'espi@ces et diminution de la biomasse. Contrairement a Vid6e ge'n6ralement admise, le groupe des N6ma- todes peut donc constituer un indicateur biologique fiable des modifi- cations de 1'6cosyst@me (Platt & Warwick, 1980) puisque leurs possibili- t6s adaptatives permettent @ certaines especes de se maintenir quelles que soient les conditions de milieu, @L d'autres de prolif6rer en quel- ques semaines pour occuper la niche laiss6e vide. La d6termination ex- p6rimentale de groupes de N4matodes a comportement similaire vis-@-vis de 1'eutrophisation ou des pollutions, apparalt donc comme une voie de recherche prometteuse pour caract6riser 1'4tat ou la dynamique des 6co- syst;_@mes perturb6s. SUMMARY Experimental study of hydrocarbon pollution in a sand microecosystem I. Effect of.the sediment contamination on meiofauna. The effects of hydrocarbon pollution, at two different intensi- ties with respect to a control, on the microecosystems were studied using recirculating experimental tanks containing 100 liters of subti- dal fine sand. Changes in the population characteristics of meiofauna (nematodes and copepods) were chosen to follow the effects of oil pol- lution. Irrespective of the intensity of pollution, the density of 'ne- matodes in the experimental tanks increased at a significantly higher rate than in the control tank during the first two months after pollu- tion and then decreased slowly. The density of harpacticoid copepods was negatively related to the intensity of oil pollution. It appears that the nematodes/copepods ratio would be an useful indicator of the degree of oil pollution. After 3 months of experimental duration, the faunal composition of the nematodes in the highly polluted tank was drastically modified. This change is evident from a sharp fall in biomass and species diver- sity; small opportunistic nematode species, known for their associa- tion with eutrophicated environment, became dominant. Changes in the meiofauna population parameters in the slightly polluted experimental tank did not show any significant variation from those in the control tank. REMERCIEMENTS Vensemble des tris de la m6iofaune a 6t6 r6alis6 par Melle L. Cras, technicienne CNRS que nous tenons plus particuli@rement @ remer- cier avec toutes les personnes ayant collabor6 i ce travail. 240 BIBLIOGRAPHIE Andrassy, 1., 1956, Die Rauminhalts-und gewichtstimmung der Fadenuiir- mer (Nematoden). Acta zool. Acad. Sci. hung. Vol. 2, pp. 1-15. Bakke, T. I T.M. Johnsen, 1979, Response of a subtidal sediment commu- nity to low levels of oil hydrocarbons in a Norvegian fjord. In Proc. 1979 oil spill Conf., Los Angeles, CA (USA). Publ. Amer. Petrol. Inst. Washington, D.C., pp. 635-639. Boucher, G., 1979, Evolution des caract6ristiques chimiques et biologi- ques des s6diments en circuit clos. II. Effet de la matU_-re orga- nique circulante sur la m6iofaune de syst;_@mes polytrophes. Collo- que national Ecotron. Publ. Sc. Tech. CNEXO, Actes Colloq, pp. 31- 47. Boucher, G. 1980, Impact of Amoco Cadiz oil spill on intertidal and sub- littoral meiofauna. Mar. Pollut. Bull., Vol. 11 (4), pp. 95-100. Boucher, G., 1981, Effets @ long terme des hydrocarbures de l'Amoco Cadiz sur la structure des communaut6s de N6matodes libres des sables fins sublittoraux. In "Amoco Cadiz. Cons6quences d'une pollution accidentelle par7les hydrocarbures". Actes Colloq.CNEXO pp. 539-549. Boucher G. & S. Chamroux, 1976, Bacteria and meiofauna in an experimen- tal sand ecosystem. I. Material and preliminary results. J. exp. mar. Biol. Ecol., Vol. 24, pp. 237-249. Boucher, G. S. Chamroux & C. Riaux, 1981, Etude d'impact 6cologique de la pollution p6troli@re de l'Amoco Cadiz dans, la r6gion de Ros- coff et de la Baie de Morlaix. Effet @ long terme sur la structu- re des 6cosyst@_-mes s6dimentaires. Rapport d'ex6cution de contrat d'6-tudes environnement CNEXO/Universit4 Paris 6 n' 79/5973, 51 pp. Chass6l C.,- 1978, The ecological impact on an near shores by the Amoco- Cadiz oil spill. Mar. Pollut. Bull., Vol. 9 (11), 298-301. Chass6, C. & A. Cu6nol6i-Bouder, 1981, Comparaison quantitative des popu- lations benthiques des plages de St Efflam et 'St Michel en Gr@ve avant et depuis le naufrage de l'Amoco Cadiz. In "Amoco Cadiz. Cons6quences d'une Dollution accidentelle par les hydrocarbures". Actes Colloq. CNEX0, pp. 513-525. Dauvin, J.C., 1979 a, Impact des hydrocarbures de 1"'Amoco Cadiz" sur le peuplement infralittoral des sables fins de la Pierre Noire (Baie de Morlaix). J. Rech. Oceanogr., Vol. 4 (1), pp. 28-29. Dauvin,, J.C., 1979 b, Recherches quantitatives sur le peuplement des sables fins de la Pierre Noire, Baie de Morlaix,, et sur sa per- turbation par les hydrocarbures de l'Amoco Cadiz. Th@se Doctorat 36 cycle Univ. Pierre et Marie Curie Paris 6, 251 pp. Dauvin, J.C., 1981, Evolution @ long terme des populations d'Amphipodes des sables fins de la Pierre Noire (Baie de Morlaix) apr6s l'im- pact des hydrocarbures de l'Amoco Cadiz. J. Rech. oceanogr. Vol. 6 (1), pp. 12-13. 241 Delaune, R.D., G.A. Hambrick & W.H. Patrick, 1980, Degradation of hy- drocarbons in oxidized and reduced sediments. Mar. Pollut. Bull. Vol. 11, pp. 103-106. De Bov6e, F., 1981, Ecologie et dynamique des N6matodes d'une vase sublittorale (Banyls-sur-Mer). Th6se Doc. Etat Sci. Nat. Univ. Pierre et Marie Curie Paris 6, 194 pp. Elmgren, R., S. Hansson, U. Larsson & B. Sundelin, 1980 a, Impact of oil on deep soft bottoms.In "The Tsesis oil spill. ed. by J. Kineman, R. Elmgren & S. ffansson, U.S. Dept. of Commerce NOAA outer Continental Shelf Environmental Assessment Program, pp- 97-126. Elmgren, R., G.A. Vargo, J.F. Grassle, J.P. Grassle, D.R. Heinle, G. Langlois & S.L. Vargo, 1980 b, Trophic Interactions in experi- mental marine ecosystems perturbed by oil. In Microcosms in e- cological Research ed. by J.0. Giesy Jr. Publ. by U.S. Techni- cal Information Center, U.S.$ Dept Energy, Symposium Series 52 (Conf. 781101) pp. 779-800. Elmgren, R. & J.B. Frithsen (in press), The use of experimental eco- systems for evaluating the environmental impact of polluants : a comparison of an oil spill in the Baltic sea and two long term low level oil addition experiments in microcosms. In Reeve, M. & G. Grice (eds), Proc. Symp. Enclosed Experimental-Ecosystems, Sidney, B.C. 12-16 Aug. 1980. Glemarec M. & C. Hily, 1981, Perturbations apport6es @i la macrofaune benthique de la Baie de Concarneau par les effluents urbains et portuaires. Acta Oecologia/Oecologia Applicata, Vol. 2 (2), pp. 139-150. Grassle, J.F., R. Elmgren & J.P. Grassle, 1981, Response of benthic communities in MERL experimental ecosystems to low level, chro- nic additions of n* 2 fuel oil. Mar. Environ. Res., Vol. 4, pp. 279-297. Guille, A. & J. Soyer, 1968, La faune benthique des substrats meubles de Banyuls-sur-mer. Premi6res donn4es qualitatives et quantitati- ves. Vie Milieu, vol. 19 (2-13), pp. 323-359. Johnston, R., 1970, The decomposition of crude oil residues in sand co- lumns. J. Mar. Ass. U.K., Vol. 50, pp. 925-937. Juario, J., 1975, Nematode species composition and seasonal fluctuation of a sublittoral meiofauna community in the german Bight. Ver6ff Inst. Meeresforsch. Bremerh., Vol. 15, pp. 283-337. 1 Lacaze, J.C., 1979, Exp6rimentations en ecosyst@mes marins control6s. Application @ la pollution par les produits p6troliers. Oc6anis, Vol. 4, pp. 423-611. Mevel, G., 1979, Conditions de nitrification de la mati6re organique dans les s6diments marins en syst@me clos. Applications aux 6le- vages de Pdn6id6s. Th6se Doc. 36me cycle Oc6anogr. Biol., Univ. Bretagne Occidentale, 130 pp. 242 Pearson & Rosenberg, 1978, Macrobenthic succession in relation to or- ganic enrichment and pollution of the marine environment. Ocea- nogr. Mar. Biol. Ann. Rev., Vol. 16, pp. 229-311. Platt, H.M. & R.M. Warwick, 1980, The significance of free-living ne- matodes to the littoral ecosystem in the shore environment, vol. 2 : Ecosystems.ed. by J.H. Price, D.E.G. Irvine and W.F. Farnham, Acad. Press London, pp. 729-759 Raffaelli, D.G. & Mason, C.F., 1981, Pollution monitoring with meio- fauna, using the ratio of nematodes to copepods. Mar. Pollut. Bull., vol. 12, pp. 158-163. Wade, T.L. & Quinn, J.G., 1980, Incorporation, distribution and fate of satured petroleum hydrocarbons in sediments from a controlled marine ecosystem. Marine Environ. Res., Vol. 3 (1), pp. 15-33. Warwick, R.M., 1981, The Nematode/Copepod ratio and its use in pollu- tion ecology. Mar. Pollut. Bull., Vol. 12 (10), pp. 329-333. Wieser, W... 1960, Benthic studies in buzzards bay. II. The meiofauna. Limn. Oceanogr. Vol. 5 '2), pp. 121-137. 243 EVOLUTION A MOYEN-TERME DU MEIOBENTHOS ET DU MICROPHYTOBENTHOS SUR QUELQUES PLAGES TOUCHEES PAR LA MAREE NOIRE DE LIAMOCO-CADIZ par Philippe BODIN et Denise BOUCHER Universite' de Bretagne Occidentale, Laboratoire d'Oce'anographie biologique, 6 avenue Le Gorgeu, 29283 Brest Ce'dex, France. ABSTRACT The ecological follow-up undertaken after the Amoco-Cadiz oil spill, on the beaches Brouennou and Corn ar Gazel (mouth of Aber Benoit) and Kersaint (near Portsall), was continued untill november 1980. Chlorophyll pigments have suffered little quantitatively from the direct effect of pollution, but the study of temporal variations in the meiofaunal densities revealed disturbances in seasonal cycles. Other factors, e.g. hydrodynamic fluctuations and macrofaunal preda- tors, could act as regulating mechanisms on the evolution of the po- pulations. The effects of pollution are particularly obvious in some fau- nistic imbalances, as the study of harpacticoid c6pepods showed. However, particular evolutionary trends between and within ecologi- cal groups of species implied that recovery was nearly complete, at least on exposed beaches. The conclusions drawn to date are tentative because of the lack of reference data, and it is intended to continue the survey annual- ly in spring. Key words : Pollution, Amoco-Cadiz, Chlorophyll pigments, Meiofauna, harpacticoids, Beaches. RESUME Le suivi e6cologique mensuel entrepris, 'a la suite de la catas- trophe de l'Amoco-Cadiz,sur les plages de Brouennou et Corn ar Gazel, 'a l1entr6e de l'Aber Benoit, et de Kersaint pr@_s de Portsall, a e6te' maintenu jusquIen novembre 1980. Alors que les pigments chlorophyl- liens ne semblent pas avoir souffert de l1action directe de la pol- lution, 11e'tude des variations temporelles d.e la densite' de la me'io- faune r6v@le une perturbation des cycles saisonniers. D'autres fac- teurs, tels que 1'hydrodynamisme et les pre'dateurs de la macrofaune, peuvent intervenir en tant que me'canismes r6gulateurs. Les effets de la pollution sont surtout sensibles au niveau de certains de'se'quilibres faunistiques, comme le montre 1'6tude des Co- pepodes Harpacticoldes. Cependant, une certaine 'evolution des groupes 6cologiques permet de penser qulun processus de retour a@ 116tat ini- tial est en cours d1ache'vement, du moins Sur les plages de mode battu. En fait, Vabsence d'e'tats de re'f6rences nous oblige encore 'a la prudence dans llinterpr6tation des re'sultats, et il est envisage une poursuite des recherches sous forme de I'veille" e'cologiques. Mots-cl&s Pollution, Amoco-Cadiz, Pigments chlorophylliens, Me'io- faune, Harpacticoldes, Plages. 245 INTRODUCTION Dans le cadre du suivi e6cologique entrepris, a la suite du nau- frage de 1111AMOCO-CADIZ11, par les Laboratoires de l'Institut d'Etudes Marines de lUniversite' de Bretagne Occidentale, la me'iofaune sensu Lato et le microphytoqbenthos de la zone intertidale ont ete' 11objet d'une 'etude realise'e par le Laboratoire d'Oce'anographie biologique. Apqres une recherche de site effectue'e sur la c8te nord-Finiste're au cours des mois de septembre et octobre 1978, deux plages A la sor- tie de l'Aber Benoit (Fig. 1), Corn ar Gazel au SW et Brouennou au NE, ont e5te retenues pour cette 'etude. Ces deux plages, situe'es dans une zone particulie'rement 6prouvq6e par la pollution due aux hydrocar- bures de 111AMOCO-CADIZ1, sont e6galement etudie'es au point de vue physico-chimique et au point de vue de la macrofaune (Le Moal, 1981) dans le cadre de ce suivi. Il a malheureusement e't& impossible de trouver dans cette rq6gion une plage ecologiquement homologue mais non pollue'e afin de servir de t6moin. Une 'etude paralle'le, mais por- tant uniquement sur la mq6iofaune sensu st-icto, a eke' re'alise'e sur la plage de Kersaint (pre's de Portsall). 48 4qN. AMOCqW j % %qV2qV Brouennou Cro ar aze Oh W 48. 34' 6qSqA stations Portsll q0 2km _4'N Kersint 438 FIGURE 1. Localisation des stations. Sur chacune des plages, unqe station (Quadrat) situe'e en dessous de la m4qi-mareq'e, dans 112q6tage m2q6diolittoral, est q1q1objet de prq6l2q6qve- ments mensuels depuis mars 1978 pour la plage de Kersaint, novembre 1978 pour les plages de Brouennqou et Corn ar Gazel. Sur cette der- nieq're, deux preq'le'vements qI'de refq4rence" ont pu 0qi2q6tre rq6alis2q4s le 17 mars 1978, avant llarriveq'e de la nappe dq1hydrocarbures. Une premiere publication (Bodin et Boucher, 1981) faisait eq'tat des rq6sultats acquis en juillet 1979. La preq'sente note les comple'te par les donne'es qobtenues jus2qquIqen novembre 1980 et tente une re'- flexion sur lq1ensemble de ce suivi q6cologique. 246 Les techniques de prelevement et d traitementes 6hantillons, ainsi que les principaux paramq@tres e'daphiques, ont de'ja' e'te' expose's dans la premiq@re publication, nous ne les reprenons donc pas ici*. Nous rappelons simplement quelques caracte'ristiques granuqlome'triques essentielles sous forme de courbes pondq6rales cumulatives (Fig. 2). De plus, nous pre'sentons le profil topographique de deux des plages prospecte'es (Fig. 3) et les variations temporelles de la temp'rature et de la vitesse du vent (Fig. 4). 75- KERSAINT CORN AR GAZEL BROUENNOU 3/80 P0 3/80 4/86 50- 1 qM190 P 0 p 2,4 so 7,15 Md 1301 NU 130 m So 1,2 So 1,31 6/80 0- 6/80- /80 - P% P 0 % P 0,4% 25. Md 190 M M130 M So 1,09 Md 145 I So119 So13Ipm m 63 8 0 ;0 125 160 200 315 500 i 80 100 125 10 2;0 315 q; 63 80 100 125 16 2;0 M 50'0 Boo i5 'trie courbes ponde'rales cumulatives. FIGURE 2. Granulome Teneur en pelites (p 2qU, M6diane (4qM), Indice de triage (So). BROUENNOU CORN AR GAZEL SUPRALITTRAL SUPRALITTORAL Galets :Mq@EDIOLITTHAL MEDIOLITTORAL MME (40) 1 PME(40) INFRALITTORAL 1 INFRALITTORAL M.M. I M.M. q: --------- BM ME (40) Q BM IME (40) E Q 100 M FIGURE 3. Profil topographique des plages. Limites des 6tages bathy- triques. Emplacement des quadrats N 5 Nq'q08q1 J' F'Mq' AqM'qJJ 'q7jgq' A 3 0 N 04q1 J F'M Aq'M Jqq'0q@Oq' A 9 0 .q0q1 A'Sq'O N qDqIJ Fq'M'Aq'M'Jq'J A'qSqo qN 9 q1 197 1980 FIGURE 4. Tempeq'rature de l1air (a vitesse du vent (b) variations 3/" q"r P NN 4q7 /0 P 4/ P 8qjq" 6/8 6/8 'go P4 So '9 4,F "9 2qL4qi 0q6q@qa4q@ des moyennes mensuelles. Une correction doit cependant 2q&tre apporte'e (Bodin et Boucher, 1981, p. 328) : 'a la place de P.M.M.E. il faut lire B.M.Mq.E., et a la place de B.M.M.E. il faut lire B.M.V.E. 247 RESULTATS Pigments chlorophylliens Un de'pouillement additionnel de carottes pour les pre'le'vements antq6rieurs 'a septemqbre 1979 modifie 1q6ge'rement les chiffres pre'cq6- demment oqbtenus et puqblie's (Bodin et Boucher, 1981). Le nomqbre de carottes utilise' a permis l'utilisation de tests statistiques : test U de Mann-Whitney et test de Kruskall-Wallis hypothe'se nulle rejetq6e au niveau 5 %. Brouennou Dans les premiers centime'tres dle'paiqsseur du sediment on observe une dq6croissance trq@s rapide des teneurs en pigments chlorophylliens, ce qui nous a permis de limiter 8qVe'tude aux quatre premiers centi- mq@tres. Pour la chlorophylle a, ce gradient est triq@s marquq6 (on retrouve en moyenne 12 % de qla teneur superficielle sous 4 cm dle'paisseur) et reguliq6rement observe' dans les prq6le'vements (a 0qVexception des mois de dq6cembre 1.98 et 1979). Llhq6teroge'nq6ite' spatiale est importante et, de ce fait, qles te- neurs moyennes calcule'es pour les deux premie'res couches (0-0,2 cm et 0,2-1 cm) ne sont pas significativement diffe'rentes pour la majo- rite' des pre'qlq@vements mensuels, alors que la prq6sence du film super- ficiel, plus riche en chlorophylle a, est constat6e dans 85 % des carott'es. Pour les phq6opigments, le gradient est moins accentue' (26 % de la teneur superficielle soqtt presents en moyenne sous 4 qdm dlepais- seur) et moins fe'quemment observe' (absent en novembre et d6cemqbre 1978, de dq6cembre 1979 a fq6vrier 1980 et de juillet 'a septemqbre 1980) que dans le cas de la chlorophylle a, ce qui peut q@@tre dqiqa en partie a ses plus faibles teneurs et donc a la moindre prq6cision du dosage. LIenrichissement superficiel en phe'ophytine nlest rencontre, que dans 63 % des carottes. La chlorophylle a est le pigment largement dominant sOrtout au sein des deux premiers centimq&tres dlepaisseur, qla' ou se limitent les variations temporelles. Sa teneur relative q6qlevq6e (71 % de la somme Ca.+ Phe'o).est un indice de la pr6sence d'une active popula- tion de microphytes dans cette zone correspondant A 11q6paisseur maximale de la couche oxygene'e. L'amplitude des variations temporelles est maximale au niveau de la couche superficielle et slattq6nue rapidement dans 11e'paisseur du sediment. La chlorophylle a pre'sente, les deux anne'es, un cycle annuel de type saisonnier (Fig. 5a). Dans la couche superficielle, le minimum de de'cembre est suivi par un fort accroissement pendant les mois dq1hiver ; il aboutit a un q11plateau printanier" entre mars et juillet 1979 (22q,1 4qpg/g) et entre f6q6vrier et juillet 1980 (15,q6 4qpg/g, en excluant le mois de juin). Entre juillet et ao6qat, une nette d2q6croissance est observeq'e elle est suivie par un "plateau automnal" entre aoqi8q@t et novembre 1979 (14 qp4qg/g). Au cours de ces deux cycles il faut noter le minimum esti- val esquiss2q6 en juin 1979, trq@s prononceq' en juin 1980, ph0q6nomq6ne d0q6j0qa observe' en zone intertidale (Colijn et Dijkema, 1981) mais non expli- cite'. 248 A-& 0.0 - 2 cm Pg/g &-,a 0.2-1.0 cm 1.0-1.8 CM 1.8-2.6 cm A s_0 2.6-3A cm Q_o 3.4 4.2 C m 20 A A 10 A' 0- -C 411-pl, 0 A I T a J MAMJJAqS^^ON D IJ F M A M j i A S N D 1979 Pi 1980 PS/9 A 10 b A \ A \A A ztq- 0 0 F A N D, J M M J J A S 0 N 1) F A'8qA J J A S 1979 1980 FIGURE 5. Variation des moyennes mensuelles de la chlorophylle a (a) et de la phe'ophytine (b) a Brouennou. Dans la couche sous-jacente les fluctuations sont similaires. Le plateau printaniersitue' entre fe'vrier et aoq5tcorrespond 'a une teneur moyenne de 15,8 qug/g en 1979 et de 11 q1jg/g en 1980 et le pla- teau qau4qtom0qbalq,entre septembre et novembre 1979, 'a une teneur moyenne de 9,4 8qpg/g. Les deux cycles annuels de la pheq'ophytine (Fig. 5b) diff2q@rent essentiellement par le minimum estival trqeq's accuse' en 1980, les te- neurs moyennes des plateauxatteiqnts de mars 'a novembre 1979 (8,8 4qjqig/g) et de feq'vrier 'a mai 1980 (8,2q5 4qPg/g)q)8q6tant semblables. Lesq"variations temporelles de ces deux pigments sont paralq- le'les, sauf au cours de 6qlq1automqne oquq' elles tendent 'a slinverserq, faisant diminuer la teneur relative en chlorophylle a. Les fortes diminutions de concentration 0qsont accqompagneq'es dq1une disparition ou dq1une att8q6qnuation du gradient dans 1q1eq'paisseur du q'diment. Cette homogeq'nqeq'isation.des couches superficielles, tr6q6s 249 apparente en de'cembre 1978 et ao5t 1979, moins prononce'e en de'cembre 1979 et juin 1980, peut 6tre attribue6e a' une 'erosion de la pellicule superficielle et a un brassage du se'diment sous Vaction des forces hydrodynamiques. Clest A la suite de ces de'croissances que se situent les plus forts accroissements relatifs (100 % entre de'cembre 1978 et janvier 1979, 74 % entre d6cembre 1979 et janvier 1980, 77 % entre juin et juillet 1980). Ces accroissements sont du migme ordre de grandeur en hiver et en e'te', et il semble donc que les facteurs climatiques (tempe'rature, 6clairement) ne soient pas limitant. La comparaison des re'sultats obtenus en 1979 et en 1980 montre pour la chlorophylle a, qulil nly a pas de diff6rences significatives entre les moyennes mensuelles de ces deux anne'es des mois de janvier a mars, tandis que celles-ci sont toujours plus 61ev6es en 1979 du mois d1avril au mois de juillet. Corn ar Gazel La distribution des pigments au sein des 12 premiers centime'tres de se'diment sle'tant re've'le'e tr@s homoge'ne, nous avons e'tudie' trois couches successives de 4 cm dle'paisseur. Les teneurs moyennes en chlorophylle a et en phe'ophytine varient peu entre les-trois couches (variation environ de 10 %). On reconnait cependant, surtout pour la chlorophylle a, deux types de distribu- tion : dans le premier type, la teneur est maximale dans la couche superficielle puis d6crolt re'gulie'rement, dans le deuxie'me type la teneur est maximale dans la couche interm6diaire (4-8 cm). Sur l1ensemble des pre'le'vements, la teneur moyenne mensuelle en chlorophylle a du s6diment est la plus faible dans la couche la plus profonde. Entre les deux premi@@res couches, comme a Brouennou, la diff'rence observe'e nlest pas, le plus souvent, significative du fait de Vhe'te'rog6ne'ite' spatiale ; elle est cependant corrobore'e.par la fr6quence, dans le pre'l6vement mensuel, de chacun des deux types de distribution verticale. La teneur relative en chlorophylle a du s6diment ne varie pas entre les trois couches. Les teneurs moyennes mensuelles en chlorophylle a (Fig. 6a) de la couche superficielle (0-4 cm) s'accroissent de janvier aN novembre 1979. Apr@s la brutale diminution de de'cembre 1979, les moyennes mensuelles mesur6es en 1980 sont, a l1exception du mois d1avril, toujours inf6rieures A celles de Vanne'e pre'c6dente. Dans la couche sous-jacente la variation des teneurs moyennes mensuelles pre'sente la m6me tendance, mais son amplitude est plus faible. Pour la phe5ophytine on remarque une 'elevation en automne (oc- tobre 1979 - septembre 1980) (Fig. 6b) des teneurs moyennes des deux premie'res couches et, pour l1ensemble des deux anne'es, une augmenta- tion des moyennes mensuelles au cours de l1anne'e 1980. Slil nly a pas de cycle de type saisonnier apparent au niveau des variations des teneurs moyennes mensuelles au sein de chacune des couches se'dimentaires e'tudie'es, un tel cycle se pr6sente (plus nettement pour la chlorophylle a) sous la forme d1une succession r6guli6re des deux types de distribution verticale. Un enrichissement subsuperficiel est constate' pendant 11hiver (de novembre a avril) alors qu'en e6t6 c'est la couche superficielle qui est la plus riche en pigments, ce qui peut 9tre d5 'a la diffe'rence saisonnieNre de sta- bilite' se'dimentaire. 250 a A-A 0-4cm 0--0 4 - 8 cm M--M 8-12cm /* N A A A A I A' I I AA 1A I- Alr--'! I Al A I AI A IA A I A 'A 'AA M A i- -i S ji N DJ, 'F M A M J J A S 0 N 0 J F M A M J J A S 1978 1979 1980 A-A 0-4 cm q-* 4-8 C. a 8-12cm b 5- A1\ EL4qX 0 . . . . .. . . . . M A @--i S P--j N D M A M J J A S 0 N D J F M A M J J A S 1978 1979 1980 q- FIGURE 6. Variation des moyennes mensuelles de, la chlorophylle, a (a) et de la phq6ophytine (b) a Corn ar Gazel. Discussion La comparaison des deux plages montre que les teneurs pigmen- taires dans les couches superficielles, au moment du minimum hiver- nal, sont tre's peu diffq6rentes (10 pg/g envqiron). Elles peuvent etre assimilees 'a la valeur intrinsq@qque de Hartwig (1978) et rq6sultent ici de llidentitq6 des mq6dianes granulomq6triques. De la mq4q@me faqqon, la faible diffe'rence existant au niveau des valeurs moyennes du taux de chlqorophylle a (6q81 % et 71 %) et de 1q1indice de diversit6q6 pigmen- taire (voisin de 2) dans la couche superficiqelle, peut 2q9tre relieq'e 'a 2qlqlidentite' du taux de pe'lites. La diff2q6rence de nature et dq1action des forces hydrodynamiques agissant sur la stabiliteq' et l'oxygq4nation du s8q6diment se refle'te dans la diff2q6rence observeq'e pour la distribution des pigments dans epaisseur du sediment (Fig. 7a, 4qb, c), ainsi que lq'ont constateq' de nombreux auteurs depuis Steele et Baird (1968). 251 a b C pt,60 Ca (Vg1g) Ph6o @Pg/g) Ca (Aglg) -4. (Vg,q) Ca (P9 Ig 0 05 10 15 0 05 10 Is 5 0 05 10 -2- -2- .2- .3. -3- .3- .4- -4- .4. .5- .5- .5. 6- 6 6 7 7 7 8 C 9 9 CM CM FIGURE 7. Rq6partition de la chlorophylle a et de la phe'ophytine 'a Brouennou (a) et A Corn ar Gazel (b, c) dans 116paisseur du sediment. Sur la plage de Corn ar Gazel, la distribution homogq@ne des diffe'rentes caracte'ristiques pigmentaires, mise en place par un brassage sous 8qVaction des vagues, peut se maintenir dans un milieu interstitiel prq6sentant de bonnes conditions d1oxqygq6nation (Gargas, 1970 ; McIntyre et aZ., 1970 ; Hunding, 1971) assure'es par l1exis- tence de houle et de courants de marq6e. Dans ces milieux instables ' les microphytes sont passqivement distribu's, llessentiel de la 84qu e flore est gq6ne'ralement constituq6 par de petites Diatomq6es q1i6es aux grains (Amspoker, 1977). Sur la plage de Brouennou la stabilite' de la surface se'dimen- taire permet le de'veloppement dans la zone photique dun film super- ficiel, tandis que, sous les deux premiers centime'tres, en milieu non oxygene, on observe un erichissement relatif en pigments de de'gradation, ceci pouvant etre da 'a la combinaison entre la migra- tion des formes mobiles vers la couche superficielle et la mort des cellules en milieu fortement reduit (Gargas, 1970). La teneur en pigments chlorophylliens du s6diment est en moyenne deux fois plus 6qlev6e 'a Brouennou qqula Corn ar Gazel lorsque sont considi_re'es les'pellicules superficielles ; elle est au contraire deux fois plus faiqble lorsque les dix premiers centimeqkres sont pris en compte. Il est donc important de pre'ciser 1q6paisseur utilisq6e pour 1q6valuation de la biomasse. Pour une comparaison avec la mq6io- faune, 'a Bruennou, ce sont les valeurs mesurees dans le premier cen- time'tre, zone oqa se concentrent les meiobenthontes, qui repre'sentent la quantite' de nourriture disponible et traduisent la stabilite de cette zone. Lq1q6tude des variations saisonni2q@res, dans le cas de sediments instables comme a Corn ar Gazel, montre qquqlun cycle quantitatif nqiest pas apparent sur une anneq'e et clest essentiellement 11insta- bilit2q6 se'dimentaire qui limite la biomasse des microphytes, la pro- duction dans la zone photique devant q9tre essentiellement exporteq'e aprq@s 8q6rosion et remise en suspension. Lqorsque la surface s6q6dimentaire est stable, comme cela est le cas a Brouennou, i2ql appara8qlt au contraire un cycle quanti tatif 0qD6 7 reproductible. Toutes les eq6tudes mene'es en zone intertidale montrent ce type de cycle annuel de's que le s8q6diment preq'sente une fraction 252 fine importante (signe de sediment stable) (Cade'e et Hegeman, 1977 Admiraal et Peletier, 1980 ; Colijn et Dijkema, 1981). Ce type de cycle se rencontre e'galement en milieu infralittoral (Boucher, 1975). La biomasse pigmentaire est constante au cours des plateaux de prin- temps et d'automne, ce qui laisse prq6sumer soit d1une productivite' plus faible qquIen hiver, soit de 11q6tablissement d1un seuil re'git par les conditions moyennes de stabilite' se'dimentaire d'une part, et dlactqivitq6 de nutrition du maillon secondaire dlautre.part. La pre- mie're hypothe'se nlest pas confirmq6e par les diffe'rentes e6qf6qtudes mene'es en milieu intertidal. On ne peut 11q6tayer, en effet, ni par une photo- inhibition (Cade'e et Hegeman, 1974 ; Colijn et Van Buuqrt, 1975), ni par un effet limitant des concentrations en sels nutritifs (Admiraal, 1977), ni par une saturation de la-zone qphotique (Admiraal et Peletier, 1980), car la constitution de denses colonies de micro- phytes n'est pas sugge're'e par les teneurs observq6es (lors de la for- mation de croq5tes de microphytes, des teneurs superieures 'a 100 qug sont mesurq6es ; Plante-Cuny et aqZ., 1981). La deuxie'me hypothe'se est en accord avec 11observation de la tendance 'a un accroissement en phe'ophytine au cours de cette pe'riode. La comparaison des re'sultats obtenus au cours de ces deux anne'es successives (Tableau 1) montre, A Brouennou comme a Corn ar Gazel, une diminution globale au cours de qla deuxiq@me anne'e deqs teneurs en chlorophylle a, alors que la teneur en phe'opigments reste inchangee ou est en augmentation. Cette variation re'sulte soit d1un effet secondaire de la pollution due aux hydrocarbures de l'IAMOCO-CADIZ", soit de la variation naturelle pluriannuelle (Cadq6e et Hegeman 1974). TABLEAU 1. Valeurs moyennes au sein de chaque tranche de se'diment calculq6es pour les pe'riodes de qjanvier a septembre 1979, 1980, et pour la dure'e totale de 11e'tude. Ca chlorophylle a q(qpg/g) - Phe'o : phq6ophytine (qjig/gq) Ca % Ca x 100 / (Ca + phe'o) - DqI : indice de diversite' pigmeqfqitaire B R 0 U E N N 0 U Epaisseur JANV. 1979 SEPT. 1979 JANV. 1980 - SEPT. 1980 NOV. 1978 SEPT. 1980 Ca Ph o Ca % DI Ca Ph6o Ca DI Ca Ph6o Ca DI 0 -0,2 18,6 6,7 73 1,99 13,2 6 69 2,3 14,9 6,1 71 2,14 0,2-1,0 14 4,9 74 2,26 101 3,6 74 2,49 11, 4,2 72,5 2,44 -1, 8,1 2,9 74 2,76 7, 2,8 72 2,72 7,2 2,8 72 2,81 1,8-2,6 4,5 2,0 69 3,20 3,6 3 54,5 3,19 1 2,7 61 3,19 2,6-3,4 1 2q$6 1,3 67 3,62 2,6 2 5695 3,67 2,7 1 98 60 3,64 3,4-4,2 1,5 q1 60 3,97 2q,0 1,7 54 3997 1q,9 1,6 54 3,89 C 0 R N A R G A Z E L Epaisseur qJqPJW. 1979 SEPT. 1979 qJANVq. 1980 SEPT. 1980 NOV. 1978 SEPT. 1980 Ca Ph o Ca % DI Ca I Phq6o Ca % DI Ca I Ph6o Ca % DI o- 4 12,9 2 87 2,24 8,4 2,75 76 2,40 q1q1 2,05 82 2,29 4- 8 12,2 1,5 87 2,30 8,2 3,0 74 2,51 10,7 q1q,9 78 2,41 8-12 10 1,5 87 2,42 7,55 2,1 78 2,64 9q,q1 1q,q8 82 2,54 253 Les conditions metq6orologiques (facteurs climatiques et hydro- dynamiques) ne peuvent q@q@tre retenues comme facteurs de'terminants de cette variation, nlayant pas e5tq6 plus particulie'rement de'favorables au cours de la deuxie'me anne'e, si ce West en automne, alors que les deux plages, pourtant dtexposition diffq6rente, ont re'agi simi- lairement et ceci dq&s le mois d1avril. La reprise des activite's de grazing est un des facteurs biolo- giques qui intervient au niveau de 4qVe'volution saisonniq6re et qui peut expliquer la diffq6rence observe'e entre les deux anne'es. Il est possible en effet de rapporter ces variations de la biomasse pig- mentaire 'a 4qVe'volution de la macrofaune au sein du processus de de5contamination (Le Moal, 1981). Ainsi, la re'apparition de 11qA1phi- pode Bathyporeia sur la plage de Corn ar Gazel peut qC-tre pour partie responsable de la diminution de la teneur en chlorophylle a, son activite' de "qbrouteur" e'tant reconnue (Sundqbqack et Persson, 1981q)-. Me'iofaune : re'sultats quantiqtatifs La Figure 8 et le Tableau 2 montrent qile'volution temporelle de la densite' (nombre dlindividus/10 cm2 de surface) des Cope'podes Har- pacticoqldes (e'chelle x 100), des Ne'matodes et de la mq6iofaune totale (sensu stricto) dans les diffq6rents prq6lq@vements recueillis aux trois stations prospecte'es. Les Tableaux 3 et 4 indiquent ltq6volution tem- porelle (en pourcentage de la mq6iofaune totale) des autres groupes du me'iobenthos vrai et de la meiofaune temporaire. Iql est q6vident que cette 6volution est diffe'rente d1une station 'a ql1autre. Brouennou La densite' moyenne (8 033 ind./10 cmq2) qy est extq@ement e'qleve'e en comparaison des donnees de la littq6rature pour la zone interti- dale (Hicks, 1977), et la me'iofaune est composq6e essentiellement de Nematodes. Les variations saisonnie'res sont assez nettes et a peu pres conservees d1une anne'e 'a 8qVautre, avec des minima en fq6vrier- mars et en sptembre (1979) ou juillet (1980), et des maxima en avril-mai et en 'dq6cembre-janvier ou octobre (1980). Cependant la densitq6 moyenne des Harpacticoqldes est nettement plus faible durant l seconde pq6riode (1979-80) : 282 ind./10 CqM2 (contre 593 pr6cq6dem- ment). Durant cette seconde pq6riode, le rapport Nq6matodes/Cop6podes oscille entre 11 (juin 1980) et 172 (janvier 1980). Les autres groupes du meiobenthos vrai representenqt un pourcen- tage relativement modeste de la me'iofaune (maximum : 26,7 % en oc- tobre 1979). De plus, ce pourcentage est en regression : iql ne d2q6- passe pas 3,0q5 % depuis juillet 1980. Les Annqe'lides constituent llessentiel du m2q6iobenthos temporaire, ce qui correspond aux donneq'es de la macrofaune (Le Moal, 1981). Corn ar Gazel La densit2q6 moyenne de la m6q6iofaune 4q0 013 ind./10 cm2) y est beaucoup plus faible 6qquqIqa 8qBrouennou. De plus, cq'est a cette station que la diffeq'rence avec la pe'riode eq'tudieq'e prq6ce'demment est la plus nette du point de vue quantitatif la densiteq' moyenne passe de 2 qt 4 697 'a 1 666 ind./10 qcqm . Cette regression n est pas le fait des 254 Wiofue tt - - - - - - - Nematodes ....... ----- Harpactcides BROUENNOU N 4 N/10cm' .......... . ............ ........... 10000 It NI 800 4qv q5000 500 Tv 200 ......... ... ......... 0. ....... ....... 0 'N'D A I q@4qVqIA'N'D q0: A'M' 1 0 N ............ 9000- .......... ....... . ....... ............ ...... ........... ... .......... ...... -BOO 6q0:qKN:::- AR GAqZ:qE:qL ... ............. ........ q5000- 500 ........... Y ...... ..... 3000- 2q00 ar ..... ..... ............ 0 - ....... ... .... ........... F:::M A M'qj 0 q@'qO'N D A0qW I 9000- ...... ............ ............ ... ....... X.: X ........... . .... .......... .... ...... .... ....... -BOO . . . . . . . ......... ............ 70OD- qAIqNT ............ ............ ............. .... . . . . . . . ......... 509 G- ........... 500 3000- . ......... ... 209 'J; 0 4qFIF'M, rA M I _O M A M I I A qS I A S,ON DqIJ'F'M'A'M' I J A' qS N q@1978 19,79 1980 PRINTEMPS ETE AUTOMNE HIVER PRINTEMPS ETE AUTOMNE HIVER PRINTEMPS ETE FIGURE 8. Evolution temporelle des densite's de, la me'iofaune aux trois stations. Copq6podes Harpacticoqldes, mais celui des Ne'matodes et des autres groupes : de 53,8 % en aoq@it 1979, ces derniers ne constituent plus que 11,5 % du meq'iobenthos vrai en ao2qi32qh 1980. Les variations saisonnie'res sont encore plus ou moins marquees durant la seconde peq'riode, aveqc un minimum classique, en feq'vrier mais aussi un maximum en septembre (1979) et deux 16q6gers pics en avril et ao6qat 1980). Dans ce biotope mieux oxyg8q6n2q6, le rapport N8q6matodes/ Copeq'podes varie dans des limites plus eq'troites : 2,2q3 (septem0qbre 1980) 'a 15 (qoctobre 1980). Parmi la meq'iofaune temporaire, les Amphipodes constituent un groupe particulie'rement inteq'ressant 'a cette station oqu' ils avaient subi de lourdes pertes d2q6s le d2q6but de la mar2q6e noire : ils ont eq't2q6 qabsents durant tout 8qIq'hiver 1979q-80, mais ont eq6te' r2q6gulie'rement 255 TABLEAU 2. Evolution temporelle des densite's de la me'iofaune aux trois stations (N110 CM2). 1 97 9 1 98 0 BROUFNNOU 6/8 21/9 8/10 5/11 20/12 17/1 19/2 18/3 30/4 14/5 12/6 10/7 11/8 11/9 23/10 20/11 N6matodes 3928 1312 6163 5840 5611 9960 5179 3444 9155 8624 6995 5293 6368 8261 7531 7904 Harpacticoides 200 33 224 325 149 58 63 68 336 611 630 440 432 160 512 400 Mgiofaune totale 4257 1574 9038 7654 6781 11022 7102 4858 13721 12104 10667 6140 7141 8968 9499 8859 CORN AR GAZEL 7/8 20/9 9/10 6/11 21/12 18/1 18/2 19/3 29/4 13/5 13/6 11/7 12/8 12/9 22/10 Ngmatodes 585 1596 1184 845 917 776 213 796 1480 1124 867 1021 1579 924 968 Harpacticoldes 189 351 272 209 200 83 32 123 180 260 332 283 409 407 65 Wiofaune totale 1870 4166 2283 1264 1444 1073 305 1181 1818 1726 1387 1434 2313 1527 1198 KERSAINT 11/7 7/8 20/9 8/10. 6/11 21/12 18/1 18/2 19/3 29/4 14/5 13/6 11/7 12/8 12 922 10 20/11 Ndmatodes 603 727 427 719 472 348 832 104 .83 200 336 341 376 599 783 660 433 Harpacticoides 208 297 181 82 425 112 123 31 27 56 39 160 232 48 77 39 289 Mdiofaune totale 2954 2309 1252 3068 3101 2352 1859 704 759 1259 3623 3168 1903 1295 1529 865 1256 TABLEAU 3. Evolution temporelle des autres groupes de la me'iofaune et des nauplii M. 1 979 1 9 8 0 BROUENNOU 6/8 21/9 8/10 S/11 20/12 17/1 19/2 18/3 30/1 14/5 12/6_ 10/7 11/8 11/9 23/10 20/11 ROTIFERES +12,1 2,9 7,2+0,5+0,9 3,4 0,51+ + TARDIGRADES ++ ++0,6 0,7 1,3 3,5 3,1 0,5+ ++ 0.6 GASTROTRICHES ++ +++ 0,5 ++ + OSTRACODES ++ ++++ +++ + ++ + TURBELLARIES 1 5,9 10,4 14,6 9,2 5,4 1,5 1,5 1,8 2,9 1,8 1,6 1,0 0,8 0,6+ DIVERS 5,3 20,0 15,1 16,2 7,9 20,5 0,7 0.7 2,9 1,2 0,5 TOTAL 6,7 10,4 26,7 12,1 12,6 7,4 22,7 18,2 23,5 16,2 23.1 1,7 2,0 3,5 1,2 I'l NAUFLII 0,8 Od 0,9 4,0 0,712,518,5 6,4 5,413,3 1,610,7 0,8112.6 4,4 CORN AR GAZEL 7/8 20/9 9/10 6/11 121/12 18/1118/2 19/3 29/4 13/5 13/6 11/7 12/8 12/9 22/10 ROTIFERES 2,3 O'g+ +++++ 1,3++ 0,8 0,8+ TARDIGRADES 1,2 1,0+ 0,7 2,0 1,5 6,6 3,1+ + + GASTROTRICRES 6,9 10,9 4,8 14,1 12,3 4,3 12,5 4,1 4,9 4,5 3,6 9,3 4,7 2,6 OSTRACODES 2,4 2,7 1,2 0,7++4,3 2,0 I'l 1,7 0,61+ +0,8 TURRELLARIES 147,9 @9,0 21,9 4,4 3,1 1,4 3,0 1,3 1,0 1,7 I'S 1,9 1,4 1,4 7,4 DIVERS 3,3 2,3 2,0 0,7 1,0 0,6 0,6+ ++ TOTAL 53,8 50,5 34,0 10,6 19,2 18,5 20,5 20,9 6,9 10,6 7,5 6,1 11,5 6,9 10,8 NATJPLII 2,9 2,1 Od 0,6++0.6 1,5 3,2++ + ++ KERSAINT 11/7 7/8 20/9_1 8/10 6/11 21/12 18/1 18/2 19/3 29/4 14/S 13/6 11/7 12/8 12/9 22/10 20/11 ROTIFERES O'g I'l 1,6 I'D 2,0 4,5 1,3+11,9+ +2,7+ 1,0 1,4 1,4 4,6 TARDIGRADES 7,4 5,7 8,1 5,6 2,6 1,3 7,6+5,4 2,9 1,3 2,0 5,1 2,3 8,9 1,8 7,4 GASTROTRICRES 14,3 1,6+7,2 +1,0 11,9 2,3 2,7 3,1 5,9 3,2 3,1 16,4 4,2 6,1 5,4 OSTRACODES 13,8 13,8 8,0 2,4 0,9 1,4 2,6 23,4 13,9 18,7 4,2 18,7 39,3 20,3 3,0 2,0 3,9 TURBELLARIES 13@,8 31,9 30,3 56,0 64,5 71,0 20 446 243 718 830 918,1 5,4 0,9 11,5 1,4 I'l DIVERS 3:5 5:8 3:4 25:5 44:6 27,7 2,8 3, 15.1 11 TOTAL 71,2 54,1 48,0 72,2 70,0 79,2 47,3 77,7 81,0 69,0 86,9 72,4 55,7 44,4 37,5 15,1125,6 NAUPLII 0,8 1 +11,611,11 +0,5+14,219,6 12,1111,2 10,2 3,8 3,2 TABLEAU 4. Evolution temporelle de certains groupes du me'iobenthos temporaire M. 1 9 7 9 19 8 0 BROUENMU 6/8 21/9 8/10 5/11 20/12 17/1 19/2 18/3 30/4 14/5 12/6 10/7 11/8 11/9_ 23/10 Annglides 0,8 2,8 1,5 3,3 1,8 0,8 0,7 0,7 0,7 2,2 2,6 3,1 1,4 0,6+ Gast6ropodes + + + CORN AR CAZEL 7/8 20/9 9/10 6@11 21/12 18/1 18/2 19/3 29/4 13/5 13/6 11/7 12/8 12/9 22/10 Ann6lides 3,8 ++1,3 3,8 3,5++ + Ianaldac6s + ++3,3 1,3++ 2,3+0,8 + ++ Cumac6s + + +1,4 0,5 3,9+ KERSAINT 11/7 7/8 20/9 8/10 6/11 21/12 18/1 18/2 19/3 29/4 14/5 13/6 11/7 12/8 12/9 - 22/10 20/11 Ann6lides . 0,9 +0,8+ 1,7 1,7+++ +++ + + Tanaldac6s 0,5 1,3 2,2 0,9 0,6 0,8 0,6 2,7 2,0 1,0 +1,4 1,5+ I'D+ + Cast6ropodes 1,8 1,3 2,9 3,8 1,7 C.-gs + + ++ ++ + 256 I presents de mai 'a octobre 1980, confirmant la reinstallation de ce groupe tres important au niveau de la macrofaune de Corn ar Gazel (Le Moal, 1981). Les Cumacq6s ont a peu prq@s le mqiq@qme comportement que les Amphipodes : presque toujours prq6sents entre mai et octobre 1980, ils atteignent 3,9 % de la population totale en septembre. Kersaint Suivie mensuellement depuis le 17 mars 1978, la me'iofaune de cette station prq6sentait une vq6ritable explosion dq6mographique en juin, juillet et aoqat 1978. Ce phe'nome'ne ne s'est pas reproduqlt par la suite, et Von est revenu a des variations saisonniq@res faible- qmenqi accentue'es, avec un minimum en fe'vrier-mars (comme aux deux autres stations) et un maximum en octobre-novembre 1979 (comme a Brouennou) et en mai-juin 1980. La densitq6 moyenne (2 063 ind./10 cm2) est la plus faible des trois, ce que laissaient prq6voir les carac- te'ristiques du sq6diment. qLle'volution temporelle de la me'iofaune de cette station a q6tq6 marque'e par une inversion du rapport Nq6matodes/ Copq6podes a partqir de mai 1978, date depuis laquelqle les Ne'matodes sont devenus pre'ponde'rants et le sont reste's : depuis le mois de juillet 1979, ce rapport oscille entre 1,1 (de'cembre 1979) et 16,9 (novembre 1980). La densitq6 moyenne des Harpacticoqldes est d1ailleurs 2, passee de 366 ind./10 cm entre mars 1978 et juin 1979, a 157 entre juillet 1979 et novembre 1980, en raison principalement du pic Itanormal" de juin 1978. C'est a Kersaint que les autres groupes du mq@e'ioqbenthos vrai sont proportionnellement les plus importants : ils, constituent pre's de 87 % de la population en mai 1980, et les valeurs dq6passant 70 % qne sont pas rares. Les Ostracodes (tous 'a des stades trq@s jeunes) constituent 39,3 % de la population en juillet 1980, et la propor- tiondes Gastrotriches s'e'lq@ve a 16,4 % en aoit de la mq@q@me annq6e mais le groupe le plus important et le plus reguliq@rement present est celui des Turbellariq6s. Parmi le me'iobenthos temporaire, les Tanaqldace's sont toujours presents (2,7 % au maximum en f6vrier 1980), les Anne'lides de- viennent de plus en plus rares a partir de fq6vrier 1980, alors qqulau contraire les Gasteropodes rq6apparaissent depuis juillet 1980 (3,8 % de la me'iofaune totale en octobr). Discussion En l1aqbsenc de donnq6es antErieures au 17 mars 1978, il est bien difficile de dire qquelle est la pq6riode la plus prche de la fnormale" du point de vu quantitatif. Les fortes densite's observees durant la premi@re periode a Corn ar Gazel et Kersaint pourraient correspondre 'a une phase deutrophisation "anormqale consecutive A lq1accumulation de mati6q@re organique dans le sq6diment, accumulation resultant elleq-mq6me de la pollution par les hydrqocarbures. Dans ces biotopes q! "haute 2q6nergieqll, 1q1hydrodynam8qisme intense a pu provoquer un retour A 0qVoligotrophie durant la secqonde pe'riode, alors 6qquqIaq' Brouennou la stabilite' du milieu maintenqait une certaine eutrophi- sationq. Malheureusement, nous ne disposoqns pas dqe donnq6es sur la teneur en mati2q@re organique des sq6diments pour q6tayer cette hypo- theq'se. 257 On peut aussi expliquer, du moins en partie, les chutes de densites de la seconde pe'riode par la re'installation dans le bio- tope de pre'dateurs provisoirement e'liminq6s par 11arrivq6e des hydro- carqbures ; en tout cas, cette rq6installation est e'vidente au ni*veau des Amphipodes. Evolution compare'e de la me'iofaune et du microphytobenthos La comparaison des re'sultats obtenus, aux deux stations de Brouennou et Corn ar Gazel, pour les pigments chlorophylliens de la couche superficielle (0-1 cm) et la me'iofaune, montre que ql1ampli- tude des variations et les valeurs maximales de la densite' et de la teneur pigmentaire sont plus q6qlevq6es a Brouennou, biotope leplus stable. A cette station, des relations de type trophique entre mi- crophytes et me'iofaune sont fortement suggq6rq6es. On observe en effet un relais entre la phase d1accroissement des pigments chlorophyl- liens (dq6cembre a mars) et celle de la me'iofaune (avril-mai), relais suivi d1une phase dle'quilibre relatif. Llaccroissement des pigments chlorophylliens apparaqlt donc en hiver, alors que 11activite' des mq6iobenthontes est ralentie et leur densite' en diminution. Les fac- teurs climatiques nlq6tant pas ici limitants pour les microphytes, on est en droit de penser que clest une diminution du "grazing" qui favorise leur accroissement. A Corn ar Gazel, ql1amplitude des variations saisonnie'res des pigments chlorophylliens et de la mq6iofaune, surtout depuis juin 1979, est fortement limitq6e par l1action de 1'hydrodynamisme. Il est possible dle'tablir une coincidence entre la distribution verticale des pigments et la valeur du rapport Ne'matodes/Cope'podes : ce rap- port prq6sente ses plus fortes valeurs en hiver, pq6riode pendant la- quelle les Cope'podes, assez infe'ode's ici 'a la surface (au contraire des Ne'matodes), sont moins nombreux et ou' iql y a aussi moins de pig- ments. Llinstabilitq6 de la couche superficielle semble donc qatre le facteur limitant de la biomasse primaire et secondaire a cetqte sta- tion et, de ce fait, une possible relation trophique est masque'e. Les Copq6podes Harpacticoqldes : 'etude qualitative Comme nous avons de'ja' eu 4qVoccasion de le montrer (Bodin et Boucher, 1981), une etude qualitative est souvent plus rq6ve'latrice des perturbations d1un peuplement qu'une simple q6tude quantitative. Variations temporelles des diffq6rents groupes q6cologiques Apre's d2q6termination, les esp6q6ceqs dq'Harpactico2qldes ont 36qk6q6 regroupees par affinit2q6s 2q6cologiques (Tableau 5) dq1apr2qes nos ob- servations personnelles et les donn2q6es de la litteq'ra4qture, operation toujours d6q6licate en raison des incertitudes qui pe'sent sur 0qI'e'co- lqogie de certaines eqspeq'ces. La compaqraison de deux aqnneq'es conseq'cu- tives : novembre 1978 'a octobre 1979 et novembre 1979 'a octobre 1980, met en 2q6vidence une certaine 2q6volutiqon des groupes 6q6qco6qlogiques au niveau de chaque station. Pour chacuqne des deux ann8q6es et pour chaque groupe, deux variables ont 2q6t0q6 calculeq'es : la somme des den- sit6q6s des esp6q@ces concerneq'es (6qE densiteq's) et la dominance geq'neq'rale moyenne (D.g.m.) (Bodin, 1977). 258 TABLEAU 5. Liste des espe'ces re'colt6es aux trois stations entre ao5t 1979 et novembre 1980 (s = sabulicole, v = vasicole, p = phytophile, e m eurytope, m = me'sopsammique). BROUENNOU CORN AR GAZEL KERSAINT (6/8/79 au 20/11/80) (7/8/79 au 22/10/80) (11/7/79 au 20/11/80) 'roupe Fr6quence Fr6quence Fr4quence 6col. D.g.m. D.g.m. D.g.m. % Canuei,ea 6uAcigm v + @6 R Canuetta peApeexa S 12,0 too c 80,3 100 c 0,4 35 F HaZectino,soma heA&nani S + 12 R + 7 R + 12 R Neudob)tadya beduina S + 7 R AAenozeteUa sp. m 0,9 18 R Tachid,W6 di6cipa e 0,6 19 R + 7 R 0,8 29 F WcAoaAthAidion Aeductum v + 12 R Thomp6onu& hyaenae S I'l 20 R HoApacticu,6 6texus S 13,7 100 c 0,2 7 R 6,7 18 R Ti6be sp. p + 13 R PaAathatut&iz dovi p + 6 R + 7 R Dactytopodia sp. p + 6 R PaAa6tenhetia spino6a butbosa p 0,1 6 R + 6 R Stenhetia (Det. ) pattatAiz bisp v O's 25 F Robextzonia cettica p 24,9 100 c + 6 R SutbampkWcu6 -imuz e 0,7 37 F Amphio,6cu6 vaAian.6 p + 7 R Amphia6cu,6 tonga&ticutatuz S + 7 R Amphiuco@dez zubdebit" p + 6 R Amphiucolde6 debZW s. str. e 39,9 100 c Amp"coZdm debitZz tim@co&A v 0,6 69 FF SchizopeAa sp. p + 6 R Apodop4yttu6 aAenicotu6 m + 6 R 10,9 100 c Ktiop,6yZeu6 consttictuh s. str. m 5,5 47 F lnteiunedopzyttu,6 inteAmediLL6 m + 12 R PaAatepta,6tacu6 6pZnicauda m 0,2 12 R 0,1 20 R 55,4 100 c Me,6ochAa pygmaea e + 7 R Enhyduzoma p4opinquum v + 19 R RhizothAix minuta S + 6 R 2,2 so c 1,9 71 FF Huntemannia jadenziz v 0,3 37 F Heteutaophonte 6t)t6mi s. str. p 5,9 81 c + 7 R HeteAotaophonte tittoxaLL6 p 4 6 It Pwatuphonte buvixost&iz sstr. p 6 R + 7 R Pa)Lonychccamptu6 cmticaudatu6 S + 6 R ksettopiz hi6pida S + 6 R Azettopsiz inte4media S 0,4 44 F 15,8 100 c 16,9 88 c 259 A Brouennou, quatre groupes ecologiques peuvent 9tre distin- gues : les sabulicoles, les vasicoles, les phytophiles et les eury- topes. D'apre's les D.g.m., ces groupes se r6partissent de la'faqon suivante Pe'riode du 3/11/78 Pe'riode du 5/11/79 au 8/10/79 au 23/10/80 E densite's D.g.m. E densites D.g.m. Sabulicoles 942 18,4 917 @25,1 Vasicoles 477 9,2 62 1,6 Phytophiles 1 628 31,8 1 043 28,5 Eurytopes 1 525 29,9 1 620 44,4 Phytophiles t Eurytopes 3 153 6197 2 663 72,9 La premiere anne'e, les phytophiles dominent, avec pre's de 32 % des Harpacticoldes ; viennent ensuite les eurytopes, puis les sabu- licoles et, enfin, les vasicoles. La seconde anne'e, les eurytopes deviennent largement pr6pond6rants, avec plus de 44 %, et les phyto- philes passent en seconde position. LIensemble phytophiles t eury- topes progresse de plus de 11 %. Les sabulicoles et les vasicoles evoluent en sens inverse, clest-a'-dire que les sabulicoles pro- gressent de pre's de 7 %, alors que les vasicoles sont r6duits d1au- tant (Fig. 9). A Corn ar Gazel, ces quatre mgmes groupes e'cologiques sont repr&sent6s la premiere ann6e, alors que les vasicoles et les eury- topes disparaissent la seconde ann6e. Mais, 'a cette station, les sabulicoles rassemblent toujours environ 99 % de la population : Pe'riode du 15/11/78 Pe'riode du 6/11/79 au 9/10/79 au 22/10/80 E densit6s D.g.m. Z densit6s D.g.m. Sabulicoles 3 247 98,6 2 905 99,8 Vasicoles 4 0@1 - - Phytophiles 2 O'l 4 O'l Eurytopes 27 .0,8 Phytophiles + Eurytopes 29 0,9 4 0,1 Il n'est donc plus question de variations entre les groupes, mais il est,inte'ressant de noter ici une variation 'a 11inte'rieur du groupe des'sabulicoles. Celui-ci est compose essentiellement de deux especes ::AseZZopsis intermedia et CanueZZa perpZexa. La premiere ann6e, A. intermedia est pre'ponde'rante, avec une D.g.m. de 71,5 % contre 18,2 % 'a C. perpZexa. L'anne'e suivante, clest C. perpZexa qui redevient largement dominante (comme cle'tait le cas en mars 1978) avec 88,7 % de la population, contre seulement 9 % a A. intermedia (Fig. 10). A Kersaint, station de sable pratiquement pur, les espe'ces vasi- coles sont e'videmment absentes. Avec une m6diane de pr6s de 200 pm, ce sable est propice a 11installation des formes typiquement inters- titielles ; il devient alors ne'cessaire de distinguer, parmi les Harpacticoldes, un groupe dlesp@ces sabulicoles me'sopsammiques et 260 N/10cm' O-c E,,,Ylopes a Phytophiles Vas,coles 300 ol 0 /* Ne, Z@ - - N D J F M A M i J A S 0 N DIJ F M A M J J A S 0 N 1978 1979 1980 FIGURE 9. Brouennou : 'evolution temporelle de'la densite' des Harpac- ticoldes regroupe's par affinite's e6cologiques. N/10, A Soo (517) 0-() 400 III///IINkIx 300 20D `/0 100 so 0-0-0 M S 0 D J F M A M i i A J F M A M J A S 1178 1979 1980 FIGURE 10. Corn ar Gazel : 'evolution temporelle de la densite' des prin cipales espe'ces d'Harpacticoldes. A 950 900 Soo @P, . A-A 700 Soo Soo 400 300 20 100 01 A . ....... so 0 11 A M i i A S 0 0 0 J F A 0 J, J A S 0 J F -A J A S 0- 1978 1979 1980 FIGURE 11. Kersaint 'evolution temporelle de la densite' des Har- pacticoldes regroupe's par affinite's e6cologiques. 261 un groupe de sabulicoles e'pi7-et endopsammiques. Par ailleurs, comme elles sont peu nombreuses et peu abondantes, les espe'ces phytophiles et les espe'ces eurytopes sont regroup6es dans un seul et mgme groupe Pe'riode du 21/11/78 Pe'riode du 6/11/79 au 8/10/79 au 22/10/80 E densit6s D.g.m. E densit6s D.g.m. Sabulicoles 1 1 120 45,9 1 181 81,2 mesopsammiques Sabulicoles 1 1 296 53,3 260 17,8 epi-endopsammiques Phytophiles + Eurytopes 20 0,8 15 1,0 L16volution de ces groupes est assez significative : durant la premi@re anne'e, les formes e6pi- et endopsammiques dominent avec plus de 53 % de la population, alors que, l'anne'e suivante, les formes mesopsammiques reprennent largement la predominance avec plus de 81 % (Fig. 11). Diversit6 D'une p6riode a l1autre, on observe une chute importante de la richesse sp6cifique : 51 esp@ces avaient 6te' recense'es dans les trois stations jusqu'en juillet 1979, on n1en compte plus que 36 entre aoat 1979 et novembre 1980 (Tableau 5). De plus, le nombre dlespe'ces dominantes (D.g.m.@> I %) diminue aux trois stations : au total, on passe de 28 esp6ces dominantes du- rant la premiere pe'riode 'a 15 durant la seconde. Parallelement, les esp@ces principales voient leur dominance g6n6rale moyenne augmen- ter. A Brouennou, la D.g.m. de Amphiascoides debiZis s. str. passe de 23 'a 40 %. A Corn ar Gazel, la D.g.m. de A. intermedia 6tait de 63 % durant la premi6re p6riode e6tudi6e, celle de C. perpZexa est de 80 % durant la seconde pe'riode. A Kersaint, durant la premi6re p6riode, la D.g.m.. de A. intermedia 6tait de 26 %, celle de Kliop- syNus constrictus s. str. de 25 %, celle de Paraleptastacus spini- cauda de 22 % ; durant la seconde pe'riode, la D.g.m. de P. spinicau- da passe 'a plus de 55 %. Enfih, le cas de K. constrictus est int6ressant 'a consid6rer cette esp@ice avait une position tout a fait pre'pond6rante jusquIen juin 1978 ; elle est reste'e fr6quente par la suite, mais sa D.g.m. est tombe'e de 24,7 a 5,5 %. Cependant, on observe une recrudescence de cette forme m6sopsammique eii novembre 1980, oa sa dominance par- tielle est de 48,2 %, ce qui nous rapproche de la situation initiale de mars 1978. Discussion Dans l1ensemble, on assiste donc a une progression des espe'ces sabulicoles et 'a une re'gression des vasicoles. A Corn ar Gazel et 'a Kersaint, 1'6volution aboutit m6me a une situation proche de celle qui pre'valait en mars 1978 ; la diminution de A. intermedia et l1augmentation du stocR des m6sopsammiques laissent supposer une de'pollution du milieu, d6pollution facilit6e par un hydrodynamisme plus intense 'a ces stations. Mais, 'a Brouennou, la progression des eurytopes est encore plus nette que celle des sabulicoles, grAce a certaines esp@ces telles que A. debilis s. str. qui occupent encore largement le biotope. 262 Doit-on considerer ces esp2qkes (A. intermedia et A. debiqZis comme des "opportunistes" au sens ou' l1entendent Bellan (1967) et Glq6marec et Hily (1981) pour la macrofaune ? Il est sans doute en- core trop tq8t pour ql1affirmer, car nous manquons dle'tats de re'fe'- rences de ce type en me'iofaune. Du point de vue de la richesse spq6cifique, clest la station de Kersaint qui a perdu le plus dlesp,@_-ces (7) par rapport a la premiq@re periode e'tudie'e (en juin 1978, 18 espe'ces e'taient prq6sentes a Ker- saint ; en juin 1980, il n'y en avait plus que 6) ; Brouennou en a perdu 5 et Corn ar Gazel en a gagne' 2. Mais le phe'nome'ne le plus significatif, a notre avis, est la rq6duction du nombre des espq@ces dominantes de chaque station et la tendance a la concentration de la faune harpacticoqldienne sur que1qques espe'ces particulie'rement bien adapte'es au biotope. A Corn ar Gazel, cette tendance est poussee 'a i1extrq9qme, clest-qA-dire qquIon a un peuplement presque monospq6ci- fique, correspondant a un biotope trq6s se'qlectif d'oq@qi les espq6ces qui avaienqt envahi le milieu a la suite de la pollution disparaissent peu a peu. CONCLUSION Le microphytobenthos est trq6s vite apparu, sur ces deux plages, qpeu sensible 'a 4qVaction directe de la pollution (dosages de pigments et observations microscopiques in vivo rqq'alise's en avril 1978) mais, partie inte'grante de 11q6cosyste'me, iql rq6agit au de'se'quilibre provo- quq6 dans celui-ci. Iql est un re've'lateur des caractq@res edaphiques du biotope et reprq6sente un maillon du rq6seau trophique benthique sous sa forme active (chlorophylle a) ou de'tritique (phe'ophytine). Son q6tude apporte des q61q6ments dans la distinction entre des fluctua- tions naturelles provoque'es par 11hydrodynamisme et une rq6action A la pollution des peuplements animaux qinterstitiels, ce quqi expliqque- rait la diffq6rence cons-11-atq6e entre les deux annq6es. Le mq6iobenthoqs, en tant que niveau trophique esentiellement lie' au substrat, est particuliq@rement sensible aux fluctuations des 'tres 'clogiques, comme 8qVont montr' de nombreux auteurs parame e e (Gray,, 1971 ; Arlt, 1975 ; Giere, 1979, Frithsen et qElmgren, 1979 Coull et Bell, 1979 ; Renaud-Mornant et Gourqbault, 1980 ; Boucher et al., 1981). Le mq6iobenthos est particuliq@remnt p're'cieux dans le cas des biotopes pauvres en macrofaune (Kersaint). Du point de vue quantitatif,, on peut constater qqulil n' 'y a pas eu d"'he'catombe" dans la me'iofaune, comme ce fq@qrit le ca's e'n,dlautres circonstances (Wormaqld, 1976). Mais on observe des perturbations au niveau des cycles saisonniers. A Kersaint, par qexempleq, il semble que la pqreq'sence d'hydrocar6qbures ait provoqueq' unqe req'gression des peu- plements (en particulier des Copeq'podes Harpactico8qldes) jus4qquIen mai 1978, ce qui a eu pour effet de retarder de deux mois le pic de printemps. Ce deq'calage nqlest plus que de uqn mois lq1anneq'e suivante et il est compl2q@tement r2q6sorb2q6 en 1980. Lq'q612q6vaqtion temporaire des densit2q6s observeq'e au bout de quelques mois peut q9tre une autre con- sequence de la pollution 6qli6q6e a une eutrophisation inhabit4quelle du milieu provoque'epar un 0q6ventuel apport de matie'res organiquesq. Dans les milieux de mode battu, 2q11hydrodynamisme a agit rapidement pour, au bout de 12 a 16 mois, op6q6qrer un retour a l'olig4q6trophie ha6qbi- tuelle de ces milieux. Dq'une certaine manie're, on retrouve ici le 263 schena' des qmq6canismes rq6gulateurs des e'cosyste'mes e'tudie's en baie de Morlaix par Boucher et aql. (1981). Mais, en l1absence d1q6tats de re'- fe'rences ante'rieurs a la marq6e noire pour ces stations, nous reste- rons prudents dans llinterpre'tation des chutes de densite's observe'es 'a Corn ar Gazel et Kersaint depuis juillet 1979, oqa intervient pro- baqblement aussi la rq6apparition des pre'dateurs de la macrofaune. Beaucoup plus re'vq6latrices sont les perturbations au niveau qua- litatif observe'es chez les Cope'podes Harpacticoqldes. Apre's ql1afflux despe'ces qui a fait suite 'a la mare'e noire (juin 1978 a Kersaint), une diminution de la richesse spe'cifique jointe a une certaine q&vo- lution des groupes e'cologiques montrent qu'un processus de retour 'a 11q6tat initial est sur le point d'aqboutir, en novembre 1980, sur les plages de mode battu. Par contre, une plage de mode abritq6 telle que Brouennou semble A un stade de d6pollution moins avancq6, A moins que ce ne soit lqa son q6tat normal ... Encore une fois, labsence de rq6fe'- rences ne nous permet pas de nous prononcer avec certitude. En tout e'tat de cause, qlle'volution de la me'iofaune en milieu pollue' par les hydrocarqbures est donc lie'e essentiellement 'a 11oxy- genation du sq6diment et, par consq6quent, a l1intensitq6 de 1'hydrody- namisme. Une I'veille e'cologiquell devrait permettre de mieux apprecier 11impact de cette marq6e noire, de pre'ciser les dq6lais de retour a' 11q6tat qdle'quilibre initial au sein des biotopes pollue's et, d1une maniq@re ge'nq6rale, de mieux comprendre les perturbations des e'cosys- temes susceptibles detre poqlluq6s dans le futur. La valeur d1une approche par une 'etude des e'cosyste'mes dans leur ensemble nlest plus a dq6montrer pour 1'q6tude des effets des pollutions sur l1environne- ment (Linden et al., 1979). Il est souhaitable que les q6tudes entreprises, tant au niveau de la macrofaune que de la me'iofaune et du microphytobenthos, puis- sent q@q@tre poursuivies encore plusieurs anne'es avec, en paralle'le, un suivi des paraqmq@tres physico-chimiques et sq6dimentologiques. REFERENCES CITEES Admiraal, W., 1977, Experiments with mixed populations of benthic estuarine Diatoms in laboratory microsystems : Botanica Marina, Vol. 20 (8), pp. 479-486 Admiraal, W. et H. Peletier, 1980, Influence of seasonal variations of temperature and light on the growth rate of cultures and na- tural populations of intertidal Diatoms : Mar. Ecol. - Prog. Ser. 2, pp. 35-43 Amspoker, M.C., 1977, The distribution of intertidal epipsammic Dia- toms on Scripps Beach, La Jolla, California, USA : Botanica Marina, vol. 20 (4), pp. 227-232 Arlt, Gq., 1975, Remarks on indicator organisms (meiofauna) in the coastal waters of the GDR : (Merentutkimuslait.) Julk./Havsforsk- ningqsinst. Skr., vol. 239, pp. 272-279 264 Bellan, G., 1967, Pollution et peuplements benthiques de substrats meubles dans la re'gion de Marseille. le're partie : Le secteur de Cortiou : Rev. int. Oceanogr. Med., Vol. 6-7, pp. 53-87 Bodin, P., 1977, Les peuplements de Cop6podes Harpacticoldes (Crus- tacea) des s6diments meubles de la zone intertidale des c8tes charentaises (Atlantique) :M6m. Mus. Nat. Hist. nat., Paris, ser. A, Zool., Vol. 104, pp. 1-120 Bodin, P. et D. Boucher, 1981, Evolution temporelle du me'iobenthos et du microphytobenthos sur quelques plages touch6es par la mar6e noire de l'Amoco-Cadiz : In "Amoco-Cadiz, Conse'quences d'une pollution accidentelle-par les hydrocarbures", Actes du Colloque International, Brest 19-22 novembre 1979, CNEXO Ed., pp. 327-345 Boucher, D., 1977, Production primaire saisonnie're du microphyto- benthos des sables envas6s en baie de Concarneau : In "Bio- logy of benthic organisms", B.F. Keegan, P.O. Ceidigh and P.J.S. Boaden Eds., Pergamon Press, N.Y., pp. 85-92 Boucher, G., S. Chamroux et C. Riaux, 1981, Etude d1impact e6colo- gique de la pollution pe'trolie're de ll"Amoco-Cadiz" dans la region de Roscoff et de la baie de Morlaix. "Effets ! long terme sur la structure des 6cosyst@mes sedimentaires" : Rapp. Contrat CNEXO / Univ. Paris VI, nO 79/5973, 51 pp. Cad6e, G.C. et J. Hegeman, 1974, Primary production of the benthic mi- croflora living on tidal flats in the Dutch Wadden Sea. Neth. J. Sea Res., Vol. 8 (2-3), pp. 260-291 Cad6e, G.C. et J. Hegeman, 1977, Distribution of primary production of the benthic microflora and accumulation of organic matter on a ti- dal flat area, Balgzand, Dutch Wadden Sea. Neth. J. Sea Res., 11 (1), pp. 24-41 Chasse', C. et M. Gle'marec, 1976, Principes ge'ne'raux de la classifi- cation des fonds pour la cartographie biose'dimentaire : J. Rech. Oce'anogr., Vol. 1 (3), pp. 1-18 Colijn, F. et G. Van Buurt, 1975, Influence of light and temperature on the photosynthetic rate of marine benthic Diatoms : Mar. Bioi., Vol. 31 (3), pp. 209- 214 Colijn, F. et K.S. Dijkema, 1981, Species composition of benthic Diatoms and distribution of chlorophyll a on an intertidal flat in the Dutch Wadden Sea : Mar. Ecol. Prog., Ser. 4, pp. 9-21 Coull, B.C. et S.S. Bell, 1979, Perspectives of marine meiofauna ecology : In "Ecological processes in coastal and marine sys- tems", Mar. Sci., Vol. 10, pp. 189-216 Frithsen, J.B. et R. Elmgren, 1979, The response of benthic meiofau- na in experimental microcosm to chronic levels of n02 fuel oil. 42e Congr@s annuel de 1'"American Society of Limnology and Oceanography Inc.", 18-21 juin 1979, New-York 265 Gargas, E., 1970, Measurement of primary production dark fixation and vertical distribution of the microbenthic algae in the Oresund Ophelia,vol. 8, pp. 231-253 Giere, 0., 1979, The impact of oil pollution on intertidal meiofauna. Field studies after the La Coruna-spill, May 1976 : Cah. Biol. Mar., vol. 20, pp. 231-252 Gle'marec,,M. et C. Hily, 1981, Perturbations apportees 'a la macro- faune benthique de la baie de Concarneau par les effluents ur- bains et portuaires : Acta Oecologica, Oecol. Applic., vol. 3, pp. 139-150 Gray, J.S., 1971, The effects of pollution on sand meiofauna commu- nities : Thal. Jugoslav., vol. 7 (1), pp. 79-86 Hartwig, E.O., 1978, Factors affecting respiration and photosynthesis by the benthic community of a subtidal siliceous sediment : Mar. Biol., vol. 46 (4), pp. 283-294 Hicks, G.R.F., 1977, Species composition and.zoogeography of marine phytal harpacticoid copepods from Cook Strait, and their contri- bution of total phytal Theiofauna : N.Z. Jour. Mar. Fresh. Res., vol. 11 (3), pp. 441-469 Hunding, C., 1971, Production of benthic microalgae in the littoral zone of a eutrophic lake : Oikos, vol. 22 (3), pp. 389-397 Le Moal, Y., 1981, Ecologie dynamique des plages touche'es par la ma- ree noire de 1"'Amoco-Cadiz" : Th@se 3e Cycle, Univ. Bretagne Occidentale Le Moal, Y. et M. Quillien-Monot, 1981, Etude des populations de la macrofaune et de leurs juveniles sur les plages des Abers Benoit et Wrac'h : In "Amoco-Cadiz, Conse'quences d1une pollution acci- dentelle par les hydrocarbures", Actes du Colloque internatio- nal, Brest, 19-22 novembre 1979, CNEXO Ed., pp. 311-326 Linden, 0., E.L. Elmgren et P. Boehm, 1979, The Tsesis oil spil : its impact on the coastal ecosystem of the Baltic Sea : Ambio, pp. 244-253 Lorenzen, C.J., 1967, Determination of chlorophyll@and phaeopigments spectrophotometric equations : Limnol. Oceanog., vol. 12, pp. 343-346 Mac Intyre, A.D., A.S. Munro et J.H. Steele, 1970, Energy flow in a sand ecosystem : In "Marine Foodchains", Oliver and Boyd Ed., Edinburgh, pp. 19-31 Plante-Cuny, M.R., T. Le Campion-Alsumard et E. Vacelet, 1981, Influence de la pollution due A l'I'Amoco-Cadiz" sur les peuple- ments bact6riens et microphytiques des marais maritimes de 1'Ile Grande. 2.- Peuplements microphytiques : In "Amoco-Cadiz, Cons6quences d'une pollution accidentelle par les hydrocarbures" Actes du Colloque International, Brest, 19-22 novembre 1979, CNEXO Ed., pp. 429-442 266 Renaud-Mornant, J. et N. Gourbault, 1980, Survie de la meiofaune apres 11e'chouement de l'I'Amoco-Cadiz" (chenal de Morlaix, gre've de Roscoff) : Bull. Mus. natn. Hist. nat., Paris, 4e se'r., Vol. 2 (3), pp. 759-772 Steele, J.H. et I.E. Baird, 1968, Production, ecology of a sandy beach : Liqmnol. Oceanog., vol. 12, pp. 14-25 Sundbqdck, K. et L.E. Persson, 1981, The effect of microbenthic gra- zing by an amphipod Bathyporeia piqZosa, Lindstq6ra : Kiel. Meer., S, pp. 573-575 Wormald, A.P., 1976, Effects of a spill qof marine diesel oil on the meiofauna of a sandy beach at Picnic Bay, Hongkong Environ. Pollut., vol. 11, pp. 117-130 Ce travail a e6tq6 en partie re'alisq6 grq9ce au Contrat NOAA/CNEXO nO 79/6184. 267. LONG-TERM IMPACT OF THE AMOCO CADIZ CRUDE OIL SPILL ON OYSTERS Crassostrea gigas AND PLAICE PZeuronectes platessa FROM ABER BENOIT AND ABER WRAC'H, BRITTANY, FRANCE I. OYSTER HISTOPATHOLOGY II. PETROLEUM CONTAMINATION AND BIOCHEMICAL INDICES OF STRESS IN OYSTERS AND PLAICE by Jerry M. NeffI and William E. Haensly2 1) Battelle New England Marine Research Laboratory, Washington Street, Duxbury, MA 02332, USA 2) Texas A&M University, Department of Vetinary Anatomy, College Station, TX 77843, USA INTRODUCTION On the evening of 16 March 1978, the.Liberian-registered super- tanker Amoco Cadiz (233,680 tons deadweight) ran aground and subse- quently broke up on Men Goulven rock, Roches de Portsall, approximately 2 km off Portsall on the Breton coast of France. Over a period of several days the complete cargo of the supertanker, which consisted of 120,000 metric tons of light Iranian crude oil, 100,000 tons of light Arabian crude oil and 4,000 tons of bunker fuel was spilled into the coastal waters. By mid April the oil had spread to and contaminated in varying degrees 375 km of the north and west coasts,of Brittany (Hess, 1978; Spooner, 1978; Southward, 1978). At the time, it was the largest oil spill in maritime history. There have been two larger spills since then. Two estuaries in the heavily impacted area, l'Aber Benoit 6 km east of the spill and l'Aber Wrac'h 9 km east of the spill, face west and became heavily contaminated with spilled oil. Aber Benoit and Aber Wrac'h are biologically rich and before the spill.supported large oyster mariculture operations and other commercial fisheries4@ It was therefore of considerable economic and hygenic impor- tance to accurately assess the progress of the long-term recovery of the estuarine biota from the impact of the oil spill. Several factors relating to this spill, including the large volume of oil spilled, the prevailing winds and currents which drove-much of the oil ashore, adverse weather conditions and large tidal prisms which resulted in the incorporation of large amounts of oil into bottom sedi- ments, and the extreme biological richness of the impacted area, all 269 conspired to create a "worst case." scenario for marine oil pollution. Therefore, the Amoco Cadiz spill offered a unique opportunity to study in detail the long-term impact and timecourse of biological recovery from a catastrophic pollution incident. While we already know that the immediate biological effects of the spill were very serioua in some areas (Cross et al., 1978; Chasse, 1978; Chasse and Morvan, 1978), there was very little information upon which to base estimates of the rate at which the impacted area would be returned to pre-spill biological productivity. We have used several biochemical parameters and histopathological examination in an ongoing biological survey to assess the health and rate of recovery of marine animals from the two heavily polluted estuaries. The primary objective of this research program was to assess the degree of chronic sublethal pollutant stress experienced by representa- tive species of benthic fauna from Aber Benoit and Aber Wrac'h. Two indices of stress were used. These are histopathology and biochemical composition. We expected the fauna of these severely'impacted estuaries to exhibit an elevated incidence of various histopathological lesions directly or indirectly related to oil pollution stress. As the estuaries recovered from the spill the incidence of these lesions was expected to diminish. Similarly, the concentrations of certain diagnostic biochemical components of the severely stressed fauna were expected to deviate sig- nificantly from normal. These diagnostic biochemical indices were expected to return to normal as the estuaries recovered and the resident fauna became less severely stressed. The results of this investigation provide valuable information for assessing the biological recovery of these severely polluted estuaries. They also provide a means of diagnos- ing.pollutant stress in other polluted environments. 1. Histopathology of Oysters Crassostrea gigas Marine animals readily accumulate petroleum hydrocarbons in their tissues from dispersion or solution in sea water and to a lesser extent from petroleum-contaminated sediments and food (see recent reviews by Neff et al., 1976 a,b; Lee, 1977; Varanasi and Malins, 1977; Neff, 1979; Neff and Anderson, 1981). The accumulated hydrocarbons and in particular the more toxic aromatic hydrocarbons interact with cellular membranes and interfere with membrane-mediated biological processes (Roubal and Collier, 1975). Two types of histopathological lesions may result from chronic contamination of marine animals with oil. 270 The first type is due to the direct toxic effects of petroleum hydrocarbons and associated heavy metals on cells. These compounds may produce a variety of histopathological lesions in the affected organ systems. There are several reports that exposure to sublethal concentrations of oil in laboratory or field studies resulted in epi- thelial sloughing and discharge of mucus glands in the gills of teleost fish (Blanton and Robinson, 1973; Gardner, 1975; Hawkes, 1977; McKeown and March, 1978). McCain et al. (1978) reported severe hepatocellular lipid vacuolization in English sole Parophrys ventulus following exposure for four months to experimentally oiled (Alaskan North Slope crude oil) sediments. Rainbow trout fed Prudhoe Bay crude oil-contaminated food showed several histopathological changes in the liver (Hawkes, 1977). These included glycogen depletion, proliferation of the endoplasmic reticulum and focal necrosis with connective tissue infiltration in necrotic regions. We have described a wide variety of histopathological lesions to embryos and fry of the killifish FunduZus heteroclitus exposed chronically during embryonic development to the water-soluble fraction of No. 2 fuel oil (Ernst et al., 1977). In a recent laboratory study of the effects of water soluble fractions of crude oil on marine fish, one of us (Eure'll and Haensly, 1981) observed a variety of histopatho- logic changes in liver and gill tissues. . Little research has been published on the histopathological effects of petroleum in benthic marine invertebrates. However lesions similar to those described in fish can be expected in the analogous organs of marine invertebrates. Although crude oil contains known carcinogens such as benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene, petroleum-induced cancer has not been unequivocally demonstrated in any marine species (Neff, 1979). However, there are several reports of increased incidence of cancer-like lesions in natural populations of marine invertebrates and fish from hydrocarbon polluted sites (See recent symposium volumes edited by Dawe et al., 1976 and Kraybill et al., 1977). The second type of histopathological lesion resulting from chronic exposure to sublethal concentrations of oil is caused by elevated suscep- tibility of contaminated animals to bacterial, virus.or parasite infection. This increased susceptibility may result from damage to protective epi- thelia in the affected animals or to deleterious effects of the pollutant hydrocarbons on the immune system of the animal (Hodgins et al., 1977; Sinderman, 1979). Marine animals which have been subjected to chronic sublethal oil pollution stress can be expected to exhibit an elevated incidence of disease in comparison to non-contaminated animals. 271 MATERIALS AND METHODS Oysters Crassostrea gigas were collected during five sampling trips to France. Dates of these trips were December 1978, April 1979, July- August 1979, February 1980, and June-July 1980. In Aber Benoit, oysters were obtained from commercial oyster parc owners in St. Pabu and Prat Ar Coum. Oysters from Aber Wrach were obtained from a commercial opera- tion near Paluden. Aber Benoit oysters were not available in August .1979. Reference oysters were obtained from several places. None were completely uncontaminated with oil. On the first two trips, December 1978 and April 1979, the oyster parc operator at St. Pabu had oysters from the Rade de Brest (supposedly uncontaminated) which he was holding for later sale. We used these as reference oysters. Subsequent hydro- carbon analysis revealed that these oysters were as heavily contaminated with petroleum as Aber Benoit oysters. They had probably become contam- inated during brief holding in the contaminated water of the Aber, as Michel and Grizel (1979) subsequently showed in transplant experiments. On the third trip, August 1979, reference oysters were obtained from the CNEXO mariculture field station at Ile Tudy. On the fourth and fifth trips, February 1980 and June 1980, reference oysters were obtained from a commercial oyster parc owner on the Rade de Brest at Plougastel. As soon as possible after collection, the oysters were shucked and the soft tissues fixed whole in freshly prepared Helly's fixative. The visceral mass was incised to insure rapid penetration of the fixative. After fixation the oysters were wash8d, dissected into several organs or body regions, dehydrated in ethyl alcohol and embedded in paraffin embedding medium. Organ systems processed for histopathological examination included: visceral mass (includes digestive tract, digestive gland, kidney and gonad), gill, and mantle. Sections were cut a 6 pm with a rotary microtome and stained with hematoxylin-eosin. All tissue blocks and prepared microslides of oyster tissues were labeled, inventoried and archived. Tissue sections were evaluated qualitatively. The qualitative pro- cedures included a description of the average and limits of normal for the histological status of each tissue. All histopathological lesions were described in full. The incidence of different types of lesions in each tissue was recorded. The incidence of different types of lesions in each tissue was recorded. These data for the three populations (2 oil-contaminated stations and one control station) were compared. Seasonal and temporal differences in the incidence of pathological lesions were also recorded. A photographic record of normal tissue histology and of all types of histopathological lesions was made and archived. 272 REsnTS Tissues from@34 specimens of Crassostrea gigas from four sites were examined for histopathologies over five sampling trips. From the speci- mens collected, tissue samples of!31 adductor muscles,127 stomach/ intestines,129 digestive glands,130 gonads,134 gills, andl30 mantles were examined for a total of 781 tissues out of a possible 804. A total of nine types of pathologies were found with an incidence of 241 occurrences (Table 1). Five-hundred and ninety of the 781 (75.6%) tissues examined were free of pathologies; or, 191.of the 781 (24.3%) tissues examined bore one or more pathologies. Of the 241 pathologies found, 77 (32.0%) were various types of symbioses, while 16.4 (68.0%) cases apparently were not correlated with symbioses. Table 2 summarizes the distribution of pathologies among the tissues examined. Adductor muscle had the lowest incidence (3.8%). Digestive gland tissue had the highest incidence (23.9%) followed by gill (22.0%), mantle (21.4%), gut (17%), and gonad (11.9%). The number of tissues with pathologies was nearly evenly distributed among the collecting sites. Oysters from reference stations had a higher incidence of lesions, particularly in gonad and gill, than oysters from oil-polluted sites. Thirty percent of the oyster tissues from both Aber Wrac'h and Aber Benoit bore one or more pathologies. Forty percent of the tissues from Rade de Brest and Ile Tudy combined contained one or more pathologies. Overall, mantle bore the lowest number of pathology types (3) while digestive gland contained the most types of pathologies (9) followed by gut (7), gill (6), gonad (5), and muscle (4). Pathologies and their distributions among organs and sites are described below. 1. Muscle. - Muscle tissues were examined from 131 C. gigas. Samples for microscopic examination were dissected from the ad(iuctor muscle and both fast and catch muscles were examined when possible. Generally, two tissue samples were taken from each muscle and oriented to give both longitudinal and cross sections. Histopathologies occurred in 4.6% (6 of 131) of the muscle samples examined. There were a total of 9 incidences of.the three pathologies described below. Muscle from reference stations contained the widest variety of pathologies. No pathologies were found in muscles from Aber Benoit. 273 Table 1. Types of pathologies, total incidence of each, affected organ and collecting site of occurrence Pathology Incidence Organ* Site+ Amoebae 3 DGMA C Ciliates 21 GU,DG,GI W,B,C Sporozoans 29 MU,GU,DG,G0,GI,MA W,B,C Copepods 23 GU,DG,GI B,C Nematodes 1 DG w Degeneration 10 MU,GU,DG,GO W'c Necrosis 9 GU,DG,GO W'B'C General leucocytosis 93 MU,GU,DG,G0,GI,MA W,B,C Focal leucocytosis 52 MU,GU,DG,G0,GI,MA W,B,C Total 241 MU - Muscle GU - Gut DG - Digestive gland GO - Gonad GI - Gill MA - Mantle + C - Control (Rade de Brest and Ile Tudy) W - Aber Wrac'h B - Aber Benoit 274 Table E. Distribution of patholo@jes in tissues of oysters Crassostred gigas from two oil contaminated estuaries and from reference stations, with sampling times combined, Orqan Digestive Station Muscle Gut Gland Gonad Gill Plantle Total Reference 5 17 20 15 21 18 96 Aber Benoit 0 18 19 6 14 17 74 Aber Wrac'h 5 6 1@ 7 18 17 71 Total 10 41 57 28 53 52 241 275 Abnormally high numbers of eosinophilic leucocytes (general leucocytosis) were apparent in 1.5% (2 of 131) of the adductor muscle samples exmained. Leucocytes were generally spread throughout the muscle rather than being in focal aggregations. Aggregated eosinophilic leucocytes were present in I of 131 adductor muscles examined. For the purposes of this report, this aggregation was classified as a focal leucocytosis although there was no central core or tight concentric arrangement of leucocytes as reported from Crassostrea virginica (Armstrong et al., 1980). This may be an inflammatory response to what appears to be a foreign body, possibly a nematode, at the edge of t@e aggregation. Five (3.8%) of the muscles examined contained areas of degenerated muscle bundles. This condition was characterized by a breakdown or liquefaction of the cellular integrity. Degenerated areas contained amorphous, light staining debris and fibers. No pyknotic nuclei were present in surrounding whole muscle fibers and no inflammation (increased number of leucocytes) was apparent. Unidentified sporozoans in the plasmodial stage were found in I of the 131 muscles examined. 2. Digestive Gland. - The digestive glands of 129 C. gigas were examined. Generally, two sample6 were taken from each specimen at differ- ent levels (anterior and posterior) of the digestive gland. Histopathologies were noted in 44.2% (57 of 129) of the digestive gland samples examined. There were a total of 64 incidences of the 9 types of pathologies described below. Twenty-seven of these or 42.2% apparently were not attributable to symbioses, while 37 (57.8%) were a type of symbiont or were clearly attributable to symbioses (i.e. inflam- mation). The distribution of these histopathologies among sampling sites is summarized in Table 3. Digestive gland samples from Aber Benoit con- tained more pathologies than samples.from the other two sites. All digestive gland samples from the December, 1978 collection at Aber Benoit bore one or more pathologies. Samples from other sites over the five collections had no more than 62% incidence of pathologies. Abnormally high numbers of eosinophilic leucocytes were dispersed throughout the leydig tissue between diverticula in 5 (3.7%) of the 129 samples examined. In some, leucocytes were also invading the diverti- cular epithelium. These cases could have been inflammatory responses to parasites such as copepods which were not included in the sectioned material. That is, the sections could be at the edge of an inflammatory response as described below. 276 Table 3. Distribution of histopathologies of C. gigas digestive gland among sampling sites. Numbers in parentheses represent the number of specimens examined. Site Pathology Aber Wrac'h (45) Aber Benoit (42) Rade de Brest & Ile Tudy (42) Total (129) Leucocytosis 1 2 2 5 (general) Leucocytosis 8 7 1 16 (focal) Degeneration 1 - - 1 Focal Necrosis 1 4 5 Amoebae - - 1 1 Ciliates 4 6 6 16 Sporozoa 2 7 7 16 Nematode 1 - - I Copepoda - 3 - 3 Total 18 29 17 64 Aggregates of eosinophilic leucocytes, were present in 12.4% (16 of 129) of the digestive glands examined. Almost all cases were in specimens from Aber Wrac'h or Aber Benoit (Table 3). For the purpose of this study, these were termed focal leucocytoses. They differed.from a general leuco- cytosis in that the leucocytes were in a dense clump, sometimes focal, rather than being dispersed throughout the tissues. General leucocytosis may possibly, in some cases, be a part of a focal inflammation viewed some distance from the focal foreign body or parasite. Some cases of focal leucocytos.is appeared to be confined to the leydig tissue surround- ing the diverticulae and were not totally "focal". -In most cases, however, the condition involved mass invasion of the lumina by leucocytes and/or phagocytes with large numbers of leucocytes and/or. phagocytes massed in the surrounding leydig tissue. Decomposed portions of copepods were present in the lumina of two specimens and no doubt were responsible for the mass inflammation. Copepods were not apparent in the leucocytic inflammations in the digestive glands of the other specimens. These inflammations, or focal leucocytoses, may also have been responses to copepods as they were identical in all aspects except for the observed presence of copepods in the section. In one case, well-formed focal aggregates were present in the leydig tissue adjacent to the digestive gland. In one, the leucocytes were con- fined to a well-formed "pocket", while in another the leucocytes were also spread from the "pocket" to adjacent leydig tissue. A massive pocket of leucocytes was present in one of the digestive glands examined. Leucocytes were confined to the large 11pocket". Adjacent leydig cells were compressed. A degeneration.of two or three diverticula was observed in one digestive gland and was associated with a copepod parasite. This involved a breakdown of the diverticular epithelia and basal membranes with leucocytic inflammation. Five (3.9%) of the samples examined bore small necrotic areas on one to four diverticula. These areas were characterized by a breakdown of cellular integrity accompanied by light staining cellular debris and a limited number of leucocytes. Necrosis appeared to be minor. Amoebae were present in the digestive gland of one C. gigas. Digestive glands of 16 (12.4%) of the C. gigas examined contained ciliates. Ciliates were evenly distributed among Aber Wrac'h, Aber Benoit and Rade de Brest oysters and were sometimes quite numerous in the diverticular lumina. Ciliates were oblong, with a somewhat pointed 278 antenai, and longitudinal spiral rows of short, stout cilia. They did not appear to damage the diverticula. Sporozoa were present in the diverticular epithelium of 16 (12.4%) of the specimens examined. All but two of the cases were from Aber Benoit or Rade de Brest and Ile Tudy (Table 3). Sporozoans were spher- ical and stained very intensely. They were often surrounded by a clear (lysed '7,) zone. The digestive gland of one specimen bore a nematode which elicited an inflammatory response, an aggregation of.eosinophilic leucocytes. Remains of copepods were noted in the diverticula of 3 C. gigas from Aber Benoit. They were accompanied by heavy leucocytic inflammation and were.be-ing phagocytized as -evidenced by the presence of leucocytes in the copepods. 3. Gut.. - Samples consisting of stomach, intestine and often esoph- ageal and rectal tissues were examined from 127 C. gigas. Generally, two tissue samples were taken from each,specimen (anterior and posterior portions of the visceral mass). Histopathologies were noted in 32.2% (41 of 127) of the gut samples examined. There were 62 cases of the 6 pathology types discussed below. Thirty-five percent (22) involved a symbiont while 40 cases (65%) apparently were not symbiotic in nature, although@there may be some question about this. Specimens from the combined reference stations bore more than two and a half times the pathologies as Aber Wrac'h specimens. Oysters from Aber Benoit contained slightly fewer pathologies than'those from the reference stations and twice the number of pathologies as speci- mens from Aber Wrach. Abnormally high numbers of eosinophilic leucocytes (general leuco- cytosis) were noted in the intestinal epithelium, and sometimes surround- ing leydig tissue, of 15% (19 of 127) of the C. gigas examined. Almost all cases were in oysters from Rade de Brest. This condition was diffi- cult to judge. Oyster intestinal epithelium normally has some leucocytes between columnar cells. However, the large number of leucocytes in the intestinal epithelium of these 14 specimens appeared abnormally high. thenumber was considered abnormally high if the basal portion of columnar cells was completely, or almost completely, obscured by leucocytes. However, there still is some doubt about whether-this is a "pathology" or a normal condition. Except for the large number of leucocytes, the intestinal tissues appeared very healthy. 279 Focal aggregates of leucocytes. were present in the intestinal epi- thelium, or adjacent to it, in 14.2% (8 of 127) of the gut tissues examined. One case involved a large, loose aggregation of leucocytes in the leydig tissue beneath the basement membrane. Another involved small clumps of leucocytes between the columnar epithelial cells. Like the general leucocytosis, this condition was difficult to judge. Leuco- cytes are normally present in the epithelium, but more or less scattered about. These clumps could be normal phagocytosis, although no foreign matter was ever observed in such clumps. The intestinal epithelium containing the above clumps appeared otherwise very healthy. Focal necrotic areas were present in the gut epithelium of 3 (2.3%) oysters examined. In two incidences, the gastric shield was involved. This condition was characterized by a breakdown of the structure of the gastric shield and/or epithelium, a concentration of debris at the affected area, and leucocytic inflammation ofthe gastric shield and/or epithelium. Ciliates were present in the gut lumen of one C. gigas., These ciliates were the same type as described above in the digestive gland. The plasmodial stage of an unidentified sporozoan was noted in the epithelium of a single C. gigas. The plasmodium was amoeboid in appear- ance with several nuclei. The gut otherwise appeared in very good condition. Copepods were present.in the stomach of 15.7% (20 of 127) of the specimens examined. No oysters from Aber Wrac'h bore copepods. None of ,the copepods observed were being phagocytized as was the case in the digestive gland. Up to three copepods were observed in some sections. 4. Gonad. - Gonadal tissues of 130 C. gigas were examined. Gener- ally, two tissue samples were taken from each.specimen (anterior and psoterior visceral mass). Gonadal tissues from C. gigas were a very difficult tissue type to assess for non-symbiotic pathologies. Possible histopathologies were noted in 38% (9 of 130) of the gonadal tissues examined. There were 50 incidences of the three non-symbiotic pathology (?) types and one symbi- otic pathology discussed below. Only six of the 50 conditions were of an apparent symbiotic nature. Half (36 of 71) of the female gonadal tissues examined exhibited moderate to heavy aggregations, both focal.and general, of eosinophilic leucocytes. This presented a perplexing problem in determining if this 280 represented an inflammatory response to a stressful condition and therefore a pathology due to such stress, or if it was a normal condi- tion in the reproductive cycle of C. gigas. This condition was present in only one Cras-sostrea virgin,@ca from South Louisiana oil platforms (Armstrong et al..1- 1980) but was observed in other bivalve species (10% of specimens examined in association with degeneration or necrosis of the gonad). Eight female C. gigas from the Pacific Northwest (Sequim, Washington) were examined for comparison. All eight appeared to be in a post-spawn condition and all had heavy aggregation of leucocytes in the gonadal tissues. The spawning cycle of the C. gigas from France could not be defin- itely determined. Undifferentiated (could not determine if it was male or female), undeveloped, developing (immature), ripe (mature) and spawned stages were present in samples from all five of the collecting periods (December 1978, April 1979, August 1979, February 1980, and June 1980). The majority of the specimens from December 1978, however, appeared to be of the spawned stage at Aber Wrac'h, ripe at Aber Benoit, and undiffer- entiated at Rade de Brest. In April 1979, the majority of the specimens appeared to be in the developing stage at all three sites. The majority of the specimens taken during August 1979 and June 1980 appeared to be of the ripe stage at all three sites, although there were some spawned- appearing.specimens from Aber Wrac'h in August 1979. The February, 1980 collection yielded more undifferentiated and developing specimens. This does somewhat indicate an early winter spawn, but as already stated, all reproductive stages were present in samples from all five collection periods. In the C. gigas from France, 13.8% (18 of 130) of the gonads examined contained large numbers of leucocytes dispersed throughout the tissues. All 18 incidences were in female gonads (18 of 76 or 23.8%). This condition was present in undeveloped, developing (immature), ripe and post spawn ovaries. In some cases it could not be determined if the ovary was in a developing stage or a post-spawn stage because of the large numbers of leucocytes present. In some, the gonad appeared fully spawned (entire gonad examined contained only a few ova and ovacytes, follicles largely empty), while in others part of the ovary was packed with ova (ripe) and the other part contained few ova and ovacytes (spawned) and many leucocytes. In gonads with large number 's of leuco-, cytes, all or almost all ova appeared normal (not degenerating or lysing). The 29 normal ovaries (no aggregations of leucocytes present) included the undeveloped, developing (immature), ripe and spawned stages. Twenty-two of the 130 (16.6%) gonads examined contained compact clumps of leucocytes ranging from foci in the follicular wall to large 281 clumps in the ovary to foci in the leydig tissue of the testes. Nine- teen (86.4%) of the cases were in female gonads. Reproductive stages varied from undeveloped to spawned. Three of the cases were found in testes. In the general and focal leucocytoses discussed above, the differ- ence between the two was in the extent (small area, tight clump vs. general dispersion over several follicles) of inflammation, but this was sometimes difficult to ascertain and the two may blend together. Necrotic appearing areas were noted in 3.1% (4 of 130) of the gonads examined. These areas were characterized by cellular debris, degenerating ova, and leucocytosis. Sporozoa were present in 4.6% (6 of laO) of the specimens examined. Sporozoans were spherical, densely staining., and were embedded in the gonadal tissue. The specimens appeared to be surrounded by a small lysed "halo" area. 5. Gill. - Gills from 134 C. gigas were examined for pathologies. Generally, threepieces of gill.(consisting of both lamellae) were dissected from one side and oriented (when possible) to give both longi- tudinal and transverse sections. Histopathologies were noted in 31.3% (42 of 134) of the gills examined. There were a total of 49 cases of the six pathology types described below. Forty-three (87.8%) were apparently not symbiotic or related to a symbiotic condition. Abnormally high numbers of eosinophilic leucocytes were present in 31.3% (42 of 134) of the gills.examined. In most incidences, the leuco- cytes were dispersed throughout several plica, but four cases appeared to be more focally organized in one or two plica. Amoebae were noted in the gills of one C. gigas. The infection appeared to be light as only two amoebae were found. The amoebae were circular in outline-with a hyaline cytoplasm. The nucleus occupied approximately one-third of the cell. A smalL, spherical inclusion body was adjacent to the nucleus. The gills of four (3%) specimens examined harbored ciliates in their water tubules. Ciliates were somewhat crescent-shaped with tufts of stout cilia extending downward from the two tips. The arms of the crescent were sometimes turned inward so that the tips of the cilia were touching, giving a partially hollow, circular shape to the ciliate. Two 282 lateral nuclei were present. The ciliates apparently provoked an inflammatory response as most were surrounded by eosinophilic leuco- cytes in the water tubules, or the surrounding tissues contained abnormally high numbers of lucocytes. One specimen contained the plasmodial stage of a sporozoan. Multi- nucleate plasmodia were subsp herical to ovate. Numerous plasmodia were dispersed throughout the gill, but most heavily in the leydig tissue of the interlamellar area. A single copepod was found on a gill filament of one C. gigas. No necrotic areas were observed on the gills examined and the outer columnar epithelium of the specimens examined appeared healthy. The number of mucous glands in sections of randomly selected-plica and term- inal grooves were counted in an effort to determine if specimens from Aber Wrac'h and Aber Benoit contained more active glands than those from Rade de Brest and Ile Tudy. The results were inconclusive. The number of mucus cells per unit area of gill was not statistically significantly different among the three populations. 6. Mantle. - Sections of mantle from 130 specimens were examined. Two or three pieces of mantle were dissected from specimens and oriented to give a transverse section across the tri-lobed edge. Histopathologies were noted in the mantle of 31.5% (41 of 130) of the specimens examined. There were a total of 47 of the three pathology types described below. The distribution of the pathologies among sampling sites was nearly equal. Abnormally high numbers of eosinophilic leucocytes were noted in 35.4% (46 of 130) of the specimens examined. Thirty-eight of the inci- dences involved large numbers of leucocytes dispersed beneath the epi- thelium or in the leydig tissue. Eight cases, however, involved leuco- cytes which were more aggregated in clusters. Sporozoans were found in the mantle of a single specimen. No necrotic areas were found on any of the mantles examined. All epithelial cells appeared healthy. In an effort to determine if the mantle epithelium of oysters from Aber Wrac'h and Aber Benoit contained significantly more mucous cells than specimens from Rade de Brest and Ile Tudy, the number of mucous cells in a high power field were counted at a level even with the circumpallial nerve and,an area three fields higher. Specimens from Aber Wrac'h contained slightly more (average of 27.5 to 34 for the five collections) mucous cells than those from Aber 283 Benoit (average of 20 to 32) and Rade de Brest and Ile Tudy (average of 22 to 31). The differences were not statistically significant. CONCLUSIONS In general, oysters Crassoatrea gigas from all five collections and all four sampling stations appeared to be extremely healthy as determined by histopathological examination. Incidence of parasitic infestation was very low, especially when compared to incidence of para- sitism in-C. virginica from the northwest Gulf of Mexico. The low incidence of parasitism in C. gigas from Brittany may be due to the fact that they are a recently-introduced maricultur6 species in the area. There probably has not been enough time for their parasites to catch up with them. According to Henri Grizell (personal communication)., parasitism and disease are increasing in these oysters. The most prevalent pathologic lesion in C.'gigas from Brittany was leucocytosis. In mollsucs, this condition is usually a response to chemical or physical irritation. It is an inflammatory defensive response. However, size and distribution of leucocyte populations varies greatly in different mollusc species under different environmental conditions. C. gigas generally seems to have more leucocytes than the closely-related C. virginica. Thus, the extent to which observed leucocytoses in C. gigas were normal or pathologic is uncertain. In any event, incidence of leucocytosis was similar in oysters from oil-contaminated Aber Benoit and Aber Wrac'h and from reference stations.in the Rade de Brest and at Ile Tudy. Necrosis was observed.several times but no definitive cases of hyperplasia, neoplasia or other precancerous conditions was noted in any of the four oyster populations. There were no consistent temporal trends in incidence of pathology in the oysters-from oiled and reference stations. Oysters collected in December 1978, nine months after the spill, had an incidence of patho- logical conditions similar to that in oysters collected in June 1980, twenty-seven months after the spill. one difference that may have obscured other effects was size. By June 1980, oysters which had been in the Abers at the time of the spill had grown to very large size. During the first year after'the spill, there was little evidence of growth in oys,ters from the two Abers. During the second year, growth appeared normal or even accelerated. 284 There was also some indication, based on observations of gonadal condition, that oysters from the Abers had an altered reproductive cycle compared to reference oysters, possibly including near complete reproduc- tive suppression for one year after the spill. Sample sizes and frequen- cres. were not great enough to demonstrate this convincingly. II. Petroleum CQntamination and Biochemical Indices of Stress in Oysters and Plaice The most obvious immediate biological effect of the Amoco Cadiz spill was a very large kill.of benthic estuarine and coastal marine organisms (Cross et al., 1978). The rate of recovery of these benthic communities would depend on the rate and success of reproduction by the surviving animals in the affected area and on the success of recruitment from adjacent unpolluted*areas. The resident benthic fauna in the oil- impacted area which survived the spill were undoubtedly severely stressed. Because of the heavy contamination of the estuarine 5ediments with oil it is highly probable that the surviving resident benthic fauna would continue for some time to be stressed and potential immigrants to the estuaries would be subjected to stress as they settled there. Considerable research has been conducted in recent years on sub- lethal physiological stress responses of marine animals to oil and other types of pollution (Neff et al., 1976a; Anderson., 1977; Johnson, 1977; Patten, 1977; Neff, 1979; Thomas et al., 1980; Neff and Anderson, 1981). A variety of sublethal physiological and biochemical responses to pollutant stress have been described. In an ecological perspective, the net effect of chronic pollutant stress on marine organisms is to shunt limited energy resources away from growth and reproductive processes to maintenance and homeostatic functions. The result is decreased growth, fecundity and reproductive success in the stressed population. A variety of biochemical parameters are altered in stressed animals and reflect the stress-induced changes in energy balance and partitioning. These biochemical parameters can be used as an index of pollutant stress in marine animals. Biochemical indices of pollutant stress chosen for use in this investigation include hemolymph glucose concentration and adductor muscle-free amino acids in oysters; and blood glucose and cholesterol, liver glycogen and ascQrbic acid, and muscle-free amino acids in plaice. We have discussed elsewhere the rationale for using these parameters as indices of pollutant stress (Thomas et al., 1986, 1981 a,b). 285 When exposed to petroleum, marine molluscs and teleost fish readily accumulate hydrocarbons in their tis'sues (Neff et al., 1976b; Varanasi andMalins, 1977; Neff and Anderson, 1981). Molluscs tend to release accumulated hydrocarbons relatively-slowly,when concentrations in the ambientmedium are reduced., However, under similar conditions, teleost fish release hydrocarbons very rapidly. Differences in hydrocarbon release rate by molluscs-and fish.can be attributed to differences in ability to convert hydrocarbQns to polar more readily excreted metabo- lites by the cytochrome P-450 mixed function oxygenase system and related pollutant-metabolizing enzyme systems (Varanasi and Malins, 1977; Neff, 1979). In the present investigation, aliphatic and aromatic hydrocarbons were analyzed in oysters and plaice from oiled and reference stations to assess patterns of hydrocarbon accumulation and release and to allow for correlations between levels. of hydrocarbon contamination of animals and histopathological/biochemical responses. MATERIALS AND METHODS I Oysters Crassostrea gqgas-were collected for biochemical analysis on the first three sampling trips. Sampling sites were as described earlier in the section on oyster histopathology. Oysters were shucked and a sample of hemolymph was collected immediately from the heart or the adductor muscle and stored frozen until analyzed. Adductor muscle was also sampled and stored at -60*C untilanalyzed. Plaice@PZeuronectes platessa were collected by otter trawl from oil-contaminated Aber Benoit and Aber Wrac'h. Reference stations for plaice samples were as follows: December 1978, Baie de Douarnenez; April 1979, Loc Tudy; August 1979, February 1980, June 1980, Ile Tudy. Fish from the Baie de Douarnenez and Loc Tudy were collected by otter or beam trawl. Fish from Ile Tudy were captured by net at the sluice gate of the CNEXO mariculture pond and held in large circular holding tanks with flowing seawater until sampled. Samples were taken as soon as possible after capture and while the fish were still alive. Tissue samples included blood, muscle and liver. Blood samples were centrifuged to remove red blood cells. Serum, muscle, and liver were frozen immediately in liquid nitrogen and kept frozen at -609 until analy-zed. Samples from 5@10 animals from each station and each trip were analyzed biochemically. Blood glucose and liver glycogen were measured with a Yellow Springs Instruments automatic glucose analyzer, Model 23A. 286 This. method, based on the glucose oxidase enzymatic reaction, is highly specific for glucose And required only 25 pl of serum. Replicate deter- mi.nations, of each serum sample were performed. Total and esterified cholesterol in serum was deterr4ined by a cholesterol oxidase assay system which is both highly snesitive and specific. for tissue-free amino acid analysis, muscle tissue was thawed, weighed and homogeni.zed in distilled water using a 2/1 ratio of distilled water/wet weight. Homogenates were deproteinized with 12.5% trichloro- acetic Acid and then centrifuged. The supernates were frozen, thawed, and centrifuged again to remove additional TCA precipitates. The super- nates were then evaporated to dryness on a rotary evaporator and the res-idue dis-solved in 0.2 M Citrate buffer adjusted to pH 2.2. The extracts were analyzed with a Beckman automatic amino acid analyzer. The amino acid composition of the extract and the concentration of individual amino acids in it were determined. Taurine/glycine molar ratios were computed. Variations in amino acid compositions and concen- trations among fish and oysters from different sampling stations were analyzed statistically. Plaice liver was analyzed for ascorbic acid. Tissue samples were thawedi weighed and homogenized in 3% metaphosphoric acid-8% acetic acid solution. After centrifugation, the supernates were analyzed immediately by the a,a-diperidyl technique of Zannoni et al. (1974). Oysters and plaice samples for hydrocarbon analysis were taken at the same times and places as samples for biochemical/histopathological analysis. Ten to twelve whole oysters were pooled for each sample. They were shucked and tissues were rinsed in distilled water, blotted dry, wrapped in hexane-cleaned aluminum foil and frozen at -60*C until analysis. For the April 1979 sample, whole fish were used. For sub- sequent samples, pooled samples of liver and muscle from 5-10 fish were used. Fish tissue samples were handled like oyster samples. Hydrocarbon analyses were performed by Dr. Paul Boehm, ERCO, Cambridge, Massachusetts using capillary gas chromatography/mass spectrometry. 287 RESULTS AND DI$CUSSIQN Petroleum Hydrocarbons Concentrations of total aliphatic.and Aromatic hydrocarbons in tissues.of oysters.Crassos-trea gigas from Aber Benoit and Aber Wrac'h, heavily cQntaminated with Amoco Cadiz oil, and from supposedly clean reference stations are summarized in Table 4. Reference oyaters for the first two collections were Rade de Brest oysters which had been held for a-short period of time in concrete holding tanks on the shore of Aber Benoit at St. Pabu. These reference oysters were heavily contaminated with Amoco Cadl;z oil as were the authentic Aber Benoit and Aber Wrac'h oysters. "Hydrocarbon status" of samples was determined by comparing GC peak profiles of fl and f2 hydrocarbon fractions of tissue extracts to GC profiles of authentic weathered Amoco Cadiz oil. Apparently, sufficient oil was still leaching from the sediments of the Aber 13 months after the spill to allow rapid and heavy contaiftination of oysters exposed to waters of the bay. Michel and Grizel (1979) reported similar rapid hydrocarbon contamination of oysters transplanted to stations in Aber Benoit and Aber Wrac'h. Subsequent reference oyster samples were obtained from sites which had not received Amoco Cadiz oil. They con- tained low levels of petroleum hydrocarbons not of Amoco Cadiz origin. Concentrations of total aliphatic and aromatic hydrocarbons in oysters from Aber Benoit and A 'ber Wrac'h did not vary substantially over the time-course of this. investigation (up to 27 months after the spill). The persistence of petroleum hydrocarbons in tissues of oysters probably represents, in part, a continuous recontamination with hydrocarbons leaching gradually into the water from the heavily contaminated sediments of the Abers. Oysters from the Baie of Morlaix, east of Aber Wrac'h and less heavily contaminated with Amoco Cadiz oil than the Abers, collected 17 months after the spill, contained about half the aromatic hydrocarbons of Aber Wrac'h oysters. It is interesting to note that Aber Benoit oysters collected in December 1978 and April 1979 had a distinctly oily taste. Oysters sampled in August 1979 and later did not taste oily. Apparently, 200 ppm aromatics is not readily detected by taste, whereas 500 ppm is. More detailed analysis of the aliphatic fraction of the oyster samples revealed some interesting trends (Tables 5-7). In all but one case (Aber Wrac'h, April 1979), the aliphatic fraction of Aber Benoit and Aber Wrac'h oysters was dominated by the low boiling,aliphatics, C10 - C20, including n-alkanes, branched and isoprenoid compounds. This is quite unlike weathered Amoco Cadiz oil or oil in the Aber 288 Table 4 - Concentrations of total aliphatic and aromatic hydrocarbons (measurcd gravimetrically) in oysters Crarsoctrea gigas from reference stations and from two estuaries contaminated with Amoco Cadiz oil. Status determined according to pattern and identity of GC peaks. Hydrocarbon Fraction (pg/g dry tissue) a Date/Sample Aliphatics Aromatics Status December 1978 (9) b Rade de Brest (reference) 47.8 208.0 AC oil Aber Benoit 136.7 552.2 AC oil Aber Wrac'h 115.4 540.0 AC oil April 1979 (13) Rade de Brest (reference) 153.9 1001.0 AC oil Aber Benoit 114.9 690.0 AC oil Aber Wrac'h 225.8 986.1 AC oil August 1979 (17) Ile Tudy (reference) 39.3 5i.9 Other oil Baie de Morlaix 134.6 206.4 Other oil Aber Wrac'h 101.0 485.4 AC oil February 1980 (23) Rade de Brest (reference) 62 87 Other oil Aber Benoit 154 275 AC oil Aber Wrac'h 217 599 AC oil June 1980 (27) :--',Rade de Brest (reference) 33 60 Other oil Aber Benoit 238 283 AC oil/Other oil Aber Wrac'h 132 430 AC oil a, AC oil - Amoco Cadiz oil; other oil - definitely petroleum, but cannot be identified as Amoco Cadiz oiZ. b, month after the Amoco Cadiz oil spill, 16 March 1978. 289 Table 5 Concentration of aliphatic hydrocarbons in tissues of oysters C'miscoctr. a oiqar from Aber Benoit, Brittany France collected at different times after the Amoco CadiL oil spill, Values are in n9/9 dry weight (parts per billion). Sample Date Dec 1978 Apr 1979 Aug 1979 Feb 1980 Jun@ 1930 Compound (,)a (13) (17) (23) (27) n-C 10 NO 32 14S NO NO n'C 11 316 328 NS 162 NO n-C 12 43 47 NO 44 NO n-C 13 99 31 NS 7 NO n-C 14 394 18 NS 16 20 Farnesane 1,343 776 NS 66 92 n-C 15 22 84 NS 68 10 n-C 16 44 46 NS 62 62 n'C 17 184 61 NS 144 22 Pristane 376 232 NS 40 22 n C18 NO NO NS 51 NO Phytane 613 375 NS 39 122 n-C 19 47 68 NS 17 42 n-C 20 77 57 NO 7 nC21 ND NO NS 18 49 nC22 NO NO NS 10 53 nC23 24 NO NS 20 57 nC24 43 NO NS 21 53 nC25 55 NO NS 21 17 n C26 51 14D NS 9 24 n-C 27 53 NO NS 37 52 n C28 37 NO NS 12 11 n-C 29 72 NO NS 18 66 nC30 NO NO NS 74 343 nC31 NO NO NS 13 27 nC32 ND NO NS 12 134 nC33 NO NO NS NO 49 n C34 NO NO NS NIO 12 Total Resolved Ali- 3,893 2,155 NS 981 1,346 phatics a, months after the Amoco Cadiz oil spill, 16 March 1978 NO, not detected NS, no sample available. 290 Table 6 Concentration of aliphatic hydrocarbons in tissues of oysters Cran.w@@trea g-@*,7ac frow Aber l1rac'h. Brittany France collected at different times after the Amoco CaJir oil spill. Values are in nqIg dry weight (parts per billion). Sampling Date Dec 1978 Apr 1979 Aug 1979 Feb 1980 Jun 1980 Compound (,)a (33) (17) (23) (27) n C10 46 NO 110 47 84 n-C 11 383 55 759 170 255 nC12 86 365 ND 126 63 nC13- 39 70 32 NO is n C14 36 650 12 519 35 Farnesane 222 617 953 370 148 n-C 15 13 NO 172 56 13 n C16 53 NO 098 198 31 n-C 17 37 177 577 218 140 Prista ne 319 52 39 44 7B n C18 NO NO NO 27 nD Phytane 571 242 141 194 64 n-C 19 187 NO NO 12 IID n-C 20 74 NO NO 12 NO n-C 21 ND 140 ND NO NO nC22 NO ND NO 140 NO nC23 NO 30 NO 141 NO nC24 15 135 NO NO RD nC25 15 222 NO 19 12 nC26 14 289 NO 27 NO n C27 11 280 NO 19 NO n*C 28 NO 241 NO 23 140 n-C 29 NO 219 ND 18 NO n-C 3D ND 132 NO 16 NO nC31 NO NO ND NO 140 n-C 32 NO NO NO NO Total Resolved Ali- 2,121 3,776 2,893 2,256 798 phatics , a. months after the Amoco Cadiz Oil spill, 16 March 1978 NO, not detected. 291 Table 7 Concentration of aliphatic hydrocarbons in tissues of oysters from reference station,, on the Brittany coast of France collected at different tillps after the eto,o c_.%;: oil spill. Value.@ are in ng/9 dry weight (parts per billion). Sampling Date Compound Dec 1978a Apr 1979a Aug 1979b Feb 1930c Jun 1930C - (9) d (13) (17) (23) (27) nC10 53 NO NO 66 226 n C11 360 441 NO 339 486 n C12 15 49 NO 86 112 n C13 17 ill NO 18 9 n C14 52 29 NO 33 18 Farnesane 11 1,346 4 45 35 n C15 172 123 29 122 66 n C16 139 55 17 69 41 n C17 252 228 61 81 46 Pristane 18 331 NO NO NO n Cis 39 ND NO 58 45 Phytane 117 529 17 NO 13 nC19 40 35 NO 22 28 n C20 43 73 NO 20 24 nC21 42 20 NO 16 22 nC22 40 47 ND 15 23 n C23 42 85 21 16 36 nC24 49 135 29 15 45 n C25 50 173 33 16 47 n C26 56 204 44 15 55 nC27 207 54 20 60 n C28 64 167 36 16 48 n C29 66 165 64 29 47 nC30 22 100 54 so 42 nC31 NO 71 59 13 22 n C32 ND NO NO NO 8 Total Resolved Ali- 1,824 4,724 522 1,200 1,604 Phatics a, from Rade de Brest, but maintained in Aber Benoit before sampling b, from oyster wariculture ponds of CNEX0 at Ile Tudy C, from a commercial oyster parc in the Rade de Brest d, months after the Amoco Cadiz oil spill, 16 March 1978 NO, not dectected. 292 sediments.which is dominated by higher boiling saturated hydrQcarbons. This phenomenon is@ unexplai'ned and could-represent selectiye accum,ula- tion and/or retention of lighter ali.phatics or more rapid metabolism and excretiQ,n qf.heavier aliphati,cq.,. The ost likely explanation is that qysters were being cpntami4ated with hydrocarbons. leaching from bottom sediments, into the water column. Lighter aliphatics, because of, their slightly higher aqueous solub.ility than heavy aliphatics, are desorbed more readilyfrom sediments and therefore aremore available for uptake by the oysters,., Aliphatic hydrocarbon fractions of reference oysters were more uniform (Table 7). Relative abundances of C10 to C32 aliphatics were similar. There were no consistent differences in characteristics of the ali- phatic hydrocarbon fraction between reference oysters and oysters from oil-polluted Aber Benoit and Aber Wrac1h (Table 8). With one exception (April 1979.), alkane/isoprenoid ratios were higher in oysters from reference stations than in those from oil-polluted stations. Pristane/ phytane ratios were quite variable and without pattern. All but two carbon preference indices were near one indicating a petroleum origin for the high molecular weight aliphatic fraction. Composition of the aromatic fraction of oysters, as determined by gas chromatography/mass spectrometry., revealed a great deal about the origin of the hydrocarbon contamination of the oysters (Tables 9-11). High concentrations of alkyl naphthalenes through alkyl dibenzothiophenes are characteristic of samples contaminated with crude oil. Amoco Cadiz oil was particularly rich in alkyl phenanthrenes and alkyl dibenzothio- phenes. These were the most abundant aromatics/heterocyclics in oyster samples from Qil-contaminated Aber Benoit and Aber Wrac'h. Aromatic hydrocarbon assemblages of crude oil origin are dominated by alkylated species, whereas aromatic assemblages of pyrogenic origin are dominated by the unalkylated parent compound (Neff, 1979). Thus we can conclude that oysters from Aber Benoit and Aber Wrach at all five sampling times, and reference oysters from the December 1978 and April 1979 collections were heavily contaminated with crude oil, resembling the Amoco Cadiz oil. The other three reference samples contained some oil, but it did not resemble, Amoco Cadiz oil. In oysters from the two Abers, there was a general trend for the concentration of aromatics/heterocyclics in the alkyl naphthalenes to alkyl dibenzothiQphenes series to decrease slowly with time. The February 1980 samples contained higher concentra- tions. of alkyl phenanthrenes and alkyl dibenzothiophenes than expected. It is pos-sible that winter storms in December and January resuspended oil-contaminated sediments causing recontamination of resident oysters. 293 Table 8 . Characteristics of the aliphatic hydrocarbon fraction of oysters Craosostrea gigas-from reference stations and from two estuaries contaminated with Amoco Cadiz oil Carbon Prefer- Date/Sample Pristane/Phytane Alkanes/Isoprinoids ence Index (C 26-C30) December 1980(9) Rade de Brest (reference) 0.15 2.34 1.27 Aber Benoit 0.61 0.18 2.0 Aber Wrac'h 0.56 0.07 1.57 April 1979(13) Rade de Brest (reference) 0.63 0.12 1.16 Aber Benoit 0.62 0.13 ND Aber Wrac'h 0.21 1.00 1.10 August 1979(17) Ile Tudy (reference) ND 5.15 1.39 Baie de Morlaix (reference) ND ND 0.93 Aber Benoit NS NS NS Aber Wrac'h 0.28 0:48 ND February 1980(23) Rade de Brest (reference) NO 4.37 1.00 Aber Benoit ND 1.30 1.02 Aber Wrac'h 0.23 0.81 ND June 1980(27) Rade de Brest (reference) ND 2.64 1.10 Aber Benoit 0.18 0.28 0.61 Aber Wrac'h 1.68 0.19 0.93 294 Table 9 Concentration of aromatic hydrocarbons in tissues of oysters Crassostrea g@gas from Aber Benoit, Brittany, France at differ- ent times after the Amoco Cadi,: oil spill. Values are in ng/g tissue (parts per billion). Sampling Date Dec 1978 Apr 1979 Aug 1979 Feb 1980 Jun 1980 Compound (,)a (13) (17) (23) (27) Alkyl naphthalenes NA 1,243 NS ND 300 Alkyl fluorenes NA 2,230 NS 891 850 Phenanthrene NA 590 NS 43 64 Alkyl phenanthrenes NA 17,345 NS 6,088 3,014 Dibenzothiophene NA 123 NS NO . ND Alkyl dibenzothiophenes NA 15,380 NS 8,860 5,420 Fluoranthene NA 665 NS 150 84 Pyrene NA 600 NS 150 87 Benz[a]anthracene NA 263 NS 200 NO Chrysene NA 490 NS 600 180 Benzofluoranthenes NA S70 NS 670 100 Benzopyrenes NA 339 NS 413 80 Perylene NA 80 NS ND NO Total Resolved Aromatics NA 39,918 NS 18,065 10,179 a, months after the Amoco Cadiz oil spill, 16 March 1978. NA, sample not analyzed by GC/MS NS, no sample available NO, not detected Table 10. Concentration of aromatic hydrocarbons in tissues of oysters Crassostrea gigas from Aber Wrac'h, Brittany, France at differ- ent times after the Amoco Cadiz oil spill. Values are in ng/g tissue (parts per billion). Sampling Date Dec 1978 Apr 1979 Aug 1979 Feb 1980 Jun 1980 Compound (,)a (13) (17) (23) (27) Alkyl naphthalenes 781 594 150 ND 10 Alkyl fluorenes 2,203 1,453 980 560 ND Phenanthrene 89 69 ND ND 170 Alkyl phenanthrenes 12,114 14,989 5,030 10,089 4,550 Dibenzothiophene 24 ND ND ND ND Alkyl dibenzothiophenes 21,748 11,521 9,900 15,820 5,530 Fluoranthene 258 58 50 150 70 Pyrene 291 105 65 190 90 Benz[a]anthracene 557 330 ND 1,100 63 Chrysene 300 230 Benzofluoranthenes ND 237 260 410 190 Benzopyrenes ND 161 50 170 .140 Perylene ND ND ND ND 50 Total Resolved Aromatics 38,065 29,517 16,785 28,489 11,093 a, months after the Amoco Cadiz oil spill, 16 March 1978; ND, Not detected. 296 Table 11 Concentration of aromatic hydrocarbons in tissues of oysters Crassostrea qiqas from "reference" stations on the Brittany coast of France at different sampling times after the Amoco Cadiz oil spill. Values are in ng/g tissue (parts per billion). Sampling Date Dec 1978a Apr 1979a Aug 1979b Feb 1980c Jun 1980c Compound (9)d (13) (17) (23) (27) Alkyl naphthalenes 327 467 ND ND 180 Alkyl fluorenes 689 1,562 ND ND 180 Phenanthrene 129 85 5 180 350 Alkyl phenanthrenes 4,375 10,679 285 527 630 Dibenzothiophene 30 56 ND ND 20 Alkyl dibenzothiophenes 3,668 10,590 283 975 675 Fluoranthene 220 171 130 180 200 Pyrene 170 98 130 180 90 Benz(a]anthracene ND 1,40 58 Chrysene 252 290 410 350 170 Benzofluoranthenes 88 48 350 410 160 Benzopyrenes 64 65 140 182 83 Perylenes ND ND ND ND NO Total Resolved Aromatics 10,012 24,111 1,733 3,124 2,796 a, from Rade de Brest, but maintained in Aber Benoit before sampling b, from oyster mariculture ponds of CNEXO at Ile Tudy C' from corrmercial oyster parc in the Rade de Brest d, months after the Amoco Cadiz oil spill, 16 March 1978 ND, not detected. 297 Higher molecular weight aromatics fluoranthene through perylene, although present in amal 1 amounts in crude oil, are.more characteristic of py:pQgenic hydrocarbon assemblages (Neff, 1979). Concentrations of these aromatics, were simillar in reference and Aber oysters and there was no consistent pattern of temporal change, These hydrocarbons probably, hAve 4 similar origin in all three populations, namely from particulate organicmatter derived from smoke of woodand fossil fuel combustion. $everal of these aromatics, including benz[a]anthracene, benzofluoranthenes, and benzopyrenes-, are known carcinogens. Thpir presence in tissues of oysters at relatively high concentration could be cause for concern. Whole fish and muscle samples of plaice Pteuronectes p,Zatessa contained low concentrations of aliphatic and aromatic hydrocarbons (Table 12). Most of the muscle samples contained aliphatic hydrocarbon 4istributions. characteristic of oil (Tables 1:3-15). Nearly tenfold higher concentrations of aliphatics were found in liver samples than in muscle samples of reference plaice and plaice f ram the oil-polluted Abers. In the August 1979 samples, some of this was identified as petroleum. In later samples, no petroleum-derived hydrocarbons were detected in liver samples. The aromatic fraction showed a distribution pattern similar to that of the aliphatic fraction. Liver aromatic fractions were dominated by biogenic squalene. Liver samples also contained high concentrations of what appeared to be naphthehic (cyclic alkanes) hydrocarbons. Aliphatic fractions from all liver samples were dominated by hydro- carbons in the C21 - C32 molecular weight range. In the three liver samples from Aber Benoit, two of the three samples from Aber Wrach, and. one reference sample, dominant aliphatics were C 27 and C29- In the remaining two samples, dominant-aliphatics were C25 and C29' With few exceptions light aliphatics, C10 - C20, were present at low or non- detectable concentrations in the plaice livers. Plaice muscle contained 1-10% of the concentration of aliphatics that liver did. Alkane distribution patterns in muscles varied consid- erably. In most cases alkanes above C24 were dominant. Concentrations of aliphatic hydrocarbons in muscle and liver were higher in summer (August 1979 and june 1980) than in winter (February 1980), suggesting a seaonal cycle of tissue hydrocarbon concentration. This seasonal pattern was, not correlated with seasonal changes in total lipid content of plaice tissues (Table 18). As in the oysters., there was no consistent difference between reference plaice and plaice from oil-contaminated Aber Benoit and Aber Wrac'h with respect to pristane/phytane ratio, alkane/ isoprenoid ratio, or carbon preference index (Table 16). 298 Table 12. Concentrations of total aliphatic and aromatic hydrocarbons (measured gravimetrically) in tissues of plaice PZcuronectes plate-ssa from reference stations and from two estuaries con- taminated with Amoco Cadiz oil. Status determined according to pattern and identify of GC peaks. Hydrocarbon Fraction (@wjg/g d y tissue) a Date/Sample Aliphatics Aromatics Status Aeril 1979 (13) b Whole Fish Loc Tudy (reference) 2.9 91 Biogenic Aber Benoit 38.0 24 Biogenic Aber Wrac'h 7.1 83 Small U.C.M. August 1979 (17) Muscle Aber Benoit 7.7 9.0 Other oil/biogenic Aber Wrac'h 19.9 12.7 Other oil/biogenic Liver Aber Benoit 801.9 235.6 Other oil/biogenic Aber Wrac'h 1034.0 317.9 Other oil/biogenic February 1980 (23) Muscle Ile Tudy (reference) 23 .19 Other oil/biogenic Aber Benoit 83 22 Other oil/biogenic Aber Wrac'h 66 12 Other oil/biogenic Liver Ile Tudy (reference) 736 548 Biogenic Aber Benoit 1510 352 Biogenic Aber Wrac'h 831 355 Biogenic June 1980 (27) Muscle Ile Tudy (reference) 16 23 Biogenic Aber Benoit 38 17 Other oil/biogenic Aber Wrac'h 146 41 Other oil Liver Ile Tudy (reference) 1210 723 Biogenic Aber Benoit 1810 682 Biogenic Aber Wrac'h 1130 511 Biogenic a, biogenic - probably of biological origin; small U.C.M. small unresolved complex mixture, typical of weathered oil; other oil - definitely petroleum but cannot be identified as Amoco Cadiz oil. b, months after the Amoco Cadiz oil spill, 16 March 1978. 299 Table 13. Concentration of aliphatic hydrocarbons in tissues of plaice Plciiro,wcte@: plate-,:a from Aber Benoit, Brittany france. collected at different tivnes after the Arioca Czzdi@@ oil spill. Values are in ng/9 dry weight (parts per billion). Date/Sample Compound Apr 1979(13) Aug 1979(17) Feb 1980(23) Jun 1980(27) Whole Fish Muscle Liver Muscle Liver Huscle Liver A-C 10 ND RD ND NO RD RD NO n-C 11 3 15 NO 6 531 NO 208 n-C 12 NO RD ND RD 324 NO RD n-C 13 No RD RD RD No 4 NO n-C 14 NO ND NO NO 14D 8 NO Farnesane ND RD NO NO RD RD RD n-C 15 3 6 RD 13 NO 8 140 n-C 16 3 12 NO 14 132 8 NO n-C 17 9 44 NO 31 336 13 RD Pristane 18 12 HD 16 RD 5 2,530 n-C 18 4 47 ND .21 299 13 NO Phytane 31 2D ND 219 NO 5 RD n-C0 3 24 RD is 331 11 RD n-C 20 6 13 NO 29 459 22 RD n-C 21 4 9 555 39 425 50 1,130 n'C 22 3 10 1,927 44 321 120 3,470 n-C 23 3 13 3,695 51 206 221 6,460 n-C 24 2 15 5,109 61 152 324 9.030 n-C 25 4 21 6,476 78 452 436 10,500 n-C 26 7 23 8,315 92 2,530 496 13,700 n-C 27 15 31 14,363 95 6,660 503 19,600 n-C 28 a 28 10,552 94 4.100 421 15,200 n-C 29 22 42 14,282 87 6,770 359 23,500 n-C 30 ND 43 4,573 57 750 299 9,040 n-E 31 10 43 4,425 '99 1,430 197 9,150 n-C 32 NO 36 1,414 20 RD 117 3,280 n-C 33 ND NO NO NO NO 86 1110 n-C 34 A-0 -LD NO No NO 51 NO Total Resolved Ali- 158 507 76,586 991 26,208 3,767 126,798 phatics a, ths after the Amooo Cadiz oil spill, 16 March 1978. 300 Table 14. Concentration of aliphatic hydrocarbons in tissue of plaice lllciaviicutc@z nZntc--@a from Aber Wrac'h, Brittany France collocted at different times after the A@oeo oil spill. Values are in ng/9 dry weight (parts per billion). Date/Saniple Apr 1979(13 )a Aug 1979(17) Feb 1980(23) Jun 1980(27) Compound Whole Fish Muscle Liver Muscle Liver t-luscle Liver n-C 10 NO 16 NO NO 562 16 152 n-C 11 NO 105 NO NO 943 46 NO n-C 12 ND NO NO NO 155 16 ND n-C 13 NO NO NO NO NO NO HD n-C 14 NO NO NO NO NO HD NO Farnesane NO NO ND NO NO NO NO nC15 140 6 NO 6 ND ND No n C16 NO a NO 11 NO 9 RD n-C 17 8 16 NO 27 ND 28 151 Pristane 10 NO NO 5 370 12 ND n-C Is 3 7 NO 19 NO 22 NO Phytane 15 ND NO S' HD 17 NO n-C19 14D NO NO 9 NO 6 NO n-C 20 NO NO NO NO ND 23 ND nC21 3 NO NO 8 164 45 1,040 n C22 3 NO 14D 7 NO 91 3,410 CC 23 2 9 NO 7 46 162 6,460 n C24 2 14 207 6 14D 223 9,150 n-C 25 3 '23 4,722 6 56 335 10,600 n-C 26 4 34 1,467 7 379 328 16,300 n-C 27 7 64 3,781 7 1,100 280 20,100 nC28 4 76 3,883 7 857 255 16,600 nCZ9 6 147 6,625 8 2,480 266 21,700 nC30 NO 189 3,032 5 599 254 10,800 nC31 NO 237 1,747 4 912 so 10,900 nC32 NO 241 NO NO 134 100 3,130 nC33 NO 237 NO NO NO 18 2,050 n C34 @'D 176 NO .10 ND 17 5$6 _L_ Total Resolved Ali- 70 1,605 25,464 157 8,914 2,649 133,129 phatics months after the A@oco Cadiz oil spill, 16 March 1978. 301 .Table 15 Concentration of aliphatic hydrocarbons in tissues of plaice Pluw@oncctes platc-u;a from reference stations on the Brittany coast of France at different times after the A@,oao Crdi_ oil spill. Values are irt ng/9 dry weight (parts per billion). Date/Sample. Apr 1979(13)a Feb 1980(23) Jun 1980 (27) Compound Whole Fish Muscle Liver Muscle Liver n-t . NO 17 99 37 231 NO 44 275 75 618 NO il 53 17 130 NO I NO NO NO 43 1%n% NO 3 NO 1 NO .F.'1f4 arpesane NO 4 AD NO NO C' 11 ND 7 NO 2 16 NO 8 NO 5@ 32 NO 16 NO 10 110 3 NO P'r Vstlarie 3 nt 4 21 NO 18 NO r., it: 18 ND 6 NO 3 8 NO 7 NO 3 7 80 9 187 n''a 2 8 302 17 1,740 21 2 7 283 41 5,510 22 "'@523 2 9 412 76 1.060 ni@c 24 1 10 571 110 14,500 wCis 2 7 771 141 23,600 C26 2 12 1,150 158 20.000 _:CV.1 3 12 2,380 158 21,400 @!CCjg 2 10 2,060 134 16,200 C 6 9 5,270 112 24.200 Z9 C NO 5 1,320 82 10,500 30 2n-C 31 1 5 3,730 58 12,500 .n-C32 NO NO 749 34 482 nC33 NO NO NO HD 402 ID n C34 NO ND NO N 1,160 Total Resolved Ali- 45 297 19,505 1,325 154,420 phatics months after the Amoco C4diz oil spill, 16 March 1978. 302 Table 16. Characteristics of the aliphatic hydrocarbon fraction of plaice Plc@4ronectcs pZatessa from reference stations and from two estuaries contaminated with Amooo Cadiz oil. Carbon Preference Date/Sample Pristane/Phytane Alkanes/Isoprenoids Index (C - C 26- .30) April 1979(13 )a Whole Fish Loc Tudy (reference 1.40 2.34 3.72 Aber Benoit 0.59 0.39 3.11 Aber Wrac'h 0.66 0.45 2.41 August 1979(17) Muscle Aber Benoit 0.61 3.38 1.19 Aber Wrac'h ND 6.14 1.12 Liver Aber Benoit 14D ND 1-64. Aber Wrac'h ND ND 1.16 February 1980(23) Muscle Ile Tudy (reference) 0.58 2.37 1.13 Aber Benoit 0.55 1.77 1.11- Aber Wrac'h 0.64 4.54 1 . 27 Liver Ile Tudy (reference) ND ND 2.32 Aber Benoit ND ND Aber Wrac'h ND ND 2.65 June 1980(27) Muscle Ile Tudy (reference) 0.56 5.22 1.07 Aber Benoit 1.00 5.00 1.06 Aber Wrac'h 0.68 2.07 1.06 Liver Ile Tudy (reference) ND ND 1.-45 Aber Benoit ND ND 1.62 Aber Wrac'h HD ND 1.39 a, months after the Amoco Cadiz oil spill, 16 March 1978. 303 The hydrocarbon data demonstrate convincingly the dramatic differences in p4tterna of petroleum hydrocarbon contamination of oysters and plaice from the s-ame oil-contaminated Abers. Oysters contained high concentrations of alka.nes, dominated by low molecular weight com ounds., while in plaice, the dominant alkanes in liver p samples were the higher molecular weight compounds. Oysters contained abundant petrogenic and pyrogenic aromatic hydrocarbons. spanning a wide molecular weight range. Plaice on the other hand contained little true aromatic hydrocarbon. These differences undoubtedly reflect the markedly different capabilities of bivalve molluscs and teleost fish to metabolize and actively excrete petroleum hydrocarbons. Most teleosts studied to date have a highly active and inducible cytochrome P-450 mixed function oxygenase system capable of converting aromatics and some aliphatics to polar and-more easily excreted matabolites (Neff, 1979). This enzyme system is absent altogether or present at very low activity in bivalve mollusc tissues. Biochemical Indices of Stress Total lipid concentration in tissues of oysters and plaice, deter- mined in connection with hydrocarbon analyses, showed no consistent patterns in relation to station or season (Tables 17-18). In June 1980, but not at other sampling times-, oysters from the two oil-contaminated Abers contained 2-3 times as much lipid as oysters from the reference station. It is. quite possible that this is related to differences between reference and Aber oysters in state of reproductive ripeness, and not directly to oil-induced effects. Hemolymph glucose concentrations in oysters were low, highly variable, and showed no relationship to station (Table 19). No statis- tically significant differences were noted in values for reference and Aber oysters. There was a trend at all stations toward increasing hemo- lymph glucose concentration between December 1978 and August 1979. Some patterns did emerge in serum glucose concentrations of plaice (Table 20). In December 1978, April 1979 and August 1979, with one exception, serpER glucose concentrations of plaice from oil-contaminated Aber Benoit and Aber Wrac'h were lower than values for reference plaice. Two of these differences were &tatistically significant. The collecting technique (otter trawl) is.highly stressful, and-maximal hyperglycemic stress response occurs rapidly in fish (Thomas et al., 1980). The data suggest, not that Aber plaice were less stressed than reference plaice, but that they had become refractory--perhaps due to chronic stress--to capture-induced hyperglycemia. Inability to respond bio- chemically to stress has been demonstrated in plaice held in the 304 Table 17 Concentration of total lipids (determined gravimetrically) in whole oysters Crassostrea gigas from reference stations and from estuaries contaminated by Amoco Cadi;:: oil. Values are in pg'/g dry tissue. Station August 1979 February 1980 June 1980 Reference 9,775 6,650 5,580 Aber Benoit NS 9,150 15,900 Aber Wrac'h 6,151 4,860 11,200 Bale de Morlaix 6,188 NS NS NS, no-sample available. Table. 18. Concentration of total lipids (determined gravimetrically) in tissues of plaice (Pleuronectes pZatessa) from reference stations and from two estuaries contaminated by Amoco Cadiz oil. Values are in pg/g dry tissue. Station Tissue August 1979 February 1980 June 1980 Reference Muscle NS 2,310 1,870 Liver NS 11,300 8,770 Aber Benoit Muscle 1,921 2,170 1,740 Liver 16,729 129700 11,200 Aber Wrac'h Muscle 3,278 1,750 1,810 Liver 14,725 20,300 4,380 NS, no sample available. 305 Table 19. Hemolymph glucose concentration in oysters Cr=;ostrea gigas from reference stations and from oil-polluted Aber Benoit and Aber Wrac'h. Values and standard deviations are in mg glucose/ 100 ml hemolymph. n = 8 replicates. Sampling Date Station Dec 1978(g)a April 1979(13) Aug 1979(17) Reference 5.12 + 3.1 13.05 + 2.9 23.53 + 4.0 Aber Benoit 3.00 + 1.8 12.57 + 2.4 NS Aber Wrac'h 4.80 + 2.6 11.25 + 3.8 23.87 + 2.8 NS, no sample analyzed a, months after the Amoco Cadiz oil spill, 16 March 1978. Table 20. Serum glucose concentration in plaice PZeuronectes- pZatessa from reference stations and from oil-polluted Aber Benoit and Aber Wrac'h. n = 10 replicates. Values and standard deviations are in mg glucose/100 nil serum. Sampling Date Station Dec 1978(9)a Apr 1979(13) Aug 1979(17) Feb 1980(23) Jun 1980(27) Reference 158.1 + 11.6 149.6 + 23.5 160.4 + 36.9 -27.1 + 19.2 37.2 + 12.8 Aber Benoit 118.3 + 32.9 57.0 + 32.9* 93.7 + 27.3* 85.9 + 33.2* 147.3 + 46.0* Aber Wrac'h NS 125.6 + 19.5 168.4 + 31.3 135.0 + 28.7* 135.0 + 55.4* significantly different from reference at a = 0.05 NS, no sample analyzed a, months after the Amoco Cadiz oil spill, 16 March 1978. 306 laboratory (Wardle, 1972). In the last two samples, February 1980 and-June 1980, reference plaice were sampled very rapidly after capture and their blood glucose values represent the normal unstressed values. Plai.ce fromthe &ber!i were s.tressed by capture and showed a hyper- glycemic response, suggesting some recovery of physiological function .wi.th-time. These data show some-of the difficulties in using blood glucose concentration as an index of stress in fish. If blood samples cannot be taken immediately after the fish are captured, capture-induced responses-mayobscure any-due to pollution. Liver glycogen concentrations in fish from the last two collections were highly variable (Table 21). Because of extremely large standard deviations., no patterns could be discerned. Total cholesterol and high density lipoprotein (HDL) cholesterol concentrations in the blood of plaice were measured in sampels from the last two collecting trips (Table 22). The general trend was for total cholesterol to be elevated and HDL cholesterol concentration to be depressed in fish from the two oil-contaminated Abers. Several of these differences were statistically significant. As a result, HDL cholesterol as percent of total cholesterol was lower in plaice from Aber Benoit and Aber Wraclh than in plaice from the reference station at Ile Tudy. Concentration of liver"free ascorbic acid was measured in plaice from all five sampling trips (Table 23). In all but the February 1980 sample, liver ascorbate concentrations in plaice from oil-contaminated Aber Benoit and Aber Wrac'h were substantially lower than concentrations in livers of plaice from reference stations. In four cases, the differ- ence was. statistically significant. In the February collection, the pattern was reversed. Reference fish con tained hepatic ascorbate concen- trations significantly lo r than con we centrations in livers of fish from the two Abers. At this time, all the reference fish were gravid females ready to spawn. Onlya few of the fish from the Abers were in this condi- tion. It is highly likely that the extreme depletion of liver ascorbate reserves in the reference fish is the result of ascorbate mobilization for gonadal maturation and ovogenesis. These gravid reference fish also had relatively low hepatic glycogen reserves (Table 21). AdductQr muscle-free amino acid profiles and concentrations were measured in oysters from the first three collecting trips (Tables 24-26). Total free amino acid concentrations were always lower in adductor muscles of oysters from oil-conta 'minated Aber Benoit and Aber Wrach than in adductors of oysters from reference stations. This difference cannot be attributed to differences in seawater salinity between Aber and reference stations, since all stations had salinities in the 30-34 o/oo 307 Table 21. Concentrations of glycogen in the liver of plaice Pleuronectes platessa from a reference station-at Ile Tudy and from oil- polluted Aber Benoit and Aber Wrac'h. Values are in mg glycogen/ 9 wet weight. Station February 1930(23)a June 1980(27) Reference 4.81 + 9.05 8.45 + 7.03 Aber Benoit 0.48 + 0.39 6.68 + 7.56 Aber Wrac'h 14.37 + 17.42 12.35 + 12.0 a, months after the Amoco Cadiz oil spill, 16 March 1978. 308 Table 22. Concentration of serum total cholesterol and HDL cholesterol in plaice Pleuronecter, platessa from a reference station at Ile Tudy and from oil-polluted Aber Benoit and Aber Wrac'h. Values are in mg cholesterol/100 ml serum. Station February 1980(23)a Sampling Date June 1980(27) Total -Chol. HDL Chol. HDL Chol. Total Chol. HDL Chol. HDL Chol.. % Total % Total w Reference 196.9 + 68.1 170.2 + 57.8 86.5 + 4.0 260.9 + 81.2 229.5 + 44.1 89.0 + 11.8 Aber Benoit 244.3 + 63.2 159.3 + 39.4 66.2 + 12.4 406.1 + 101.3* 189.6 + 23.3* 48.4 + 9.4 Aber Wrac'h 271.7 + 75.0* 107.3 + 43.5* 70.7 + 13.8 269.9 + 53.0 165.5 + 36.7* 62.3 + 11.4 *' significantly different from reference at d = 0.05 . n = 10. a, months after the knoco Cadiz oil spill, 16 March 1978. Table 23. Concentration of ascorbic acid in the liver of plaice Plouronectee plate.,;sa from reference stations and from oil-polluted Aber Benoit and Aber Wrac'h. Values are in mg ascorbate/g wet weight. Sampling Date Station Dec 1978(g)a Apr 1979(13) Aug 1979(17) Feb 1980(23) Jun 1980(27) Reference 136.6 + 22.5 137.4 + 16.2 131.2 + 15,2 5.4 + 2.3 69.7 + 21.2 Aber Benoit 108.7 + 20.9 80.6 + 29.8* 88.3 + 37,1 17.3 + 3.5* 44.7 + 14.8* Aber Wrac'h NS 93.1 + 28.5* 64.3 + 32.5* 25.1 + 6.0* 50.1 + 8.7* significantly dif'Ferent from reference at a = 0.05 . n 8-10. a, months after the Amoco Cadiz oil spill, 16 March 11978. 310 Table -24. Concentrations of free amino acids in the adductor muscle of oysters Crassortrea gigas from a reference station in the Rade de Brest and from oil-polluted stations in Aber Benoit and Aber Wrac'h. n = 5 unless stated otherwise. December 1978 (nine months after spill) FAA Concentration (VM/q wet weight and standard deviation) Amino Acid -kade de Brest I'Aber Benoit l'Aber Wrac'h LYS 0.65 + 0.06 1.02 + 0.16* 0.68 + 0.11 HIS 0.25 + 0.04 0.53 + 0.09 0.25 + 0.03 ARG 6.71 + 0.45 6.68 + 0.29 5.67 + 1.05 TAU 62.90 + 8.61 68.70 + 7.99 63.91 + 3.41 ASP 3.57 + 0.55 0.78 + 0.06* 0.41 + 0.12* THR 2.83 ++ SER 2.85 + 1.96 3.93.+ 0.52 3.05 + 0.16 GLU 8.81 + 1.12 7.59 + 1.09 5.14 + 0.091, PRO 28.41 + 6-39 40.82 + 9.97* 24.01 + 3.01 GLY 67.69 + 3.44 30.56 + 7.82* 26.54 + 2.88* ALA 18.98 + 1.52 13.50 + 2.90 10.15 + 0.49 CYS VAL 0.33 0.19 + 0.06 MET 0.11 + 0.01 + 0.31 + Ml 0.18 + 0.09+ ILE 0.15 + 0.03+ 0.12 + 0.06+ 0.10 + 0.03 + LEU 0.20 + 0.03 + 0.41 + 0.21 + 0 .19+ 0.01 + TYR 0.12 ++ 0.33 + 0.18 + PHE 0.22 ++ NH3 2.55 + 0.45 2.74 + 0.33 2.33 + 0.19 Total FAA 201.73 + 24.21 175.50 + 31.52 143.30 + 11.54 not detected detected in two samples detected in one sample significantly different from reference at a 0.05 311 Table 25 Concentration of free amino acids in the adductor muscle of oysters Crassortrea gigas from a reference station in the Rade do Brest and from oil-polluted stations in Aber Benoit and Aber Wrac'h. n = 5 unless otherwise stated. April 1979 (thirteen months after spill) FAA Concentration (1,M/g wet weight and standard deviation) Amino Acid Rade de Brest Aber Benoit Aber Wrac'F LYS 0.91 + 0.15 0.79 + 0.15 1.04 + 0.46 HIS 0.29 + 0.14 0.24 + 0.07 0.33 + 0.10 ARG 5.06 + 0.84 4.31 + 0.65 4.64 + 0.72 TAU 57.41 + 9.77 64.39 + 6.19 65-58 + 3.87 ASP 1.46 + 0.58 1.31 + 0.58 0.92 + 0.39 THR SER 2.40 + 0.82 3.59 + 0.92 2.76 + 0.46 GLU 9.89 + 2.08 10.27 + 1.21 10.37 + 1.11 PRO 26.15 + 9.53 20.76 + 9.29 6.76 + 5.28* GLY 28.02 + 4.38 25.62 + 20.46 26-09 + 4.38 ALA 11.59 + 3.74 10.33 + 2.87 10.94 + 0.96 CYS 0.170++ VAL 0.26 + 0.03 + MET 0.10 + 0.02 + 0.21 + 0.05 + 0.11 + 0.06 + ILE 0.15 + 0.06+ 0.13 + 0.01 + 0.11 + 0.03 + LEU 0.27 + 0.11 0.26 + 0.02 0.21 + 0.06 TYR 0.163 ++ 0.134++ PHE NH3 2.44 + 0.79 1.82 + 0.62 1.87 + 0.44 Total FAA 146.94 143.60 129.86 not detected detected in two samples ++, detected in one sample significantly different from reference at a = 0.05. 312 Table 26. Concentration of free amino acids in the adductor muscle of oysters Cra-,soctrea gigas from a reference station at Ile Tudy and from an oil-polluted station in Aber Wrac'h. n = 5 unless otherwise stated. Auqust 1979 (sixteen inonths after spill) FAA Concentration (jrj/g wet weight and standard deviation) Amino Acid Ile Tudy Aber Wrac'h LYS 0.43 + 0.20 0.21 + 0.27 HIS 0.35 + 0.17 0.28 + 0.13 ARG 3.82 + 0.27 6.67 + 1.83 TAU 62.09 + 6.54 63.32 + 5.58 ASP 2.83 + 0.92 0.97 + 0.49* THR SER 1.59 + 0.71 2.73 + 1.23 GLU 8.07 + 4.59 7.66 + 2.13 PRO 33.46 + 25.72 28.57 + 9.74 GLY 41.11 + 16.13 26.11 + 11.52 ALA 11.92 + 5.31 14.09 + 3.72 CYS VAL MET 0.29 + 0.02+ 0.23 + 0.13 + ILE 0.03 + 0.03+ 0.23 + 0.30 + LEU 0.11 + 0.07 0.19 + 0.003 TYR PHE NH3 2.44 + 1.13 2.43 + 0.56 Total FAA 166.12 151-77 not detected detected in two samples significantly different from reference at a 0.05. 313 range. Dominant tissue-free.amino a 'cids in all samples were taurine (TAU), glycine (PLY)., proline (PRO) and alanine (ALA). In all samples from all collectiQns.-and stations@ taurine concentration was maintained nearly cQnstant (range pf means., 57.4 - 68.7 g14/g wet weight). There was-a trend for glycine and aspar-tic acid conce ntrations to be lower in adductors. pf pysters. from the two pil-contaminated Abers than in adductors:o,f reference py-sters., The result was that free taurine:- glycine molar ratios (A recommended index of pollutant stress) were significantly higher in adductor i4us-cles of oysters from Aber Benoit and Aber Wrac!h than in adductors of reference oysters in all but one instance (Table 27). Jefferies (1972) has suggested that taurine:- glycine ratios higher than about 2.0 in mollusc tissues may be a good index of stress-. As indicated above, the high taurine:glycine ratios are attributed almost exclusively to a decrease in free glycine concen- tration. This, in turn, may be attributed to poorer nutritional status or altered patterns of amino acid@metabolism in oil-stressed oysters. Similar patterns were observed in free amino acid profiles and concentrations. in skeletal muscle of plaice (Table 28-32). Total free amino acid concentrations were much lower in plaice muscle than in oyster muscle, reflecting the well-developed capability of plaice to regulate body fluid concentration hypoosmotic to the ambient seawater medium. As in oyster muscle, taurine, glycine and alanine were the dominant free amino acids in plaice muscle. Concentrations of several free amino acids were statistically significantly different in muscle of plaice from Aber Benoit and/or Aber Wrac'h than in muscle of refer- ence plaice. However, there was no consistent pattern of change. Free glycine concentration was lower in muscle of plaice from the Abers than @in muscle of plaice from reference stations in December 1978 and August 1979.. In February and June 1980, free taurine concentration in muscle of Aber Wra0h plaice was lower than in muscle of reference fish. In February 1980, it was higher. Despite these as yet unexplained varia- tions, in seven out of nine cases Where comparative data were available, mean free taurine;glycine molar rat-ios in muscle of plaice from Aber Benoit and Aber Wrac'h were statistically significantly different from ratios in muse .,le of reference fish (Table 33). Because of seasonal variations in free taurine:glycine ratios in muscle tissue of oysters and pl4ice, it is important when using this parameter as an index of stress, to compare Values. for ppllutant-impacted and reference animals collected at the same time from nearby locations. Several biochemical parameters were evaluated as potential indices of pollutant stress in oysters and plaice from oil-contaminated Aber Benoit and Aber Wrac'h. Values of some of these parameters were statistically significantly different in populations from the 314 Table 27 Mean free taurine:glycine molar ratios in adductor muscle of oysters Crassostrea gigas from reference stations (Rade de Brest or Ile Tudy) and from oil-contaminated estuaries (Aber Benoit and Aber Wrac'h). Seven replicate samples from each station were analyzed. Sampling Date Station Dec 1978(g)a April 1979(13) July 1979(16) Reference 0.93 2.05 1.51 Aber Benoit 2.25* 2.51 NS Aber Wrac'h 2.41* 2.51* 2.42* *' significantly different from reference sample at d = 0.05. NS, no sample analyzed. a, months after the Amoco Cadiz oil spill, 16 March 1978. 315 Table 28 Concentration of free amino acids in skeletal muscle plaice Pleuronectes platessa from a reference station in Baie de Douarnenez and from oil-polluted Aber Benoit. n 5 unless stated otherwise. December 1978 (nine months after spill) FAA Concentration (pM/q wet weight and standard deviation) Amino Acid Bale de Douarnenez -Aber Benoit LYS 0.28 +0.10 1.47 +0.76* HIS 0.40 +0.04 0.53 +0.14 ARG 0.37 +0.09 0.26 +0.08 TAU 11.33 +2.68 11.23 +3.06 ASP THR 0.84 +0.19 0.66 +3.06 SER 0.92 +0.68 0.71 +0.18 GLU 0.35 +0.13 0.15 +0.06+ PRO 0.27 +0.11 0.48 +0.18 GLY 11.86 +4.81 5.58 +1.48* ALA 2.91 +0.99 1.34 +0.09* CYS VAL 0.12 +0.01+ 0.23 +0.17+ MET 0.07 +0.01 + 0.07 +0.02 ILE 0.06 +0.01+ 0.11 +0.07 LEU 0.11 +0.01 + 0.13 +0.08 TYR PHE NH 3 6.36 + 0.39 6.29 + 0.34 Total FAA 29.89 + 9.86 22.95 + 6.58 Not detected two samples significantly different from reference at 0.05. 316 Table 29 . Concentration of free amino acids in skeletal muscle of plaice Pleuronectec platessa from a reference station at Loc Tudy and from oil-polluted Aber Benoit and Aber Wrac'h. n = 5 unless otherwise stated. April 1979 (thirteen months after spill) FAA Concentration (iiM/g wet weight and standard deviation) -Amino Acid Loc Tudy Aber Benoit Aber Wrac'h LYS 0.47 + 0.23 0.88 + 0.21* 0.18 + 0.11 HIS 0.92 + 0.31 0.99 + 0.21 0.56 + 0.35 ARG 0.21 + 0.12 + TAU 14.67 + 3.19 11.41 + 1.38 8.94 + 2.43* ASP 0.13 + 0.06 0.05 + 0.03 0.06 + 0.02 THR 0.89 + 0.31 1.21 + 0.58 0.59 + 0.27 SER 0.77 + 0.19 0.97 + 0.09 0.66 + 0.16 GLU 0.29 + 0.14 0.29 + 0.07 0.30 + 0.05 PRO 0.25 + 0.03 0.87 + 0.24* 0.56 + 0.31 GLY 7.38 + 0.47 9.81 + 1.50 8.86 + 3.01 ALA 1.77 + -0.28 1.28 + 0.18 1.15 + 0.24* CYS 0.12 ++ VAL 0.11 + 0.01 + 0.12 + 0.01+ MET 0.49 + 0.01 + 0.06 + 0.02 + 0.06 + 0.02+ ILE 0.09 + 0.03+ 0.04 + 0..03+ 0.07 + 0.01+ LEU 0.08 + 0.03 + 0.11 + 0.02 + 0.06 + 0.05+ TYR PHE NH 3 4.69 + 0.74 5.07 + 0.61 5.30 + 1.24 Total FAA 28.21 28.41 22.18 not detected detected in two samples ++* detected in one sample signi -ficantly different from reference at 0.05 317 Table 30. Concentration of free amino acids in skeletal muscle of plaice PLeuronectes platessa from a reference station at Ile Tudy and from oil-polluted Aber Benoit and Aber Wrac'h. n = 5 unless otherwise stated. August 1979 (seventeen months after spill) .FAA Concentration (uM/q wet weight and standard deviation) Amino Acid Ile Tud@, Aber Benoit Aber Wr_ae h LYS 0.96 + 0.65 0.64 + 0.33 1.40 + 0.70 HIS 0.97 + 0.66 0.66 + 0.12 1.50 + 0.21 ARG 0.17 + 0.02 0.21 + 0.01+ TAU 9.28 + 1.22 10.08 + 1.32 8.01 + 1.16 ASP 0.03 + 0.01 0.05 + 0.03' 0.06 + 0.05 THR 0.36 + 0.15 0.48 + 0.07 0.65 + 0.24 SER 0.27 + 0.22 0.41 + 0.22 0.47 + 0.25 GLU 0.10 + 0.01 0.18 + 0.05+ 0.17 + 0.11 PRO 0.50 + 0.03+ 0.74 + 0.20+ 1.78 + 1.41 GLY 9.96 + 3.40 6.57 + 3.14 3.70 + 1.56* ALA 1.67 + 0.81 0.86 + 0.26 1.27 + 0.29 CYS 0.18 VAL 0.16 + 0.15+ MET 0.03 + 0.04+ 0.39 + 0.53 + 0.17 + 0. + ILE 0.41 + 0.03 + 0.07 + 0.03 + 0.11 + 0.09+ LEU 0.04 + 0.001 + 0.10 + .0.08+ 0.16 + 0.11+ TYR PHE NH 3 5.65 + 0.58 5.38 + 2.19 6.34 + 1.27 Tot&T FAA 24.40 21.39 20.29 --, Not detected +; detected in two samples ++1 detected in one sample significantly different from reference at a 0.05. 318 Table 31. Concentration of free amino acids in skeletal muscle of plaice Pleuronectes ptatetsa from a referEnce station at Ile Tudy and from oil-polluted Aber Benoit and Aber Wrac'h. n = 8 to 10 unless otherwise stated. February 1980 (23 months after spill) FAA Concentration (PM/g wet weight and standard deviation) Amino Acid Ile Tudy Aber Benoit Aber Wrac'h LYS 0.57 + 0.40 0.95 + 0.58 0.55 + 0.42 1 HIS 0.50 +0.27(7) 0.70 + 0.39 ARG TAU 7.55 + 2.90 12.57 + 3.38* 12.36 + 2.33* ASP 0.20 + 0.15(8) THR 0.33 + 0.11 0.63 + 0.20* 0.78 +0.42* SER 0.70 + 0.52 0.93 + 0.50 0.79 +0.36 GLU 0.57 + 0.20 0.24 + 0.13* 0.27 +0.10* PRO 1.51 +0.72(3) GLY 7.42 + 2.64 15.49 + 7.96* 18.61 +5.32* ALA 3.70 + 0.74 1.63 + 0.59* 1.53 +0.40* CYS VAL MET 0.32 + 0.08(5) 0.03(l) ILE 0.21 + 0.14(7) LEU 0.29 + 0.17(7) 0.16(l) TYR PHE NH3 NA NA NA Total FAA 20.86 32.94 38.80 not detected NA, not analyzed significantly different from reference at a 0.05 number of samples in which amino acid was detected. 319 Table 32 . Concentration of free amino acids in skeletal muscle of plaice Neuronectes plate,,sa from a reference station at Ile Tudy and from oil-polluted Aber Benoit and Aber Wrac'h. n = 8 to 10 unless otherwise stated. June 1980 (twenity-seven mo ths after spill) FAA Concentration (pM/g wet weight and standard deviation) Amino Acid Ile Tudy Aber Benoit Aber Wrac'h LYS 0.34 + 0.30 0.63 +0.52 1.05 + 0.32* HIS 0.29 + 0.13 1.39 +0.63* 1.87 + 0.46* ARG TAU 16.92 + 3.96 12.37 + 3.63 11.81+ 2.30* ASP 0.13 + 0.14(7) 0.08+ 0.03 0.06 + 0.02 THR 0.33 + 0.17 0.87+ 0.37 0.77 + 0.35 SER 0.80 + 0.32 0.59+ 0.38 0.63 + 0.37 GLU 0.26 + 0.19 0.25+ 0.08 0.19 + 0.06 PRO 0.24 + 0.28 1.73+ 2.34(g) 2.30 + 2.33(9) GLY 2.50 + 2.08 14.65+ 6.95* 8.77 + 4.52* ALA 2.26 + 0.60. 1.28+ 0.43* 1.07 + 0.32* CYS VAL 0.34 + 0.37(3) 0.15+ 0.02(6) 0.16 + 0.02(6) MET 0.15 + 0.08 0.12+ 0.02(9) 0.14 + 0.06(9) ILE 0.12 + 0.18 0.08+ 0.04 0.09 + 0.02(8) LEU 0.16 + 0.25 0.16+ 0.05 0.16 + 0.03(7) TYR PHE NH3 NA NA NA Total FAA 24.84 35.37 29.07 not detected NA, not analyzed significantly different from reference at a = 0.05 number of samples in which amino acid was detected. 320 Table 33. Mean free taurine: glycine molar ratios in skeletal muscle of plaice Pleuronectes platessa from rEference stations (Baie de Douarnenez,- Loc Tudy, or Ile Tudy) and from oil-contaminated estuaries (Aber Benoit and Aber Wrac'h). Seven or ten replicate samples from each station were analyzed. Sampling Date -Station Dec 1978(g)a Apr 1979(13) Aug 1979(17) Feb 1980(23) Jun 1980(27) Reference 0.96 1.99 0.93 0.66 1.35 Aber Benoit 2.10* 1.16* 1.53* 0.81 0.84 Aber Wrac'h NS 1.01* 2.16*. 1.02* 6.77* significantly different from reference sample at a = 0.05. NS, no sample analyzed. a, months after the Amoco Cadiz oil spill, 16 Mitrch 1978. 321 oil-polluted Abers and from nearby reference stations. These may be useful indices of polluta'nt stress. They include blood cholesterol/ HDL choles.terol, liVer ascorbic acid, and skeletal muscle-free amino acid ratios in fish; and adductor 14uqcle-free amino acid ratios in qypters. Blood glucose also, has. potential as an index of stress in fish, if the fish can be captured and blood samples, taken ver quickly. y Alternatively, useful information can be obtained if degree And duration of capture@induced stress can be standardized for reference and experi- mental fish. In such a case, the tndex of chronic pollutant stress is hypoglycemia, reflecting a loss or dimin.uAtion of the capacity of the hypophyseal-interrena.1 system to respond to stress. Several of the alterations in biochemical parameters in oil-polluted fish and oysters are indicative or symptomatic of poor nutritional status (e.g., depressed muscle glycine, depletion of liver glycogen and ascor- bate, etc.). Thismay be related to histopathological lesions, reported by Haensly and Neff in this publication in the gut and liver of plaice from the oil-contamianted Abers. One difficulty in using biochemical and histopathological parameters as indices.of pollutant stress is that one is riot always certain that animals from the impacted and reference sites are from the same popula- tion and therefore can be compared biochemically and histopathologically. The only way to establish convincingly that differences observed in indi- cator parameters in reference and impacted populations are due solely or primarily to the pollution incident under investigation, is to have comparative data collected before the pollution incident. This usually is not available. The oysters used in this investigation are a recently introduced species Crassostrea gigas and are from a common breeding stock throughout Brittany. Genetic differences between reference oysters and oysters from the Abers are -therefore extremely unlikely. However, the extent to which plaice from, the west and south coast of Brittany (reference sites)-mix and interbreed with plaice from the northwest coast (site of the Abers) is not known. Some intermixing undoubtedly occurs. It seems likely, therefore, thatmany of the differences we have reported between oysters and plaice from the Abers and those from reference stations are attributable directly or indirectly to impacts of the Amoco Cadiz oil spill. There has been substantial improvement in condition of oysters and plaice in the Abers during the timecourse of this investigation (up to 27 months after the spill). Recovery is still not complete, however. 322 A most interesting observation from our investigations is that oysters, which were heavily contamianted by the oil spill and remained so for the duration of the investigation, showed little evidence of histopathological or biochemical damage, whereas plaice from the Abers, although not heavily contaminated with oil, showed evidence of serious and progressive histopathological and biochemical damage. This may be due to differences in the sensitivity of molluscs and fish to petrol- um. However, an alternative hypothesis is that the metabolites of petroleum hydrocarbons, particularly of the polycyclic aromatic hydro- carbons, are much more toxic than the unmetabolized parent compounds and cause much of the damage in a chronic pollution situation. It is well-established that the phenolic, epoxide and diol metabolites of polycyclic aromatic hydrocarbons are much more elctrophilic and bio- logically reactive than the unoxygenated parent compounds (Neff, 1979). Since oysters have little or no capability to oxygenate polycyclic aro- matic hydrocarbons to reactive metabolites, they are quite tolerant to oil. Fish on the other hand have a highly active mixed-function oxygen- ase system and so rapidly convert polycyclic aromatics to reactive metabolites which cause tissue damage. This hypothesis warrents further investigation. 323 REFERENCES CITED Anderson, J.W., 1977, Responses to sublethal levels of petroleum hydro- carbons: are they sensitive indicators and do they correlate with tissue contamination? pp. 95-114 In: D.A. Wolfe (ed.) Fate and Effects of Petroleum Hydrocarbons in Marine Organisms and Ecosystems. New York: Pergamon Press. Armstrong, H.W., J.M. Neffand R. Sis, 1980, Histopathology of benthic invertebrates and demersal fish: central Gulf of Mexico oil plat- form monitoring study. U.S. Dept. of Interior, Bureau of Land Management, New Orleans, LA. 132 pp. Blanton, W.G. and M.C. Robinson, 1973, Some acute effects of low-boiling petroleum fractions on the cellular structures of fish gills under field conditions. pp. 265-273 In: D.G. Ahern and L.P. Meyers (eds.) The Microbial Degradation of Oil Pollutants. Baton Rouge, LA:L.S.U. Center for Wetland Resources. Chasse, C., 1978, The ecological impact on and near shores by the Amoco Cadiz oil spill. Mar. Pollut. Bull., 9:298-301. Chasse, C. and D. Morvan, 1978, Six mois apres la maree noire de i'Amoco Cadiz bilan provisoire de l'impact ecologique. Penn ar Bed, 11: 311-338. Cross, F.A., W.P. Davis, D.E. Hoss, and D.A. Wolfe, 1978, Biological observations. pp. 197-215 In: The Amoco Cadiz oil spill. A preliminary report. Ed.. by W.N. Hess. NOAA/EPA Special Report. Dax@e, C-J., D.G. Scarpelli, and S.R. Wellings (eds.), 1976, Tum.ors in aquatic animals. In: Progress in Experimental Tumor Research. Vol. 20. Basel, Switzerland: S. Karger. 438 pp. Ernst, V.V., J.M. Neff, and J.W. Anderson, 1977, Effects of the water- solublefraction of No. 2 fuel-oil on the development of the estuarine fish, FunduZus grandis Baird and Girard. Environ. Pollut. 14:25-35. Eurell, J.A. and W.E. Haensly, 1981, The effects of exposure to water- soluble fractions of crude oil on selected histochemical parameters of the liver of Atlantic croaker, Micropogon unduZatus L. J. Fish Dis. 4:197-194. 324 Gardner, G.R., 1975, Chemically-induced lesions in estuarine or marine teleosts. pp. 657@693 In: W.E. Ribelin and G. Magaki (eds.) The Pathology of Fishes. Madison, WI: U. of Wisconsin Press. Haensly, W.E. and J.M. Neff, 1982, Histopathology of Pleuronectes pLatessa from Aber Wrac'h and Aber Benoit, France 9 to 27 months after the Amoco Cadiz oil spill (this volume). Hawkes, J.W., 1977, The effects of petroleum hydrocarbon exposure on the structure of fish tissues.. pp. 115-128. In: D.A * Wolfe (ed.) Fate and Effects of Petroleum Hydrocarbons in Marine Organisms and Ecosystems. New York: Pergamon Press. Hess, W.N. (ed.), 1978, TheAmoco Cadiz oil spill. A preliminary report. NOAA/EPA Special Report. 283 pp. Hodgins, H.O., B.B. McCain, and J.W. Hawkes, 1977, Marine fish and invertebrate diseases, host disease resistance, and pathological effects of petroleum. pp. 95-174 In: D.C. Malins (ed.), Effects of Petroleum on Arctic and Subarctic Marine Environments and Organisms. Vol. II. Biological Effects. New York: Academic Press. Jeffries, H.P., 1972, A stress syndrome in the hard clam, MercentTria mercenaria. J. Invert. Pathol. 20:242-251. Johnson, F.G., 1977, Sublethal biological effects of petroleum hydro- .carbon exposures: bacteria, algae, and invertebrates. pp. 271- 318 In: D.C. Malins (ed.), Effects of Petroleum on Arctic and Subarctic Marine Environments and Organisms. Vol. II. Biological Effects. New York: Academic Press. Kraybill, H.F., C.J. Dawe, J.C. Harshbarger, and R.G. Tardiff (eds.), 1977, Aquatic pollutants and biologic effects with emphasis on neoplasia. Ann. N.Y. Acad. Sci. 298. 604 pp. Lee, R.F., 1977, Accumulation and turnover of petroleum hydrocarbons in marine organisms. pp. 60-70 In: D.A. Wolfe (ed.) Fate and Effects of Petroleum Hydrocarbons in Marine Organisms and Eco- systems. New York: Pergamon Press. 325 McCain, B.B., H.O. Hodgins, W.D. Gronlund, J.W. Hawkes, D.W. Brown, M.S. Myers, and J.H. Vandermeulen, 1978, Bioavailability of crude oil from experimentally oiled sediments to English sole (Faro- phrys vetulus), and pathological consequences. J. Fish. Res. Bd. Can. 35:657-667. McKeown, B.A. and G.L. March, 1978, The acute effect of bunker C oil and an oil dispersant on: 1 serum glucose, serum sodium and gill morphology in both freshwater and seawater acclimated rainbow trout (SaLmo gairdmri). Water Res. 12:157-163. Michel, P. and H. Grizel, 1979, Etude de-impact de la pollution petroliere de 1-Amoco Cadiz sur les huitres-Mars 1978/Avril 1979. CNEXO con- tract #78/5719. Centre Oceanologique de Bretagne. Brest, France. 32 pp. Neff, J.M., 1979, Polycyclic aromatic hydrocarbons in the aquatic environment: sources, fates and biological effects. Applied Science Publishers. Barking Essex, England. 266 pp. Neff, J.M. and J.W. Anderson, 1981, Responses of marine animals to petroleum and specific petroleum hydrocarbons. Applied Science Publishers, Barking Essex, England. 177 pp. Neff, J.M., J.W. Anderson, B.A. Cox, R.B. Laughlin, Jr., S.S. Rossi, and H.E. Tatem, 1976a, Effects of petroleum on survival, respiration and growth of marine animals. pp. 515-539. In: Sources, Effects and Sinks of Hydrocarbons in the Aquatic Environment. Washington, D.C.: American Institute of Biological Sciences. Neff, J.M., B.A. Cox, D. Dixit, and J.W. Anderson, 1976b, Accumulation and release of petroleum-derived aromatic hydrocarbons by four species of marine animals. Mar. Biol. 38:279-289. Patten, B.G., 1977, Sublethal biological effects of petroleum hydro- carbon exposures: fish. pp. 319-336 In: D.C. Malins (ed.), Effects of Petroleum on Arctic and Subarctic Marine Environments and Organisms. Vol. II. Biological Effects. New York: Academic Press. Roubal, W.T. and T.K. Collier, 1975, Spin-labeling techniques for studying mode of action of petroleum hydrocarbons on marine organisms. Fish. Bull. 73:299-305. 326 Sinderman, C.J., 1979, Pollution-associated diseases and abnormalities of fish and shellfish: a review. Fish. Bull. 76:717-749. Southward, A., 1978, Marine life and the Amoco Cadiz. New Scientist, 20 July 1978, 174-176. Spooner, M.F., 1978, Editorial introduction. Amoco Cadiz oil spill. Mar. Pollut. Bull., 9:281,284. Thomas, P., M. Bally, and J.M. Neff, 1982, Ascorbic acid status of mullet, Alugil cephaZus Linn. exposed to cadmium. J. Fish Biol. 20 (in press). Thomas, P., R.S. Carr, and J.M. Neff, 1981, Biochemical stress responses. of mullet Muqil cephalus and polychaete worms Neanthes virens to pentachlorop6nol.' pp. 73-103. In: J. Vernberg, A. Calabrese, F.P. Thurberg, and W.B. Vernberg (eds.) Biological Monitoring of Marine Pollutants. New York: Academic Pres-s. Thomas, P., B.R. Woodin, and J.M. Neff, 1980, Biochemical responses of striped mullet A&giZ cephalus to oil exposure. 1. Acute responses- interrenal activation and secondary stress responses. Mar. Biol. 59:141-149. Varanasi, U. and D.C. Malins, 1977, Metabolism of petroleum hydrocarbons: accumulation and biotransformation in marine organisms. pp. 175- 270 In: D.C. Malins. (ed.) Effects of Petroleum on Arctic and Subarctic Marine Environments and Organisms. Vol. II. Biological Effects. Academic Press, New York. Wardle, C.S., 1972, The changes in blood glucose in Pleuronectes ptatessa following capture from the wild: a stress reaction. J. Mar. Biol. Ass. U.K. 52:635-651. Zannoni, V.C., M. Lynch, S. Goldstein, and P. Sato, 1974, A rapid micro- method for the determination of ascorbic acid in plasma and tissues. Biochem. Med. 11:41-48. 327 RETABLISSEMENT NATUREL D'UNE VEGETATION DE MARAIS MARITIMES ALTEREE PAR LES HYDROCARBURES DE L'AMOCO-CADIZ:MODALITES ET TENDANCES par Jacques E. LEVASSEUR et Marie-L. JORY Laboratoire cle Botani.que G6n6raLe Campus Sci.enti.,fi-que cle BeauLieu 35042 -RENNES C6clex - France R E S U M E Le r6tabLissement cle La v6g6tation des marais cle L'ILe Grande partieL- Lement d6truite par Les hydrocarbures est significativement engag6 et ce depuis 1980. Les modaLit6s et La chronoLogie du r6tabLissement sont fonction cle La domi- nance reLative, en chaque point, cle cleux processus : reg6n6ration in situ d'indi- viclus p6rennes, germination cle graines et semences procluites sur pLace ou clans Le voisinage. La coLonisation est surtout Le fait d'esp&ces annueLLes, aLors que La germination des esp&ces p6rennes est tr&s peu fr6quente, sauf dans Les zones abri- t6es 6 substrat meubLe et propre. ELLe est toutefois raLentie ou nuLLe clans Les secteurs expos6s aux effets directs cle La mar6e et/ou pi6tin6s intens&ment Lors clu nettoiement cle 1978. CeL6 justifie Les efforts cle restauration voLontaire, au moyen cle pLantation.S. tent&s clans cle teLs sites et clont un des int6r&ts est d'ac- ck6rer Les ph6nom6nes cle d6p6t des s6climents et des semences. D'autre part, des esp6ces initiaLement "r6sistantes" pr6sentent actueL- Lement une sensibiLit6 marqu6e A La poLLution enclog6e toujours activeuqui se tra- cluit par Le d6cLin et, h terme, par La clisparition sur de Larges espaces de popu- Lations enti6res (cf. Juncus maritimus Lam.). Ainsi, ces processus, agissant simuLtan6ment ou successivement conclui- sent-iLs b des s6quences cle r6tabLissement vari6es, 6 des stacles transitoires (?) marqu6s par une redistribution spatiaLe des esp6ces,qui s'L&carte notabLement cle La distribution ant6rieure. A B S T R A C T Recovery of ILe Grande saLt-marsh vegetation, partiaLLy destroyed by hydrocarbons has been significantLy started up since 1980. Ways and timing of re- covery are due to the reLative dominance, in each point, of two processes viz. in situ regeneration of perenniaL individuaLs., germination of seeds producted near or 0 e site. CoLonisation is mainLy due to annuaL species whiLe germination of perenniaLs is a rare event, except in shades pLaces with Loose and cLean substrate. However, it is impeded either in tide exposed points or in formerLy heaviLy tram- pLed pLaces. So, efforts of voLontary restoration are justified in such Locations pLanting actsbesides by the speeding up of the aggregation of sediments and seeds cLose to the transpLants. 329 In an other hand, some species, initiaLLy "resistant" show a marked sensibiLity to underground actuaL poLLution and consequentLy,Large popuLations may decLine or even die (cf..Juncus maritimus Lam.). FinaLLy, these processes, acting in simuLtaneous or successive manners, wiLL Lead to varied recovery sequences, to transitory (?) stages characterized by a. spatiaL species redistribution which may be quite different from the origi- naL pattern. M 0 T S C L E S RdtabLissement, reg6n6ration, restauration, successions secon- .daire et primaire, poLLution par Les hydrocarbures, nettoiement, v6g6tation cle marais maritimes. K E Y W 0 R D S Recovery, regeneration, restoration, secondary and primary successions, hydrocarbons poLLution, cLeaning up, saLt marsh vegetation. 330 INTRODUCTION Lerg-tablissement dun couvert vqigqg-tal perturbqd est un processus complexe qui recouvre des rqialit0qh et prq6sente des modalqitq6s trqas diver- ses, d'autaqnt que les causes perturbantes n1qont pas eu le mqeqme impact suivant les lieux et suivant les espq6ces composaqnt le tapis vqgqgiqgtal (Baker,, 1979 ; Levasseur etal., 1981). Les marais maritimes constituent un ensemble hq6tq6rogqaqne, qui quoi- que fondamentalement organqiqs6qVen habitats q9-tagqds, aux conditions mqiso- logiq8ques varqiq6es, supporte-une vq6gq6tatqion qualqitatqi'vement ou dans les espaces intrazonaux,quantitativement variqg-e. Cependant, q6tant donnqg qu'il slagit d'envqiroqnnements pqfLysqiquement dq6termqiqnqgs, surtout dans les par- ties moyennes et baqssesdeqsmarais, la diversitqi spq6cqifique est faible, ce qui sqiq@gn2qf2qf4qie qu'une perturbation peut avoqir un effet drastique sur des communautq6s vq6gq6tales et ceci d'autant plus qu'elles eront pauci- spqA-cifiques et/bu partqiculiq@rement sensibles, de par leurs composantes, 8q& une cause perturbatrice particuliq6re Ayant dans un travail antqgrieur (Levaseur et al., 1. c.) dqg-taill6 cet aspect des choses, nouqsne prq6senterons, dans cq;qtte communication, que que1ques donnqges relatives au rdtabqlqisement de la vqdg6tation au courqT des trois annqges q9-coul6es-.et ceci aussi Bien dans les espaces non modifiqgs par qIhomme que dansceux qui ont qA-t6 transformqds du fait des opgrations, de nettoqiement. Pour ce faire, nou utqilqierona les documents q&uivant, quoique non exhau$tifs des diffg-rents cas- de figures rencontr6s : - carqtographqie chronologique dun marais choisi pour sa diversitqa iqntrinqsqaque qinitiale, mais ausi pour la diversitg des perburba- tion ql'ayant affect6 depuis marqa 1978 ; - transects permanents, rqiguliqarement relev6s deqpuis 1979 et desti- ng-s-, au plar-populations vqggqdtales)q& qilluqstrer 4q& la foqis la chro- nolgqie des reprises, le dq6veloppement vq6gq6tatiqf ultqdrieur, les r6organisations spatqiales interclone-eqtla colonisation directe par le qindqividus@nouveaux. LE RETABLISSEMENT DEFINITION ET PROCESSUS GENERAUX DqA-fqinition 11 y a lieu de disti4qnguer entre I le r4q6tablisqqq-ement dansq: unqmara6qis dqonqn4qg du couvert v4q6qg4q6tal, qui se traduit par une cicatrisation se d4q6roulant non nq6cessairement lin8qgai- rement danle temps et non syqnchroniquement dan8qs 6qVes20qpace et dont la dur8qge probable, pour qatre men4qge 8q& son termeq5est fonction de nombreux pa- ram8qatres,essentiellement 331 - le degre de destructuration et/ou de destruction initiales - les nouvelles conditions 6cologiques (p. ex. la permanence* dans et sur le sol, et sous diffq6rentes formes, de quantitq6s importantes de pq6trole est un nouveau facteur de 0qVenvironne- ment). Cette cicatrisatqion (qi. e. gains en recouvrement) est indq6pendan- te des vo4qtes@ suivqieqa et des mo8qyens mis en oeuvre. Elle peut quelquefois avoir pour conaqgquence la constitution dtun peuplement vqggqgtal qualqitativement et/'ou structuralement diffq6rent du peuplement qd'orqigine5maqis la dimension temps manque pour qEvaluer le de- grq6 de permanence de 1 q6tat atteint au moment du constat car, en cette matiq6ire, tous les-q6-tatssont conditionnels et contractuels ! 2 - le rq6tablqisement d'une communaquti vq6g6tale partqiculiare, en quqalitqg et en structure, danqs un lieu donnq6 Cet 9-tat, nq6cesitant des. rq6fq6renceqs antgrqieure prq6cises es-t beaucoup plus dqiffqicqile q1 q6valuer que le premier cqitqg, qimqmg-diqatement apprqg-hendabqile car qil s'it du degrqg- de recouvrement par la vq9g6tatqion, au temps t, d'un espace donng. Cependant, le rq6tablqiss-ement apriqis perturbation d-une-qV6gq6tation peut qZtre etqimg, lorsquqil est comprqi comme q6tant la r6paration natu relle des dommages- subis, au moyen de constats q6ta5lis A intervalles rq6gulqkers. Procesqsus-gq6nq6raux. Le r6taqblissement est a la foi un processus et un rqisultaqt, lbrs- que Von considqare qu'iql est menqg a terme. De ce point de vue il est largement engag6 en de nombreux sites et mqgme localement achev6. Mais sous ce ph6nomq6ne,en dq6pit des expressions patiales et des chronolo- gies sqi diverse.% actuellement, sqe retrouvent les mq@qmes qmq6canismes. Ceux-cqi sont fondamentalement au nomqbre de deux auxquels il faudrait ajouter les-actqionqvolontaqireqs de res-tauraqtionpar plantations : Succeaqions-secondai're d'une part, p6qE8qtqlaire d'utre part. qEn fait, la distinction dqanqs un lieu donqng de ces@ deuqx processu est loin d'atre nette car frq6quemment ils agissent synchroniquement et non sq6quentielleq- ment et de plus ils inter-et rqitroagissnt quelqquefois continuement. La rqiparation aturelle des destructions peut se faire soit dans un premier temps q1 partir de la rq6gq6n6ration in __ situ qVqglqiqments vqi'vaces ayant survq6cu, que ceuxq-cqi oqtent situg-s-9 11itq6iie-qUr de la zone attein te ou pqgrqipqhiriquement,s-oit, dansun second temps, par des implantations nouvelles A partir de migrules provenant 0qVindividus ou de clones situq6s en dehors- de la zone qintqgress-4-eyou dindividu ou de portions de clones autochtones ayant pu pourqsuiqvre ou retrouver un cycle p6qhq6nologique nor- mal ou ayant retrouvqg cette capaci0qtq6 [email protected] d8q6laqi plus ou moins long de survqivance en vie ralentie. Dans ce caqs, il s'agit alors d'une colo- nis2qation intersqtit6qielle et6qtou sq6quentielle pu6qiqsque commandq6e spatiale-, ment par 32qVordre de r0q6apparition, la localisation, le nombre et la na@ ture des individus vivaces ayant surv4q6cu, ma6qis auqssi par lqes conditions. 8q6cologiqueqs riqiq_@gnant dans le lieu. Ainsi, un procesus de r0qdg0q6nqgration qu6qi q1 notre qsens se rapportqe d'abord aux espq6ceqs pq6rennes qimpl0qique la poursuite norqmale du cycle v4qdgq6q-. tatif dqtqind8qi'vidus q6pargn6q6s et la reprise de dq6veloppement 0q6pqig4qg dq'indi- vidus survqivants du fait de dispositions- morpholog2qiques part2qicuqli4q;_@res ou de conditionsq-d'habitatqs plus 8qfavorables. 332 Nous distinguerons alors les phases suivantes 1 - Apres une pq6riode plus ou moins longue de vie ralentie, repri- se du cycle phqdnologique normal 2 - Extension vq6gq6tative qgventuellement centrifuge consqg-cutive ou concomittante de la premiq6re phase et/ou formation de graines et semen- ces viables 3 - Poursuite du processus si leg conditions mq6sologiques restent adq6quates et si leg conditions 0qVintqgractions coenotqiques (concurrence) le permettent. Le second proces-sus de colonisation (qi, e.succession primaire a, 1.) implqiqque : 1 - La formation de.graines et semenceqa dans, en pe-rqiphqdrie ou a 1'eqxtqgrieur delazone concernqge 2 - La non-exportation pour celles produites sur place ou inverse- ment ql'accesgibilitqg des qItqeux pour les autres 3 - Des conditions de germination et de dqg-veloppement qfavorables 4 - Le maqintqien de ces conditions auxquelleqs vont s'ajouter les con- ditions coenotqiqueqg leqgunes et leg autreqs variant avec le temps,,dan 4qVespqacejdu fait du proceqssus fondamental de rqitroactqion, La colonigation peut auqgaqi qP-tre le fait de fragmentavqg-gqgtatifs dq6@ tachq6qgde piedsm-eres, dans le lieu ou y ayant accqas.par le jeu des cou- rants. (qf. dispersion actuelle,qui utilise ces deux: modalitq6s,de Sparti- na cf. anglica, dans- le marais 2, mais ausqgi, a lqOuest du pout, dans le marais 4). ILE GRANDE q71 AN INIZIGO IVq-4q-q- marais maritimqes RUN N remblais A R ilegrande (q3 AM q@k rlqaqvoqs. qr08q;8q;q-4q020q9 iocalisation des transects q0 loom 0q.1q0okm Fir2qgure q.1. Carte de localisation des 0qmarais de 1'6qlle Grande. 333 INVENTAIRE S014MAIRE DES SITUATIONS HERITEES A la fin de 1978, les cinq situations suivantes ont qitqg distinguq6es, sur la base du recouvrement ou non par lea hydrocarbures, de l'intensitqg du piqgtinement ou du passage rqgpqg-tqd dengina lors du nettoiement interve- nu en 1978, desopqgrations connexes: de ce nettoiement telles le dqicapage au bulldozer de la couche superfqicielle du sol.ou , 8qVqdtablissement de reqmqblaqtqa.q1 0qVemplacement des foss@eqade stockage- du pq6trole. Groupe A : 1) zones non touchqges ou touchqges. seulement margqinalement par l'q6pandage d'qfqrydrocarbures, maqis quqi ont pu q;-tre secondairement pig- tqinqies. Groupe 0qR : zones-ayant q6tqg soumises a l'impact direct du pq6trole. 2) zone pq6trolqge mats non netto8qyq6e intensqiqvement (i. e. sans pqiqdtqinement qi'mportantyant entrainqg une compaction durable des couches@supqgrqieures du sol). 3) zone pq6trolqge et nettoyqde qintens-qi-4qyement. Groupe C : zones profondqgment modiqfqi-!qEes-par rapport 8q& leur statut antq6rieur : 4) zones remblaqyqdes (mort-terrains, sq6diments meublea) 5) zones q6trq6pq6es- au bulldozer. Caractqdrqiqatiques gq6nq6rqales d ces: zones. Situation 1) Secteurs non atteints ou peu atteints par les hydrocarbures. du fait de leur situation topograph6q�qque ou des7 dispositions prqies: qi'qmmq6dqiatement prqBs-,la catastrophe (cf. maraqis 1 et 2, -a qI'Est du Pont), Localisation : Partqie.T qinternes@dea maraqiqm ou dunes hordiqZres. Processus en cours : Succession secondaqire de cqicatrisqatqion dans les zones piq6tinqges, Situation 2) Terrqitoires-polluqgs mats non ou peu piq6tinqgs. Dq6pq6t initial de pq6trole sur le parties aqdriennes des plantes et sur le sol formant ensuqite sur celui-ci un revq9tement coqhq6rent qui se desqquame localement avec le. temps, danqa lea sites exposq6s (modalitqg 1). Dq6pq5t intrasq6dimentaire de pq6trole ; celui-ci encore actuellement sous: forme semi-liquide dans les chenaux, dans la partie haute de la slikke, maqis aussi en haut-schorre, dans- les zones saturqges en eaux douces par les sources venant du doma0qine terresq-q-tre et qui constituent des marqgcageq-s-supraq@ l2qittoraux saumqitres comme danqs le mara6qis- 6 (modalit0qg 2). Procesqsuqs en cours-: Tr8qZs variables. Successions secondaires ayant d8q6but2q6 d6q6s, 4qlq'autom- ne 1978 et qui s4qont caractqgr0qis4qges: es4qsentielleqment par une r4q6gq6n8q6ration in situ dq'eqspq6ces p4q6rennes 4q9pargnq6es- et survi- vantes. Lorsque la destruction du tapis v4qgg4qdtal a 4q6tq6 plus compl8qate, il y a possibilitqg de colonisation directe par des 0q610q6ments allochtones si leqa conditions s'y pr0qatent. Il y a ainsi possiqbilit4qd dq'une succession primaire inter- stitielle. 334 Situation 3) Territoirfortement piqfqtin6s, ettou soumis A. des passages dlengqiqn.%, notamment d'engqinqs c4qhenqillq6r. quqi.dq6truqfsent les organes endogqda de pq6rennance. Deatruction quasi-totale de la vq6gq6tation ; survivance qdun tr'ea faible pourcentage(quelquefoqis!qinqfqg-rqieur 1 5 d'indi'vqiqdu de type gq6opqhqyte q& -rT3.izome et hq6mqicryptophqyte q1 souche. Compaction secondaire forte avec pour cons:6qquences le tasement du sol et la pq6nq6tration forcqge du pq6trole dans, ses premiers centimqatres. Peu de changements: en trois ans de cet qEtat, sauf en mode expoaqg oa des- dq6lqitations at deqzquamatqion ae produqisentsauf encore dan les zones pro- ximales oumiaeqs 9-a6dqimentatqion. Localqisatqion Cette situation se rencontre urtout danqsle pqArties des marais lea pluaprocheqades chenaux et cqiques, 11 oqa 6tait eqffe- tuq6 le pompage du pq6trole. Lescammunautqiqiqales: plus frqgquem- ment des-tructurq6ea ou dq6truitesFsont les, suiqvantes peuplement A Spartina maritima, of q!.^alicornia perennis, a 6qHalimione portulacoides, A.. Puccinellia maritima et Trqigloc4qhin maritima, qP.P. A Juncus marqitmu du schorre moyen (vqgg6tation .4q1 Limonium vulgare et Plantaqgo martima.(qAl Proceqasus-en cours, Le rqdtablissement de la vq6gdtation, par des voies naturelles y est trq@,s lent sinon nul,. actuellement encore ; succession secondaqire possible A partir des 61q6ments 6pargn6s, mais ceux-ci-sont en quantit6 qinsuffisante pour permettre une cicatrisation rapide de ce lieux. En fait la reprise de la v6gq6tation y est quas-i-nulle, Le rq6tablisement d'un cou- vertvq6ggtal-ne peut q9tre que lqa-consqgqquence dune succes- qaion pri'maire. en qqelque. sqorteoblqigatoqire dans.le lieu, mats qut pour de nembreux aqttes n'eat encore que potentielle, ou bien encore d'opqgrations de restauration par plantations ad hoc.d'esp6ces vivaces. Sqiuq4ti,on 4) Rbqiai Lea partqies-deqmaraqis. remblyq6es ont localis6ea iql qVempla- cement de fosse1ppe qui expliqque lea tasements ultqdrieurs oqbevqg-, depuqi 1980, et la rqdintq6gration de certains de ces espaceadans le domaqine marqi-time s. lq(-qmarai 2 p. ex.) Proceqasus en cours- : Succession primaire obligatoire. Dailleurs trq4s rapidement ini- ti4q6e et qui, en 1981, en est d4qgjq.-"L au stade dq6veloppement vq6q.q- g4q6tatif horizontal d'esqp4qaces- vqivaces- surtout Puccinellia maritima. Il faut noter que dans-ces-lieux la germination 36qVepi4q6ceqs vqi00qvaces a 4qgtqg obqservqge et que la colonisation Sqtest fa0qite 8q1 part0qir degraines etLqaemences ayant eu acc2qis au site et ayant pu germer sur place, alors que ces germi- nations n'ont pas qgt4qg observqges dans les autres sites des marais a s0q6diqments non meubles. 48qMNomencLature d'aprqL@s Abbayes H.des et a4qL.(1970) 335 Situation 5) Territoires q6trq6pq6s. Solution la plus extrq@qhe de nettoqiement puisque la roche- mare, ici limons quaternaires dq6calcifiq6s, eat mise A nu. A part qqquelqques exceptions trqas locales, A la fois le con- tingent de graines produites avant 1978, et 6qVensemble de la masse vq6gq6tale 6pi - et endogq6e a q6tqg dq6truite sur de vastes espaces comme par exemple dans 8qVestuaqire de Ker- lavos, ituq6 A quelques kilomatres q1 qI'Est de l'Ile Grande. Aqilleurs (maraqis -q1, 3, 4, 5, 6, 7) c'est essentiellement la partie proximale des maraqts qui a Stqg amputq6e de la orte. Proceqsus en courqs Initiation dune nouvelle pq6doge'nq6se. Le ritablissement d'un couvert vq6gqgtal ne peut proveqnir que d'une succession primaire obligatoire effectivement enga- gq6e aprqZs une pq6riode de latence de I an via l'qinstalla- tion de populations th6rophytiques essentiellement cons- tituges de diffq6rentes espqaces annuelles du genre Salicor- nqia. CARTOGRAPHI4qE CHqRONOLqOqGIqQU6qE DUN MARAIS PRIS COMM EXEMqPLE : le MARAIS NORD WAN INIZIGO, Les caract6res particuliers de ce marais dont la localisation eat prq6cis6e sur la figure 1, sous le numq6ro 6, et qui nous qVont fait choi- air comme exemple rq6side dan la variq6tg des habitats contigq;s, parti- culiq6rement 6qVexisqtence de prairies saumqitres- upra-littorales dans sa partqie N, domqinqge soit par deqa roselqi6res a Scirpus taqbernaemontani et Scirpus maritimus oit par des prairies qa0qluncus maritimus, espqace quasi- exclusive sur de grands epaces. Morphologiquement, ce marais se relq6ve pdriphq6riquement, le rqdseau hqyqdrographqique con0qverqgeant en son centre seq6soud en deux chenaux ma- qjeurs,orqientq6s Et-0qDuesqt. S6qa partie Ouest a 6tq6 partiellement 6trq6pq6e tand6qk.qque sa patqie Sud a fortement qdtqg piq6tinq6e, le sous-ensemble Nord 6tant de ce point de vue, peu touc0qhq6. La figure 2 rensqeigne chqdmatiquement ur la distribution spatiale des aqgreqasions qu'il a ubien 1978 q: absence de pollution, pollution san. nettoqy-age, pollution puis7 nettyaqge, q6tr6paqge proximal q1 lOuest. La partie Nord-Eat a q6t9 remqblayge q1 diqff6rents moments depuis 1978. Aiqn8qsqi, lea cqinq qsqituation8qa prq6citesq-se retrouvent ici, mais les troiqs premiq6reqs dom6qinent. Les figures 3, 4, 5 se rapportent q1 des conqstatqs q6q-t8qdblis en Aoqat 197q9, 1980 et 1981. 11 ne qsq'agqit pas 4q1 proprement parler de cartes de v4q6gq6tatiqon, puqique 36qVaspect qualitat0qif n'eqat pas le but de ces repre- sentatqionqs. Celui-ci en effet, eat de fixer,. 8q1q-un moment aonnq6, lq'q6tat de la v4q6gq6tation dq'un double point devue - progr6q6qs de la cicatris6qation, - n0qiveau de reconstitution des communautq6s: v8q6-g8q6tales. 336 Plus precisemen le codage corresDond Pour le premier groupe (A) a des peuplements recouvrant la quasi- totalitqg du sol, au moment de la cartographie ; le caractqare diffqirentiel intergroupe q6tant celui de la diversitqg spq6cifique a ce moment 1qA, - Pour le second groupe (B), il s'agit de peuplements en cours de rq6gqinqgration ou soumis 8q1 succession prqimaire. La destructuration, la dq6- nudatqion a pu 8qy avoir q9tqd trq6s forte et le recouvrement (hormis en niveau 6 pour lea esp-eces-annuelles) eat toujours@qinfq6rieur .q1 50 q% de la surface. Code des fqigureqs-3, 4 et 5. Groupe A recouvrement des especes perennes pouvant atteqindre 100 Niveau I : Vqggqdtation plurispqicifique, non touchq6e par lea hyqdrocarbures ou vq6gq6tation ayant recouvrqg, a la fois sa diversqitqg originelle, mais aussi sa structure antqgrieure. Dans ce cas on parlera de rqitablissement achevq6. Nqiveau 2 : Vq6gq6tation paucispq6cifique (n inf. 1 3 espq6ces). Niveau 3 : qWgqdtation monospq6cifique. Ce rq6sultat peut etre atteint de deux faqgons ; la rqigq6nqiration eat le fait d'une seule espq6cqe oqu It fait du rq6tablqisseq- ment complet d'un clone (cf. roseliqares). Groupe B recouvrement des especes pq6rennes infq6rieur q1 50 q% ; diversi- tq9 spq9cifique indiffq6rente : Niveau 4 : Recouvrement compris entre 50"et 25 % Nqiveau 5 : Recouvrement compris-entre 25 et 5 q7 Nqiveau 6 : Recouvrement infq6rieur a 5 % pour les espqaces pqGrennes, mais pouvant dq6passer 80 q% en ce qui concerne lea thqgropliytes. Dane ce dernier cas, il s'agit d'un recouvrement saqisonnier. Niveau 7 : Recouvrement nul aussi.qbien pour les espq6ces pq6-- rennes que pour les- espq4ces annuelles. 11 eat ainsi possible, en un lieu donnq6, de suivre les changements de statuts, et de comparer d'un site 2q1 0qVautre, 6qVamplitude de ces chan- gements et leur vitesse reqlative@et ceci pour une mqame ganme dhabqitats (cf. ci-aprqas). La figure 6, quant 4q5- elle, eat plus qsy4qnt24qMt6qique pu6qi8qsqu'elle i4qndique les 8q6carts des 2q9tats constat8qEs en Ao6qat 1981 par rapport 8q& ceux dq'Ao2qi0q3t 1979. 337 Code de la figure 6 I Pas de changement Passage du niveau. 7 au niveau 6 (i. e. acquisition d'un cou- vert phangrogamique saisonnier a therophytes). Changement d'Stat correspondant k I niveau Changement d'etat correspondant 2 niveaux Changement d'6tat correspondant A 3 niveaux et ce, jusqu'A la categorie 4 incluse. Passage d'un niveau inferieur a 3 A. un niveau 3, 2 ou 1 Passage du niveau 3 au niveau 2, de celui-ci au niveau. 1 Passage du ni'veau 3 au niveau 1 cas particuliers Evolution regressive, mais qui peut correspondre 1 des I gains spatiaux d'une espece sociale, au detriment d'une vegetation non concurrentielle. Secteurs plantes (zone de restauration de l equipe "amiri- caine" Pr. Seneca-Raleiga). 338 % FRARATS 6 %r4- 2 % ILE C2A',,N-Dr %. % 7one non toucli6e % M" Zone polluA-e non nettoyp-.e % 0 % Zone pollu6.e piftinge % Rem I- !a i % % % rt 0) Zone ftr6p6e rt % 0 ON (D I % % (D rr (D . :j 1-., Ell H. . % 0 0 F@ (D Ja @:3 0 FL rt r ,Dt 25m 06- OF dop op A" op 4L dF -or q LIO Figure 3. Marais 6-Etat en 1979 (16gende dans le texte). 340 0 J* CD dolp Ap 4e do ZZ xp J* imi C4 7 rigure 4. Marais 6-Etat en 1980 (16gende dans le texte). 341 op do A. e .00. 40 dp dP' 00 jo. JV J* e, op 00 E CN 01 lob- z -.mmmmmmmm@: 7 . ...... 79 Figure 5. Marais 6-Etat en 1981 (le'gende dans le texte). 342 f Ago df lop Or E o C4 -0,00 owl N Figure 6. Marais 6- YAstribution spatiale des 4carts de niveaux observds entre 1979 et jql.@j S16gende dans le texte) 343 Commentaires Mis a part un petit nombre de points depourvus de toute vegetation, y compris therophytique, tous les autres ont vu leur couvert vegetal, meme fragmentaire evoluer avec le temps. Ces transformations sont, dans la plupart des cas et selon nos conventionsprogressives - i. e. tendent vers une cicatrisation des espaces denudds. Celle-ci peut etre-accompa- gnee par une augmentation de la diversite specifique, lorsqu'il y a eu destruction selective d'une partie du contingent specifique de depart, mais non denudation. extensive. Cependant des- tendances regressives s'ob- servent plus-localement. Il existe une certaine opposition entre les parties Nord et Sud du marais (gauche et droite surles figures 2 a 6), de part et d'autre des deux chenaux majeurs medians. L'etat de la vegetation, en secteur Sud, ne depasse pas le stade peuplement therophyt, ce qui correspond, 43 mois apres-le dep0t de petrole, a la destruction effective presque totale des especes perennes dans-ces zones bordiares. On constate neanmoins (cf.figure 6) que la cicatrisation I partir de regenerations autochtones est comparativement assez rapide le long de la route bordant An Inizigo au Nord.. En trois ans, Vamplitude du changement est de ordre de deux ou trois ni'veaux, Une cause possible de cette regeneration plus rapide, qui interesse surtout les hemicryp- tophytes souche et lea chamaephytes ligneuses mais aussi les hemicryp- tophytes stoloniferes-telles- Puccinellia maritima, est a rechercher dans le ruissellement permanent ou la percolation laterale de sources provenant de I'lle. Vest effectivement en tate des chenaux et sur les marges que ce processus est le plus rapide. Lorsque Von compare les figures 2 et 6, on remarque que ce sont les zones pitinges qui prdsentent le plus faible taux: de reprises, exprimg par le progrs-, en recouvrement, des espZices vivaces. Il faut noter toutefois que la vegetation initiale, dansla haute slikke, le bas-schorre et le schorre moyen tait pauci - ou monosp6cifique. Le piftinement, ajoutg a Veffet imm6diat du p6trole sur des plantes appareil vegetatif essentiellement pigg a des effets drastiques et surtout durables. AinsI ces deux facteurs, l'un initial, Vautre consqe- cutif, maisencore opgrant actuellement, constituent-ils, en premire hypothZse, la raisonmajeure-du retard dans-le retablissement d'une veggdtation, du fait de leurs effets multiples.-, direct,- ou indirects. L'exposition aux effetadynamiques de la marige joue A-galement, aussi bien sur le plan de la reprise que.celui de Vinstallation d'in- dividus- nouveaux dans les espaces tr6s dnudst. L'exposition favorise, Gtant donng Vabsence de couverture vegetale, 1'6ro4qs6qion des berges, ce qu6qi conduit 4q9 des effondrementqs locaux qu6qi, a terme, peuvent entra2qTner des modifications dans les drainages, par occultation des chenaux les plus 2q6troitqs ou les moinqs.profonds. Maiqs cette exposition, comme nous 4qlq'avons dit plus haut, aq"q'p'our cons4q66qquence une destruction acc4qgl4q6r4qde de la couc20qhe coh8qdrente de p4q6trole d8q6pos8qge sur les s4q6dimentqs, 0qm6q@8qhe lorsque celleq-ci a 4qdt4qi tass0qGe. Des souches suqrvivantes ainsi lib0q6r4qges ont pu alors reprendre leur d4q6veloppement, par formation de nouvelles pousses a2qgrienneqsq. a partir de 4qbourgeqons advent0qifs. Par contre, en ce qui concerne la coloniqsatiqon dqe ces eqspaceqs, 32qVexposition aux effets directs de la mer joue comme un facteur contrai- 344 re. 11 n'en est pour preuve que installation reussie, dans des situa- tions misologiques homologues, des esp6ces annuelles, en Vabsence de ce facteur U. e. dans lea zones abritges, en position plus interne). La partie Nord du marais se difffirencie des autres pour deux rai- sons - sa pli2qysiographie, - la nature de sa couverture vegetale. La topographie et Vexistence de nomBreuse-s sources qui ont assurg un. auto-nettoyage- pr6coce du petrole deposg- dans cessecteurs distaux font que le degrE de destruction et de.destructuration du couvert veiged- tal 0qy a t6 comparativement plus fatble. Structuralement en effet, lea gophytes 1 rhizomes dominent et les parties ariennes, qui de toutes facons, ont une durge de vie limitee, ont joug le role de pige 1 pitro- le, sans que lea organes endoggis de perennance n'aient dti immdiatement atteints. Seule, la sous--strate des peuplements, composgie d'hemicrypto- phytes stolonifres, a gtg endommage, mais la reconstitution de ce ta- pis interstitiel eat en cours et mgme localement, dans lea zones de stagnation des eaux continentales, presque acheve, 11 resort de ces oBservations que dans un mme marais lea condi- tions et lea modalitg-s-de rdtablissement de la vegegetation sont trs va riges. Si la regnration, 1 proprement parler, eat engag6e, 1 des de- grgs divers et pour des raisons et avec des moyens divers, Vinstalla- tion de nouveaux individus, par colonisation directe ne s'observe pas, hormis le cas des esp6ces annuelles dans les lieux protq6g6s. Aussi peut- on dire que le retablissement est assurg par le developpement vegetatif des seuls individus ou clones survivants, sana qu'il y ait ajout quan- tttatif ultg-rieur significatif. Les evolutions ultdrieurea sont cependant dans 1 majorit- des cas conditionnelles et dependantes de la resistance des populations regene- rees 1 la pollution remanente diffuse ou directe, de 1 lea delais va- riable observes dans Itinitiation du processus ! (cf. figure 6). ANALYSE DIACHRONIQUE DE QUELQUES TRANSECTS PERMANENTS. Parmi les transects etablis en 1979 et regulierement suivis depuis, nous en avons selectionne trois, deux dans lea marais de 1'11e Grande, le dernier dans Vestuaire de Kerlavos, La localisation des deux premiers est indiqu6e sur la figure 1. - A Kerlavos, il a'agit d'un schorre moyen A Armeria maritima et Plan- tago maritima, ftrepe par decapage des dix premiers centimetres du sol. Le substrat plan est constitue de limon. Le mode d'exposition eat abri- t. En utilisant le code deqa figures 3 1 5, le niveau ae depart, en 1979 eat 7. - Marais 5-Eat d'An inizigo Transect d'orientation ENE-WSW, etabli dans une zone exposee en partie aux actions dynamiques de la maree et qui fut a la fois tres polluee tres pietinee et soumiee egalement au passage d'enginee. Deux petits chenaux recoupent ce transect qui se rele- ve legarement vers I'Est. Selon les sections de celuici, lea niveaux, de depart sont de 4 ou 5. 345 - Marais 3. Ce transect, situe en mode tras abrit6, et orientg NW- SE. Son point de dq6part haut eat situqg sur une levqA-e artificqielle non touchqge par le pq6trole, mais tr6s pidtinge. Il se. termine dans une zone mar6cageuse occup6e par une prairie 1 Juncus maritimus, encore actuelle- ment tras polluEe, mais non piktinie. Les ni'veaux de d6part vont de 1 Ces transects: illustrent ainsi lea differents cas de figures ren- contris, que ce soit sur le plan de la vegetation, des habitats, des destructions. Legendes. des figures 7A - 7E, 8A-8B , 8C, 9A -9E, Figures 7A, 8A, 9A. Les transects- sont constitug-s de carrge contigils de 0,50 x 0,50 m subdivisgs chacun en 25 cases-de 0,10 x 0,10 m, Les presences apcifiques sont relevg-ee dans chacume des cases. Vik chelle des hauteurs, dans le diagramme eat tablie co=e ci-apras -55-1010"1515- . -25 Presence de 1'espce dans n cases Afin de faciliter la comparaison des distributions limaires: et des, recouvrements sur la ligne, lea donnges relatives 1 chaque ann6e, pour une eapqace donnge, sont rapprochges. L'ordre de presentation des espaces, dans lea diagrammes eat con- ventionnel. Il s'appuie sur lea types suivants : - plante 1 organe de resistance endogg. [gophyte 1 rhizome micryptophyte A, souche - plante s-ans organe de resistance en- [chamaephyte ligneuse dog hemicryptophyte stolonifare - terophytes (y, compris: espces Uannuelles) Figures 7B, 8B, 9B : sont indiqus: - le nombre de cases- par carrd occupdea par la vegetatqion, toutes- espaces confondues. - le nomBre d'eapZices prd-sentes dans, chaque carrg, Figure 8C : Pour ce qui eat du transect 5, nous avons prgsentg diffrem- ment leqa donnges, pa00qrla distinction des especes-annuelles. et p rennes et leur importance, exprim6e comme la somme des cases occu- pes- par chacune des especes. Ce.somme8q,8q- qsqont ensu6qite cumul4qgeqs, de 1q.4q1 lesq-d4q6pasq-qsementsq-posq7qsi0qbleqs danqs le ca8qs de cooccurence ou m0qgme de d8q6but de qatrat0qification. La qsomme des recouvrements individuels peut a0qins0qi 0qZtre sup8qgr6qieure aux 100 8q% d'un carr8qi 4q614q64qmentaire. C6qqd 4q@q-A8qT2qm mar. e8qi8qia.marit6qim q. .=Aster tri0qpoqliumq-Co 1q*1 gl q'qvQc0qblea- 84qVe 6qR. a i2qi a q10q1qfqtqlo e or20qWasqct20qA4q@92qR2q;12qf12qtncuqs mqa 0qJ2qNn A qmc4qi0ql2ql6qg0q@--u2qA4qTq"qI2qmqo 6qs0q'4q@8q'92q* 20qR4qi6qmq-qv80qNaoqrqr qr qS rq4 qI00qNsaq.8qFqI84qN mqcars eq-6q@ qU qn4q@4q@ 0q�,cQ qu q'rq'm=qOq'76qNqjq-ccine la 88qM20qpq"6qAq- al IV tilt 8qSq=8q@aqnq'2q@76qK88qH0qXe6qlsqn84qt6q@36qM96qlqca; 6q;2qra 76qUq=4qRqara0qRqnoqlq,6q@96qqq= 0qEa erequn.=8q8a18q;cqQr0qYq1a 0q?erennqI88qM32qPr2qE. m0qd2qfq.=0qT56q%q-6qg8qSq1ar1a marina2q;q-76qM2qA76qW68qRqA q.= q,uae a 4qArqitqim8qi; och n marqnq- 346 0 5 10 15 17 rn A OIQ r_ f--" rl M Trigl mar. MEIN- Salic. peren. a) CA rt P. rt Pucc mar. . ...... ....... P, :e o r- 'o :j rltM H. 9 J-40 0 0 0 Salic. div. SP. (Th) U) (D ....... ....... . . . . . rt 03 rlt rt ;J Fl En 0) (D rt Sperg, mar. H 0 M 0- Co M H ri. U) tl, 1979 SSW NNE (D 1980 ei U) rL M 10 M 1981 25. .............. .............................................. . . ......................... 0 (DI 0 n M i-D w FA M 20. M 0 rt M 15 rt CL Nombre de cases (D 1981 occup6es par la 10. v6,a-6tation, par carr6 1980 un carre- -- 5 25 cases de 0;10 ...... rn x 0,10 M. f 979 rt r--i r 0 17m t-3 7 ffrM .. ........ . ....... ...... . ...... . ...... Mz= Juncus mar. '71 M . .......... ........ .............1 ....... ..... Trigi. mar 1-3 ............. .......... ........... ..................... ................ . Inw. !....... ...... .................. Ln In U) rt M Plant. mar. ar r_ _j rt (D -o-Limon. vulg. V@ 0 -Arm. mar. In M 10 0 ........ .......... . ....... . ....... ....... ....... cczzzcc@ W Halim port. Ft INN&. (D H Salic. peren. LO .g, (D (n OC) Cj) Ln r ------- I ............. . . ....... . ..... Pucc. mar. ih _J111111111111 a MdMLme.. UMML_ (D Vh P. C== C 4 ............ . . ................. ................. ....... Salic. div. sp. (Th.) vMML_ ...... .................. Suaecla mar. .............. 07", nrrq Sperg. mar. 3Cochl. angi. Mwb=M -------- nT- ENE -Paraph. strig. 0 0 5 10 15 20 22m 25. FtBI TRANSECT MARAIS 5 r 20. r i I I r 15. r r r FU OQ r -t L L 10. r m J J- LJ 00 5. L 1 1979 .............. L-j... ............. . CL 1-3 .............. r- " .... (-I ---] .......... . ..." M CA Am 0 m o n ------Nombre de cases occup6es par carr6 de rt 25. 8B2 25 cases de 0,10 x 0,10 -,m m 20, ............... Nombre d'esp&ces par carr6 m 0 r rt :J m m 0 1& r rt fl r r P, @ I I I I I I I I r m L J I I I I X 10. L J r LJ 5. ............ Ln : :.... 1980 ............... M I " L7J k.... rn ......... H. 4' ib 16 rt 0) 1 io i2m (D% H Fh ID 25. 8B3 r? F@ L o LJ :1 L 20. r-L -1 r -r- H L---j 0 1 L J L r--j Fl (D 5.- ....... ........ 1981 . .......... J- ..... LJ .... i2m 8CJ Nombre cumulg- de points-esp6ces par carrE. TRANSECT MARAIS 5 1'4 30 OQ 0 H m 00 25 ---------------------------------------------------------------------------------------------------- --------------------------- ---- ---------------------- ... ........... c H 20. 1979 m co 0 M 15. 0 0 ... r. ct .4 L-LT m 10 rt:j m 5. Ca. 0 m ft U) I @1. --I- r- I --i -r' 0 5 1@ io 22m rt &0 m In 8c2 0 vivaces 35. m H El annuelles m rt 30. 0 m :3 rt 25 - --------------------------------------- ---- ----------------------------------------------- ---------- < 0 4 2Q n rt m r. a. 1980 15, . ...... -X-: m 10, 6F' 5 L* -, J. 0 5 lb 20 22m 8 C3 Nombre cumulg de points esp6ces 90. par carrd. 09 80 : vivaces m : annuelles 00 TRANSECT 14ARAIS 5 CL @3 70. M nm 0n r. m 60. m rt :j m r. ... ... ... 50. . . . . . . . . . . . . . . M ... ... ... t-n X., 0 40. m m m F4 ... ... X- 1981 rt0 30. .... ... ... ... 0 ... ... 0OQ ------------ ... ----------- -------------- ...... ..... MP. U) rr 20. . . . . . . . . . . . . . . . . . . . . mm rt 10- L L-LI 0 5 ib io i2m A 0 1P 15 2P -22 M H* -L F Juncus mac . ...... .. ..... ...... awl 1@ MAR=l M > I j t3 Trigl. mar. In ...... rt (D rt ED =3 rt (D 11 Limon. vulg 0*9 0. -- 1.., 7. V) (D 0) ct rtI Plant. mar. 13, (n (D 2D IF tj Arm. mar '0 . (DI MCQ rn C: M "Halim. port. -.7-Z Salic. peien. C=53 F-1 r= Pucc. nsar. WNW -IM Salic.div. sp(Th) ...... ....... UMM Sperg. mar. Nona- Suaeda mar-s- am" @O Cochl angl. Ast. trip. 25. 9BI r -i F--l 20. 1 1 1 1 TRANSECT MARAIS 3 L CQ L L J 10. L ti LO 1979 ............... ...... ...... ....... rt C? H. -1 L L ....... 0 10 1- A -1 ;j m 0 ------- Nombre de cases occup6es par Carr& de ................... Nombre d'espaces par carrg 22m 25 cases de 0,10 x 0,10 m 0 0 25. 9B2 m m F4 0 11 1, rt rt L J 20. r H, i-n 0 10. (D r f.... 1980 0 5 f I..... .............. t.......... i2m m lb 1@ I is lb 2b ;j rt (D 25- 9B3 (D 20. r 15. 17-1 10. L-j F4 rt 1981 . . . . . . . . '61 .......... ............... 0 i2m Commentaires - Estuaire de Kerlavos (figures 7A, 7B) Les especes vivaces ont un recouvrement presque nul en 1979, tandis que s'installent quelques pieds de Salicornia ramosissima et Salicornia. gr. herbacea (non distingudes sur les transects !). Le fait remarquable est la rapidit! avec laquelle,en moins de troia ans, les therophytes ont saturg lespace disponiqble, ce peuplement saison- nier ayant un recouvrement qui peut atteindre 100 %. Il s'agit d'une co- lonisation primaire active (stade I - especes "opportunistes" dites 1 stratigie 'r'). 11 faut noter 1 importance des centres de dispersion ini- tiaux (cf.Brereton, 1971). Les pied&-meres etablis en 1979 ont produit des grainies qui n'ont pas ete exportees etant donni le niveau topogra- phique des lieux et leurs-positions tres internes-dans,1'estuaire. Ll a ainsi capitalisation locale des graines-qui germent pratiquement aur place. La prisence saisonniere d'un revetement algal microphytique exer- ce une influence aussi bien dana le pigeage des graines deposees Sur le sol que sur leur germination, au printemps suivant (protection ther'- mique, hydrique, photique, mais aussi protection contre les agents dyna- miques). - Marais 5 (E-An inizigo) (figures 8A, 8B et 8C) Augmentation, entre 1979 et 1981 de la richesse specifique dans certaines sections du transect, notamment celles correspondant au schor- re moyen. Cette augmentation vient de reprises relativement tardives d'hemicryptophytes 1 souche et rosette telles Plantago maritima, Limo- nium vulgare et Armeria maritima dont Vimportance numerique reste nean- moins tras faible. De nouvelles therophytes apparaissent en 1981 : Coch- learia anglica et Parapholis strigosa. En ce qui concerne les-plantes perennes, et si Von met a part Juncus maritimus, il a gain dans le recouvrement de chaque population, Le phenomeane interessant est celui. du decalage progressif des optimums, d'une annie sur 1 Vautre, ce qui peut traduire deux faits : . l'importance de la competition interspEcifique, du fait de la mise en contiguitS de clonea regeneres, et qui se sont ensuite develop- pea lateralement d'une facon centrifuge. . la biologie de la regeneration qui favorise un developpement plus actif des parties du clone les moins end agges. Si cette partie est en situation marginale, par rapport au clone ancien, il peut y avoir double-mouvement, normalement vers VextErieur, mais aussi vers VintA6 rieur (developpement centripate) assurant afnsI une reconquete des posi- tions anterieures. Un clone epargn, developpement plagiotropique dominant, se cica trise lui-mgme tout en gagnant des espaces, 1 sa peripherie, qu'il n'oc- cupt pas prEcEdemment du fait de 1 occupation deslieux par d'autres especes ou par des clones de la meme espece (on peut rattacher ce pheno mane 1 celui du "die-back" observE chez lea especes A extension vegetati- ve). Cleat ainsi que ce processus peut conduire Ides. redistributions spatiales de dominance, apres perturbation, telles celles signalees par Balker (1973). 354 De ce point de vue, des especes stoloniferes ou radicantes comme Puccinellia maritima et Halimione portulacoides ont un developpement spatial qui s'accelere avec le temps et qui est rapide compte-tenu du nombre de pieds survivants au depart. Comme dans la section du transect oqa ces deux espaces se developpent, le terrain est plan, donc la pente, via les durges et friquences de submersion n'est pas un facteur limitant, Vaire actuelle de ces plantes tend A rejoindre leur aire. potentielle en Vabsence de concurrence. occupation de Vespace par les annuelles montre la mgme tendance qu'A Kerlavos. Ilya saturation des espaces interstitiels. Le mime pro- cessus de capitalisation 1 partir des pieds-meres s'observe encore ici. De plus, apparalt un phenomene de nucleation (Yarranton et Morrisson, 1974). En effet, toute vegetation dans un espace intertidal soumis A submersion est un obstacle pour les particules sedimentaires-et les se- mences-veghiculges par Veau. Dans un second temps, ces semences qui peu- vent provenir de plantes perennes galement, germent dans le lieu oq elles se sont deposges A la faveurde&etdans.-les sediments meubles depo- sds A la base de 1'dbstacle. On observe de la sorteldans les secteurs oil ce phenomene se produitya la fois une acceleration autoentretenue de la sedimentation.et,d'une facon concomittante, une acceleration de la colo- nisation accompagnee d'une augmentation de la diversite specifique loca- le. Cette phase d'augmentation transitoire de la diversite- est bien con- nue: transitoire car elle est suivie (cf. marais 2,remblai artificiel, non etudie dans cette communication) par une legere decroissance de la richesse specifique, consequence du comportement "impErialiste" de cer- taines espaces "couvre-sol". L'observation de la figure 8B montre, a un autre niveau le phenome- ne dejA decrit A Kerlavos de saturation du plan -toutes especes confon- dues- egalement 1 augmentation de la diversite dans certaines sections, en meme temps que la variation horizontale de celle-ci, entre 1979 et 1981, traduisant 1'heterogeneite, A grande chelle d'un tel espace. Vexplication est plus A rechercher du cStg de la competition interspeci- fique que de celui de contraintes mesologiques (types biomorphologiques compatibles ou incompatibles, cf. discussion in Levasseur et al., l.c.). La figure 8C montre la part prise, des 1980, par les therophytes, part croissant tres brutalement en 1981. Mais l acceleration la plus forte se tient precisement la oa une vegetation vivace dejA installge fait protection alors que dans la partie gauche du transect, plus expo- see, la regeneration d'esperes vivaces est moins active-ou la des- truction initiale de celles-ci plus complete les annuelles son' nume- riquement moins abondantes. - Marais 3 dit de Notenno (figures 9A, et 9B) Ce transect est d'une interpretation plus delicate que les precedentS etant donne l heterogeneite des situations presentes et les tendan- ces en quelque sorte inverses qui s'y developpent depuis 1980. Ce qui frappe A premiere vue dans la figure 9A est le desequilibre entre les parties gauche et droite du diagramme pour ce qui est de la diversite specifique en chaque point. Deux raisons peuvent etre evoquees 355 1 - appel A la notion de composition floristique initiale (-Cf- Egler, 1954), fonction, dans ce cas particulier, de gradients msologi- ques le long de cette toposdquence induisant en particulier une riches- se specifique maximale dans la partie moyenne du gradient topographi- que, mais avec cooccurence des limites basses. 2 - l'opposition peut aussi etre interprete comme le resultat des perturbations differenciees qui se sont exercdes et qui continuent de slexercer dans ces lieux, notamment dans les parties deprimes du tran- sect et qui marquetl'influence effective de 1'epandage du petrole, et ce jusqu'A la limite supdrieure de depot. On notera a ce propos que la cicatrisation a 6tq6 rapidement effective au dessus de cette lqimite et ceci des 1980, alors que le recouvrement vegetal reste discontinu en dessous (cf figure 9B). Le fait A retenir reste cependant le remplacement de la population primitive a 'Juncus maritimus, seule espece rescapee en 1978 par une es- pece de type biomorphologique different, Halimione 2ortulacoides -cha- maephyte ligneuse A tiges radicantes- qui etend son aire, dans les par- ties basses, depuis 1980. Dans Vespace disponible libre, au moins au niveau et au dessus du sol ' le meme phenomane de colonisation en nap- pe par les Annuelles observe, de la part de Salicornia du groupe her- bacea, tandis que les autres therophytes -y compris des formes annuelles d''Aster tripolium- S'installent, A leur niveau. bionomique habituel sur les parties moyennes et hautes du transect. L'evolution des performances de Juncus maritimus, geophyte a rhi- zome est intgressante car trs representative de la tendance generale au declin qui affecte, en divers lieux, et ce depuis 1980,des populations entieres de cette espece. Celle-ci, rappelons-le, est une des plantes qui occupait et qui occupe encore la plus grande partie des marais 3, 4, 5 et 6 et qui a etqi considgrge par nous en 1979 comme une plante rg- sistante. Comme son rele physionomique, structural et coenotique est con- sidgrable, toutes modifications dans sa distribution et-son abondance spatiale pourront avoir, a plus long terme, des incidences certaines. Les donnAkes quantitatives suivantes (figures 10 et 11) illustrent de telles tendances regressives. La figure 10 presente 1'evolution sai- sonuiere du rapport biomasse dpigge sur necromasse agrienne exprimees en poids sec. Les donnges de base sont le resultat de fauches effectuees au ras du sol, dans trois quadrats de 0,50 x 0,50 m, pris au hasard a trois niveaux topographiques, les recoltes 6tant ensuite sechges 48 heures a 65*C. On notera Vinversion du rapport . partir de la fin de l'annge 1980 et A terme, ceci conduira a la disparition de Vespace dans ce lieu. D'une faqon concomittante, les capacites de reproduction sont alte- rees et se traduisent, selon les cas, par une reduction du nombre de ti- ges fertiles', par des malformations de 1 inflorescence et, pour ces der- nieres , par une diminution du nombre de rameaux floriferes, par 1 absence d'etamines ou la non-formation de capsules et de graines ou seulement par la production d'un petit nombre de graines avortees. Dans le meme temps, les pieces. du perianthe peuvent revetir un aspect brac- tiforme (cf. figure 12). De plus, l'observation de coupes transversales de rhizomes montre une aecrose des parenchymes; ceci, ajoute 1 un dessechement des apex vegetatifs, pose le probleme de la nutrition hydrique et minerale de (*) cf.figure 11. 356 Altitude relative des populations de Juncus maritimus 6chantillonn6es dans le marais 3: Rapport biomasse 6pigde (PS) Le ni,@eau de r6f6rence est celui atteint par les marges de n6cromasse (PS) coefficient 100 0 1 0 cm ........... Il -60 cm 111 -100 Cm R 3.8. 3,4. '4 3. S. Nil 4& 2.2. V, 1,4. ......... A@ All F 0.6- 0.2 J; it t 6 m 94 A 1@ it N 79 791 :80 81 81 I Figure 10. Evolution saisonni6re du rapport biomasse 9pigge sur n6cro- masse agrienne. PROPORTION DE TICES FERTILES PAR RAPPORT AU NOMBRE TOTAL DE TICES AERIENNES 26, Altitude relative des populations de Juncus maritimus 6chantillonn6es dans le marais 3: le niveau de r6f6rence est celui atteint par les 24 mar6es de coefficient 100 0 22. 0 cm A ................ A 11 60 cm 1 100 cm 20, 18 16 14 Ln 00 12- 6 V A .......... 'A A - - - - - - 4 I.A \ 2 V Jn it A S 0 NJ tA m it S N D F m A it S N 79 791 ISO, so 81 81 Figure 11. Evolution saisonni6re du nombre de tiges fertiles parrapport au nombre de tiges agriennes (s=0,150 m2). AI 1979 1981 .Figure 12. Yorphologie compar6e d'inflorescences de Juncus maritimus Lam. provenant dun m9me clone situ6 au S du marais 4 (cf.figu- re 1) pr6lev6es en Aoat 1979 et Aoiat 1981. 359 ces organes, donc des effets A long terme du p6trole, toujours present, sur la physiologie et sur certains mq6tabolismes de la plante. Apparem- ment, il n'y a pas renouvellement des reserves dans le rhizome : celles subsistant aprqas 1978 ont maintenant q6tq6 consommqSes. Il faut rappeler A ce propos les conclusions de Baker (l.c.) relatives A la sensibilitqg de cette espq6ce A des pollutions chroniques par les hydrocarbures. L'observation simultanqge des deux courbes montre la rqdalitqg d'une tendance exprimq6e de deux faqSons diffq6rentes et qui ne touche pas seu- lement 0qVappareil reproducteur ! Rappelons cependant qu'une "allocation d'qEnergie" plus forte en faveur de 0qVappareil vq6gq6tatif est considqgrq6e comme classique par Ranwell (1972) Chez les plantes des marais mari- times4qAa veritable question relative A la nature et aux dqilais de rqita- blissement de la vegetation, dans les marais de l'Ile Grande,est 1qA. Il est vraisemblable que ce rq6tablissement aura lieu, naturellement mais aussi avec 2qVaide des plantations volontaiqres; mais, sur de grands espa- ces encore occupq6s par le qJonc, il aura lieu sans lui, ce qui pourra mo- difier passablement le paysage vq6gq6tal, mais aussi les strategies de res- tauration, celle-ci ne pouvant alors 0qkre.uniquement foqc-alisqge sur les seuls secteurs actuellement denudes. CONCLUSIONS Les quelques exemples prq6sentq6s montrent qu'au delqa de la variqgtqg constatq6e, un processus de rq6tablissement du couvert vq6gq6tal est effec- tif en de nombreux lieuxq; les gains en recouvrementypar rapport A la situation de 1978, repqrq6sentent environ 35 qZ. Il est vraisemblable que localement on assistera a une acceleration du phq6nomqane puisque par nu- c1qdations en challne, le nombre de points d'agglutination des sediments et des semences va croqltre d'une faron non linqgaire. Il reste encore des sites oqa la destruction de la vq6qgq6tation a q9tqe totale. Pour diffq6rentes raisons ils demeurent stqgriles,en ce sens que des germinations ne peuvent s'y effectuer. Aussi un rq6tablissement natu- rel y est-il peu probable au moins dans un avenir proche. Il semble alors que des restaurations au moyen de plantations soient la seule voie rq6aliste possible, comme en tq6moignent les succq@-s enregistrq6s, A la sui- te des deux annqges d'expqgrimentations menqges dans ces secteurs par le Dr. Seneca et son q6quipe. En effet, 2qVintroduction de boutures, selon les cas, avec ou sans sol, leur reprise et leur dqivqgloppement ultq6rieur, montre a contrario que c'est la phase "germination" qui est inhibqg-e et donc qu'en la court-circuitant, on accqglq6re lqa cicatrisation. De la mqgme maniqare, comme ces boutures font obstacle, d'autres espq6ces peuvent alors qsq'6qimplanter naturellement, mais dans un second temps, dans ces lieux, r4q6tablissant une diversit8qg sp8q6cifique qui nq'q*existait pas au depart lors de 1q'exp4qgrience. Notons quq'un ph4q6nom8q6ne similaire peut 0qetre initi2qg par les nappes de th4q4rophytes qui marquent fr4q6quemment la premi0q6re phase de la recolonisation. Si Von compare 32qV8qitat actuel de la vegetation avec lq'8q9tat primitif, on constate que le r8q6tablissement de celleq-ci passe en de nombreux lieux par une redistribution spatiale des especes, au profit dq'un petit nombre dq'entre elles. Cette redistribution, qui peut quel6qquefois aller jusquq'4qA la monopolisation (transitoire ?) dq'un espace peut avoir deux causes 360 les plantes survivantes -initialement "resistantes ne possedent pas des capacites d'extension vegetative suffisantes pour cicatriser les espaces interstitiels denudes, alors que d'autres especes,"sensi- bles" celles-la aux premieres perturbations presentent ces qualites de part leur organisation et leur ethologie; de 1 cart actuellement observe entre la composition floristique initiale du site et la compo- sition du moment. - des especes tout-A-fait resistantes, telles Juncus maritimus et dans une moindre mesure, Triglochin maritima, deviennent sensi- bles A la pollution chronique qui affecte maintenant ces marais. Leurs populations, en declin, sont envahies peripheriquement par des especes autrefois cantonnges, du fait de la saturation de 1 espace par les pre- miares en dehors des clones les plus denses. Ce sont d'ailleurs les memes espaces qui jouent ce rele dans les deux cas, a savoir Puccinellia maritima et Halimione portulacoides,toutes deux capables, par stolons ou tiges radicantes,de couvrir le sol, mike lorsque celui-ci est encom- brd, a un niveau endogg,par des souches ou des rhizomes qui se maintien- nent longtempsapres la mort de la plante. La resistance d'une plante est donc une notion tres relative, elle est en quelque sorte individuelle et thematique mais 1 organisation fu- ture d1uncouvert vegetal apres perturbation doit autant aux plantes dites resilientes qu'a des espAces resistantes" en nombre insuffi- sant ou devenant sensibles d'autres causes que celles qui avaient au- torisg la resistance de depart. La soi-disant robustesse d'un tel cosystZzme tient plus A ses ca- pacites.de cicatrisation via des colonisations periphEriques ou implan- tations directes, lorsqu'elles sont possibles, qu'a la resistance alea- toire, A plus long terme, d'autres espekes. Mais encore y a-t-il une nuance fondamentale entre la reaction vis-a-vis d'une perturbation exceptionnelle, mais finie dans le temps et une perturbation qui devient chronique et qui n'a pas et integrge dans le pass par exemple au moyen d'une selection particuliere d'espaces. Vest peut-etre ce qui est en train de se dessiner actuellement. Encore faudrait-il, usteurs vaut mieux en effet parler ecosystemes superposes ou inclus dont les caracteres qualitatifs, structuraux et dynamiques sont differents au travers des types biomorphologiques reprgsentgs,-deleur abondance rela- tive de leur distribution spatiale. Une mgme perturbation s'exercera alor; d'une fagon selective et differenciee sur les elements composant une vegetation locale, del& les delais et les modalites differentes du retablissement consecutif. Celui-ci pourra mame ne pas PEtre possible : une Spartinaie altgrge ne pourra etre reconstitue que par la Spartine ellememe. Plus fondamentalement, la nature, 1 abondance relative, la dis- tribution spatiale des especes presentes ou apparaissant pendant la succession pourront etre soumis a variation, changeant dans un premier temps la composition mais aussi la structure des peuplements en cours de retablissement, ceci a l'interieur de certaines limites imposees par 1 environnement mesologique. Ces ecarts et ces divergences, par rap- port a 1 etat ancien, ne constituent pas des phenomenes "anormaux" et/ ou eventuellement inquietants. Ils representent seulement la materiali- sation instantanee du processus fondamental qui conduit a une saturation par la vegetation de 1 espace disponible, lorsque l'opportunite s'y pre- te, comme c'est le cas en ce moment. 361 REFERENCES Abbayes, H. des, G. Claustres, R. Corillion et P. Dupont, 1971, Flore et v6g6tation du Massif armoricain. I. Flore vasculaire, P.U.B. St-Brieuc, 1226 pp. Baker, J.M., 1973, Reco,very of salt marsh vegetation from successive oil spillages :,Environ. Pollut., vol. 4, pp. 223-230. Baker, J.M., 1979, Responses of salt marsh vegetation to oil spills and refinery effluents:in Jefferies R.L. and A.J. Davy (eds.), Ecological processes in--coastal environments, Blackwell Scien- tific Publ., Oxford, 684 pp. Brereton, A.J.,1971, The structure of the species populations in the initial stages of salt-marsh succession : J. Ecol., Vol. 59, pp. 321-338. Egler, F.E., 1954, Vegetation science concepts : I. Initial floristic composition, a factor in old field vegetation development : Ve- getatio., vol. 4, pp. 412-417. Levasseur, J., M.-A. Durand et M.-L. Jory, 1981, Aspects biomorphologi- ques et floristiques de la reconstitution d'un couvert v6g6tal phangrogamique doublement alt6rg par les hydrocarbures et les opgrations subs6quentes de nettoiement (cas particulier des ma- rais maritimes de l'Ile Grande, C6tes du Nord) : in AMOCO-CADIZ, Consg-quences d'une pollution accidentelle par les7hydrocarbures, C.N.E.X.O., Paris, 881 pp. Ranwell, D.S., 1972, Ecology of salt marshes and san.d dunes, Chapman and Hall, London, 258 pp. Yarranton, G.A. and R.G. Morrisson, 1974, Spatial dynamics of a prima- ry succession : Nucleation : J. Ecol., Vol. 62, pp. 417-428. 362 RESTORATION OF MARSH VEGETATION IMPACTED BY THE AMOCO CADIZ OIL SPILL AND SUBSEQUENT CLEANUP OPERATIONS AT ILE GRANDE, FRANCE Ernest D. Seneca and Stephen W. Broomel INTRODUCTION General Tidal salt marshes functon to stabilize estuarine shorelines, to exchange nutrients with sediments and the surrounding waters, to provide energy as detrital material to the estuarine food web, and to serve as nursery grounds for many commercially important fish and shellfish. Because competing land uses have resulted in a decrease in areal extent of these valuable resources in the past, there have been concerted efforts recently to preserve the remaining marshlands and to reestablish marshes at selected sites. Techniques and procedures have been developed to: (1) reestablish marsh in areas where Man has destroyed natural marsh, (2) reestablish marsh along shorelines where storms have damaged or destroyed natural marsh, (3) establish marsh along canal banks and shorelines to stabilize the substrate and retard erosion, and (4) establish marsh on dredged material (Woodhouse et al., 1974; Garbisch et al., 1975; Seneca et al., 1976). Our research efforts in marsh establishment along the southeastern coast of the United States led us to respond to an invitation from the joint scientific commission of the National Oceanic and Atmospheric Administration (NOAA)/Centre National pour 1'Exploration des oceans to study the effects of the Amoco Cadiz oil spill. We developed a proposal for restoring*marsh at the Ile Grande site adapting techniques and procedures developed for S2artina alterniflora Loisel. in North Carolina (Woodhouse et al., 1974; Seneca et al., 1976) to restoration of a part of the Ile Grande marsh using vegetation indigenous to that region. This interim report contains results from 2 years' marsh rehabilitation research at Ile Grande and a nearby estuary at Kerlavos. Literature Review The effects of oil pollution on salt marsh vegetation have been studied and reported by European researchers. Based on observations of Welsh salt marshes affected by oil spills from the chryssi P. Goulandris in January 1967 and the Torrey Canyon in March 1967, Cowell (1969) rated susceptibility of marsh plants to crude oil and concluded that salt marshes dominated by@ Spartina townsendii H. and J. Groves and Puccinellia maritima (Huds.) Parl. were most subject to damage. Stebbings (1970) studied the effects of oil from the Torrey Canyon spill on salt marshes in Brittany and found that these marshes were 1) Department of Botany and Department-of Soil Science, respectively North Carolina State University, Raleigh, North Carolina U.S.A. 27650 363 able to withstand 2 to 10 cm of oil with only slight, short-term, floral composition changes. Apparently, most of the toxic fractions had been lost from the Torrey Canyon oil, since it had been weathered at sea for 2 to 18 days. Stebbings noted that oil appeared to form an impervious layer on the substrate preventing gaseous interchange between soil and air, causing reducing conditions in the mud, and ultimately chlorotic symptoms in plants. Stands of Agropyron pungens (Pers.) R. and S., Festuca rubra L., Juncus maritimus Lam., and Scirpus maritimus L. were extremely vigorous and seemed to derive some nutritional benefit from the breakdown products of this Torrey Canyon oil. Cowell and Baker (1969) noted that populations of annuals such as Suaeda maritima (L.) Dum. and Salicornia spp. near Pembroke, Southwest Wales, were reduced initiallyTut were recovering a year after oiling from the Chryssi P. Goulandris. Halimione portulacoides (L.) Aell. was the plant most badly damaged. in June 1968 the plant species with the greatest coverage in the upper, middle, and lower marsh (Festuca rubra, Puccinellia maritima, and Spartina townsendii, respectively) had recovered completely (Cowell and Baker,, 1969). Baker (197la-i) reported on several aspects of the effects of oil pollution on salt marsh and concluded that single oil spillages do not cause long-term damage to marsh vegetation (Baker, 1971a). These studies indicate that marsh vegetation is resilient and often can recover from single oil spills. Baker (1971e) suggests that it is best to let an oiled marsh recover naturally. However, persistent oil pollution has killed Spartina marsh at Southampton Water (Ranwell, 1968). Such sites may develop extremely anaerobic conditions in the mud so that higher plants can no longer grow on them. Cowell (1969) states that repeated contamination is likely to have increasingly serious effects if anaerobic conditions are created due to bacterial use of oxygen in the biological oxidation of the oil. We found no account of marsh recovery after removal of the upper layer of marsh substrate and vegetation. Study Sites The Ile Grande site is a relatively protected estuary with a mean tide range of ca. 6 m, a spring tide range of ca. 8 m, and a mean tide level of ca. 5 m. Our first visit to Ile Grande was in December 1978. Our NOAA liason representative, Douglas Wolfe, indicated that the marsh west of the bridge at Ile Grande was to be our primary study site (Fig. 1). There were extensive stands of Juncus maritimus on both sides of the estuary with lesser stands composed of a mixture of species including Puccinellia maritima, Triglochin maritima L., Limonium vulgare Mill., Spartina maritima (Curtis) Fern., and Halimione portulacoides. There were vast areas with no vegetation cover, the result of cleanup operations by the French military to rid the marsh of Amoco Cadiz oil. In many areas only the aboveground marsh vegetation and associated oil had been removed and in other areas the entire marsh surface including the root mat had been removed to a depth of over 30 cm. The intertidal creek banks were almost completely lacking in vegetation cover. A limited number of substrate samples from the disturbed sites were taken which subsequently indicated a 364 91ML AWNAA.&,@ AMP. -nv 2, 74 ww FIGURE 1. Marsh west of the bridge at Ile Grande. Area without vegetation is due to removal of oil and vegetation during Amoco Cadiz cleanup operations. material which was sandy loam in texture and low in nitrogen and phosphorus. Marsh vegetation adjacent to the disturbed sites indicated that prior to the oil spill the natural marsh was composed primarily of Juncus maritimus, Puccinellia maritima, Triglochin maritima, Limonium vulgare, with ' lesser amounts of Spartina maritima. Halimione portulacoides was dominant along the creek banks. We noted considerable variation in the relative dominance of these species and others within marshes in the vicinity. Spartina anglica C. E. Hubbard was present only at a single site at Ile Grande as a small clump less than 3 m in diameter. This species is abundant in the Bay of Mt. 'St. Michel some 125 km to the east of Ile Grande. Juncus stands generally occupied the highest elevations of the marsh relative to the other species mentioned. Subsequent observations indicated that the Juncus marsh is flooded for about 3 days each spring tide cycle. Above the level of Juncus there was in some areas a narrow fringe of Festuca rubra and Agropyron pungens with associated species. many salt marsh ecologists consider this vegetation to be a part of the marsh. This higher zone of vegetation which extends up to ca. I m above the Juncus marsh is flooded relatively infrequently on extremely high storm tides and spring tides. It lies above the marsh impacted by Amoco Cadiz oil and cleanup operations. Our marsh rehabilitation efforts were confined to elevations from 0.8 m below to 0.3 m above that of the Juncus marsh. 365 Because we wanted to compare the marsh at Ile Grande with other marshes in the vicinity, we also visited the marsh in the estuary at Kerlavos ca. 5 km from Ile Grande (Fig. 2). This marsh contained less Juncus, no Spartina, and was dominated by Puccinellia maritima, Armeria maritima (Mill.) Willd., and Triglochin maritima along with Plantago maritima L., Cochleria officinalis L., Halimione portulacoides and Aster tripolium L. There was evidence of marsh removal by cleanup operations in the Kerlavos marsh also, but it appeared that the marsh was much less heavily impacted than that at Ile Grande. We chose to use this marsh area as a supplemental study site. NOW- 77W FIGURE 2. Marsh in estuary at Kerlavos. Areas without vegetation represent sites of marsh removal during Amoco Cadiz oil cleanup operations. PROCEDURE Substrate Substrate samples were taken at the transplant source sites for each species and also at the experimental planting sites each year. These samples were analyzed for elemental concentrations, pH, organic matter, and volume weight by the North Carolina Department of Agriculture Soil Testing Division using their routine methods. 366 1979 Plantings Based on our preliminary plantings made in December 1978 and the nutrient analysis of initial substrate samples, we established 9 experimental plantings in May 1979, using primarily Puccinellia maritima (Fig. 3), to a lesser extent Juncus maritimus (Fig. 4), and to a lesser extent still because transplants were not locally abundant, Spartina maritima (Fig. 5). These experimental plantings were designed to determine transplant response to conventional ammonium sulfate + concentrated superphosphate and slow release (Mag Amp and Osmocote) fertilizer materials at different rates over a wide range of tidal elevations. All transplants were taken from the natural marshes at Ile Grande and Kerlavos. Digging of transplants was confined to smal 1 areas along narrow drainageways (Fig. 6) and protected sites so as to impact the marsh as little as possible. Half of the 2900 May transplants were plugs (16 to 15 cm deep cores from 5 to 7 cm in diameter composed of root material with attached substrate) and half were sprigs (root material only) (Figs. 7, 8, 9). Holes for the transplants were made with a 6-5-cm diameter soil auger (Fig. 10). Transplants were spaced 0.5 m apart and the appropriate amount of fertilizer material was placed into the transplant hole prior to insertion of the transplant (Fig. 11). Planting was conducted just prior to the spring tide cycle so that transplants would be flooded shortly after planting. TZ PT @A' IRS FIGURE 3. Puccinellia maritima. 367 z, /A T FIGURE 4. Juncus maritimus. %i4 LO 10 �r. FIGURE 5. Spartina maritima. 368 -S. FIGURE 6. Digging Puccinellia along a narrow drainageway. e 15z' -6@el' WU@ FIGURE 7. Transplants of Puccinellia: sprig on left, plug on right. 0 I j-1 F@ 369 WrM W 4- FIGURE S. Plug type transplants of Juncus. N -4 cx FIGURE 9. Plug type transplants of Spartina. 370 It FIGURE 10. A 6.5-cm diameter soil auger used to make holes for transplants in experimental plantings. FIGURE 11. Osmocote (a slow release fertilizer material) + concentrated superphosphate. Black cup measures a dose of fertilizer (2.8 g N + 1.2 g P) per transplant. Holes for transplants are spaced 0.5 m apart. 371 J FIGURE 12. Triglochin maritima. 101, 4 3f _7 T S FIGURE 13. Plug type transplants of Triglochin. 372 In September, we made 9 additional plantings of Juncus maritimus, Puccinellia maritima, and Spartina maritima and established initial plantings of another species, Triglochi maritima. (Figs. 12, 13). Although not recognized as such on our initial visits to the site, the latter species appears to be a common pioneer species on disturbed sites alone or with Puccinellia maritima. Both Puccinellia and Triglochin appear to invade by seed. Height, number of stems, cover (a measure of spread) and aboveground dry weight per transplant were determined in September 1979, 4 months after planting. Cover determinations were made by measuring the average maximal diameter and the average minimal diameter of the transplant and using these dimensions in the formula for the area of an ellipse. Percent survival by transplant type and species was also assessed at this time and at each subsequent visit. Because our major objective was to establish vegetation, destructive sampling for biomass determinations was held to a minimum of three samples per treatment per location. A photographic record of all plantings was initiated. 1980 Plantings Based on results of our 1979 plantings, we established 14 additional plantings at higher elevations in May 1980 utilizing plugs of Puccinellia, Juncus, Spartina, and Triglochin and sprigs of Halimione (Figs. 14, 15). Remains of stems and intact root systems of Halimione indicated that this species was the dominant along the creek banks prior to the Amoco Cadiz oil (Fig. 16). Consequently, we began preliminary tests of reestablishing this species along the creek banks (Fig. 17) and included it in an experiment to determine the feasibility of nursery production for transplants. Like the earlier plantings, these 1980 plantings were designed to determine transplant response to fertilizer materials at different rates over a range of substrate and exposure conditions. Cover was determined for selected plantings. All experimental plantings were surveyed to determine relative elevations, i.e. relative to the natural marshes (Fig. 18). In September we made 8 additional plantings using primarily Puccinellia and Halimione with some Spartina. Based on results from our earlier plantingsr further planting of Triglochin seemed impractical. As in September 1979, all earlier plantings were assessed for survival, height and cover with sampling for dry weight determinations limited to only two plantings. Photographic surveillance was considered even more important because we did not conduct intensive destructive sampling. 373 ";7 AFAA FIGURE 14. Hali-mione portulacoides. 4@ A" FIGURE 15. Sprig type transplants of Halimione. 374 V, FIGURE 16. Creek bank without vegetation as a result of removal of oil and vegetation in cleanup- operations@, Old root systems of Halimione are visible on lower portion of banks. Note lack of natural marsh plant invasion of these sites at time of photo, may 1980. AW WABI& FIGURE 17. Making transplant holes along creek banks, May 1981. Experimental plantings made in May 1980 are visible in background on cre6k'bank. 375 7 4 FIGURE 18. Surveying to determine the elevation of our plantings in relation to that of the natural marsh at Ile Grande, May 1981. Note transplants on creek banks between surveyors and on right creek bank toward the village from the white stake 1981 Plantings Based on results of all earlier plantings, we established 21 additional experimental plantings in May 1981 utilizing about 4900 transplants at Ile Grande. Most of the planting effort was concentrated on establishing cover on the bare creek banks which were at this time beginning to erode due to decay of the binding root mat and undercutting by tidal waters (Fig. 19). Two rows of Halimione transplants (sprigs) were planted on the edge of the creek banks with two rows of Puccinellia transplants (plugs) adjacent to and toward the marsh along several intertidal creeks (Fig. 20). Many other areas, still bare of vegetation 2 years after the catastrophe, were planted to increase the probability of revegetation (Figs. 21, 22, 23). All transplants were spaced 0.5 m apart. Cover was determined for selected plantings. All experimental plantings were surveyed to determine relative elevations, and the photographic record was continued. 376 'j, e zF "NO ;77, FIGURE 19. Eroding creek bank at Ile Grande, May 1981. zT V@ I 4ZV le- L FIGURE 20. Creek banks that had no vegetation over 2 years af ter the catastrophe. . Each bank was planted with two rows of Halimione sprigs in May 1981. 377 FIGURE 21. Site at Ile Grande without vegetation prior to planting in May 1981. A, FIGURE 22. Making holes for transplants and applying fertilizer in preparation for planting at same site as shown in Figure 21, May 1981. 378 p 4: FIGURE 23. Same site as that in Figures 21 and 22 just after planting Puccinellia in interior and Halimione on the perimeter of area, May 1981. Nursery Plantings It' was obvious from our initial visit to Ile Grande that transplant sources could become exhausted as we began scaling up the planting operation. With rehabilitation of larger areas as a goal, we explored @the possibility of establishing nursery areas for two of the most promising species, Puccinellia and Halimione. The Puccinellia nursery a:rea was established at Kerlavos in May 1979 in conjunction with a type of transplant and fertilizer materials experiment. The nursery area for Halimione was incorporated into a fertilizer materials experiment with three other species at Ile Grande in May 1980. Both areas were refertilized with Mag Amp + Osmocote to determine the effect of fertilizer in addition to that applied at planting (Fig. 24). A limited. number of transplants were taken from each nursery area in May 1981 and compared with transplants of the same species taken from the natural marsh in experimental plantings at Ile Grande. Another approach to the problem of transplant propagation was undertaken in a joint venture with Monsieur Levasseur in 1981. He took Puccinellia plants from a natural marsh, transplanted them into small plastic pots, and grew them in his garden in Rennes for several weeks in the spring (Fig. 25). These transplants were planted in an experimental plot at Ile Grande in May 1981 to compare their growth response with transplants taken from the natural marsh at the time of planting (Fig. 26). 379 A 'T' it' FIGURE 24. Puccinellia transplants in September 1981 being refertilized with Mag Amp + Osmocote 16 months after planting at Kerlavos. 16 2, FIGURE 25. Puccinellia transplant grown by Monsieur Levasseur in his garden for several weeks prior to planting at Ile Grande, May 1981. 380 .,10 1. *MEW - ;40' i 'oor 10- FIGURE 26. Experimental planting established with transplants shown in Figure 25, May 1981. Data Analysis Data were analyzed using the Statistical Analyses System (SAS) programs for analysis of variance and least significant difference (LSD) (Barr et al., 1976). All statistically significant differences were determined at the .05 level. Variability was generally high and not all data could be analyzed statistically. We feel that in f ield experiments of the type conducted on disturbed marsh sites that overall observations, photographs, and at times fragmentary data have to be interpreted and used as best they can even when statistical significance cannot be documented. Consequently, because of these conditions and the fact that data of this type are not readily available, we have included data in this report that we consider important even though the variability is high. 381 RESULTS AND DISCUSSION Substrate Results of analyses of substrate samples from Ile Grande indicated important differences among sites which affect plant growth. Samples from relatively undisturbed areas of marsh from which transplants were taken and from the root mat of marsh killed by oil but with no surface manipulation, had relatively high ammonium, phosphorus and organic matter concentrations and low volume weight and pH values compared to substrate below the undisturbed root mat and that exposed by the cleanup operations (Table 1). The low ammonium and phosphorus concentrations of the subsurface material definitely were limiting to plant growth. Data to be presented later in the report indicate that fertilizer materials were necessary for significant plant growth in these disturbed substrates. TABLE 1. Values for five substrate variables for seven sites (source of transplant sites for four species, two strata of marsh without vegetation but with surface not removed, upper stratum of a creek bank without vegetation but with surface not removed, and a site from which the marsh surface was removed) at Ile Grande. Organic Volume Ammonium Phosphorus matter weight Site (mg/dm3) (mg/dm3) M pH (g/cc) Puccinellia site 88 12 8.9 3.4 0.7 Triglochin site 67 30 8.3 5.4 0.7 Juncus site 56 19 6.5 6.0 1.0 Spartina site 117 15 5.8 3.3 0.8 Marsh without vegetationa upper 10 am (root mat) 50 32 4.8 5.9 0.7 below root mat 14 7 0.6 7.1 1.4 Creek bank (upper 10 cm)a 40 17 3.5 4.7 1.0 Site with marsh removedb 5 2 0.0 7.0 1.2 a Marsh vegetation removed as a result of cleanup operations but marsh surface not removed; examples of site in Figures 1 and 16. b Marsh surface including root mat was removed during cleanup operations. 382 Elevation All elevations are given in relation to the average elevation of the natural marsh at the particular site, Ile Grande or Kerlavos. At Ile Grande the average elevation of the Juncus marsh on the southwest side of the bridge was the reference datum., This average elevation is tied to a white mark on the rock wall of the bridge for which Mademoiselle Odile Guerin, who has worked on*the Ile Grande project, has elevation tied to a national datum. The average elevation is also tied to the concrete foundation of the bridge itself. At Kerlavos the average elevation of the marsh is tied to a bench mark at an electric station tower ca. 0.5 km from the study site. Based on relating water levels at the two sites, the Juncus marsh at Ile Grande is about 0.1 m above the elevation of the natural marsh at Kerlavos. At Ile Grande we planted Juncus, Puccinellia, Spartina and Triglochin over a range of elevatfon from 0.8 m below to 0.3 m above the average elevation of the natural Juncus marsh. Juncus transplants did not survive at elevations below that of the natural Juncus marsh and best survival occurred at 0.3 m above that of the natural Juncus marsh. Puccinellia transplants did not survive at elevations of 0.7 m below that of the natural Juncus marsh and survival was less than 10% at elevations of 0.5 m below that of the natural Juncus marsh. The best growth and survival of Puccinellia transplants was achieved in the range of elevation between 0. 1 m below and 0 - 3 m above that of the natural Juncus marsh. Sbpartina transplants survived at the lowest elevations of all species tested. Although Spartina transplants survived at elevations of 0.8 m below that of the natural Juncus marsh, growth was best at 0. 3 m below that of the natural Juncus marsh. Survival and growth of Triglochin was generally poor but its elevation response was similar to that of Puccinellia with no survival at elevations of 0.7 m below that of the natural Juncus marsh. At Kerlavos experimental plantings of Puccinellia and Triglochin were established over a range of elevation from 0.5 m below to 0.1 m below that of the natural marsh. The best survival and growth of these Puccinellia transplants occurred at 0.2 m below that of the natural marsh. Transplants at 0.4 m below the elevation of the natural marsh did very poorly. Triglochin transplants responded in a similar manner. Halimione was planted at elevations from 0.1 m below to 0.3 m above that of the natural marsh at Ile Grande. Survival and growth of these transplants were best at about 0.3 m above the elevation of the natural marsh, but survival was good throughout the range of elevations planted. At Kerlavos, Halimione was planted from 0.4 m below to 0.2 m below the elevation of the natural marsh. Survival and growth was best in the upper half of this elevation range. Plantings in General About 9,700 transplants have been planted at Ile grande and about 1,800 others at Kerlavos over the period May 1979 through May 1981 (Table 2). Although half of these transplants were those of 383 TABLE 2. Number of transplants planted at Ile Grande and Kerlavos for five species-for five dates from May 1979 to May 1981. Number of transplantsa by year by month by siteb Species 1979 1980 1981 May Sep May Sep May IG K IG K IG K IG K IG K Halimionec 0 0 0 0 332 108 220 0 2756 0 Juncus 518d 0 173 0 360 0 0 0 0 0 Puccinellia 1298d 718d 180 so 448 645 179 40 2186 0 Spartina 258d 0 62 0 105 0 85 0 0 0 Triglochin 0 0 117 40 447 105 0 0 0 0 Total 2074d 718d 532 120 1692 858 484 40 4942 0 a All transplants were plugs except as otherwise noted. b IG = Ile Grande, K = Kerlavos. C All transplants were sprigs. d Half were sprigs and half were plugs in May 1979. Puccinelliaf four other species were also studied intensively. We tested two different types of transplants of four species, spring versus fall planting for four species, conventional and slow release fertilizer materials over a wide range of substrate and elevation conditions for five species, and developed nursery areas for two species. These comparisons and tests resulted in the establishment of 61 separate experiments and plantings over about 0.3 ha (Figs. 27, 28, 29, 30). The smallest experiment contained only 27 transplants while the largest contained over 1,000 transplants. The results from selected plantings are contained in the sections of this report that follow. Although quantitative measures (survival, cover and dry weight) are important in assessing transplant response to 'fertilizer materials and local site conditions, qualitative measures of plant response such as sequential photographs can also be revealing and supplement data. One of our best documented experimental plantings is at Kerlavos where we compared sprigs and plugs of Puccinellia in several fertilizer treatments. The planting was established in May 1979 and after realizing the initial objectives, we refertilized the area to develop a nursery for Puccinellia transplants. The pictorial sequence shows the site prior to planting (Fig. 31), immediately after planting and initial fertilization (Fig. 32), 1 year after planting (Fig. 33), and 2 years after planting, 8 months after refertilization (Fig. 34). 384 74 Uj Z 7(- in 6 C4 FIGURE 27. Map of study area on northwest side of the bridge at Ile Grande showing location of experimental plantings (stippled areas). Base map by Monsieur Levasseur. 385 Tk + + a, 1 110 44 & 0 OL FIGURE 28. Map of study area on southeast side of bridge at Ile Grande showing location of experimental plantings (stippled areas). Base map by Monsieur Levasseur. 386 XLE C@ FIGURE 29. Map of study area on northwest side of estuary at Ile Grande beyond area in Figure 28 from bridge showing location of experimental plantings (stippled areas). Area is just beyond road from village to several houses on edge of estuary. Base map by Monsieur Levasseur. 387 ce% *C 0, P1 14 Jb 0 K toll C) Kerlavos 1/1100 FIGURE 30. Map of study area in estuary at Kerlavos showing location of experimental plantings (stippled areas). Base map by Monsieur Levasseur. 388 o Woft il M,O* wv@ FIGURE 31. Site for experimental planting at Kerlavos prior to planting in May 1979. FIGURE 32. Same site as in Figure 31 just after planting with Puccinellia sprigs and plugs on 0.5-m spacing in May 1979. 389 -4@ 7' AW, FIGURE 33. Same site as in Figure 32 in May 1980, 1 year af ter planting. There are six rows of sprigs and six rows of plugs which alternate with each other beginning with a row of sprigs on the extreme left. The tallest plants in the center are plugs in the Mag Amp + Osmocote fertilizer treatment. FIGURE 34. Same site as in Figure 32 in May 1981, 2 years af ter planting. The planting is now a Puccinellia nursery area. 390 Survival In September, 4 months after our initial planting, the survival of plugs was significantly greater than that of sprigs for all three species (Table 3). Survival averaged over transplant type was about 65% for both Puccinellia and Spartina but only about 50% for Juncus. Reevaluation of these May 1979 transplants 1 year after planting indicated that significant mortality of both transplant types occurred over winter (Table 3). Greater overwinter mortality occurred in plug than in sprig transplants for both Puccinellia and Juncus. These results suggest that whatever factors were causing mortality in the sprigs were still affecting the plugs and that it was simply taking longer to cause mortality in the larger transplant type. overall survival was less than 50% for both transplant types for all species, but still significantly higher for plugs than for sprigs. These relatively low survival percentages included plantings in unfavorable (low elevation, exposed, poorly drained) locations, since we were trying to determine response over a wide range of conditions. In the more favorable sites, at about the elevation of the natural Juncus marsh, survival of plug transplants was consistently above. 70% for Puccinellia. on those planting sites where 10% or more of the transplants survived through the second year, plugs continued to survive better than sprigs for Puccinellia and Juncus (Table 4). Of all the 1979 transplants, only those of Puccinellia on the better sites yielded survival values of greater than 60%. TABLE 3. Percent survival at 4 and 12 months for two types of transplantsa for three species for the combined plantings made at Ile Grande and Kerlavos in May 1979. Survival (%) by timeb by type Species 4 months 12 months Sprig Plug Sprig Plug Juncus 22 80 14 37 Puccinellia 48 84 31 47 Spartina 56 so 10 26 Averaged over species 44 84 29 37 a There were 230, 979, and 100 transplants of each type for Juncus, Puccinellia and Spartina, respectively. b Survival of plugs was significantly greater than that of sprigs within and over all three species at each of the two sampling periods based on chi-square analysis. The reduced survival of both transplant types over species between the two sampling periods was also significant based on chi-square. 391 TABLE 4. Survival of two types of transplants of three species 1 and 2 years af ter planting (May 1979) averaged over all locations on more favorable sites. Survival (%)a Species May 1980 May 1981 Sprig Plug Sprig Plug Puccinellia 64 87 63 82 Juncus 4 39 2 23 Spartina 10 26 4 10 a There were 379, 120 and 50 transplants planted of each type for Puccinellia, Juncus and Spartina, respectively. Except for Puccinellia transplants, these survival data from May 1979 plantings are not impressive. We were in the process of learning where to plant with regard to elevation, the best type of transplant to use, the appropriate species, whether spring was better than fall planting, and how to satisfy the nutrient requirements of the transplants with fertilizer materials. Survival data for the May 1980 plantings indicate a significant increase in survival over those of the earlier plantings (Table 5). These results indicate that we were making progress and that except for Juncus, survival values greater than 50% were achieved for all species tested. The seemingly erroneous survival value for Halimione 4 months after planting is due to aboveground material appearing dead but the underground material being alive and giving rise to new aboveground growth the following spring. A decrease in survival overwinter of greater than 14% was only noted for Juncus which experienced a 41% decrease. Spring appears to be a better season to transplant than fall but the data are incomplete at this time. Type of Transplant The best comparison between sprig and plug transplants of Puccinellia from our experiments is the data from Kerlavos where cover and survival were higher for plug transplants except for the conventional ammonium sulfate + concentrated superphosphate treatment where the cover of sprigs was higher than that of plugs (Table 6, Fig. 33). The reduced survival for both types of transplants and especially sprigs in the ammonium sulfate + concentrated superphosphate (2.8 g N + 4.1 g per transplant) treatment was due to a small depression without exterior drainage which occupied a portion of this treatment and in which transplants did not survive (Fig. 33). Increased salinity due to evaporation or waterlogged substrate conditions due to prolonged ponding could have contributed to the reduced survival of this treatment. Although this drainage condition was restricted to a small 392 TABLE S. Survival of sprig type Halimione transplants and plug type transplants of four other species at two sampling dates averaged over all locations; planted May 1980. Survival (%)a Species Sep 1980 May 1981 Halimione 45 52 Juncus 80 39 Puccinellia 95 83 Spartina 96 89 Triglochin 82 68 a There were 440 Halimione, 360 Juncus, 823 Puccinellia, 105 Spartina, and 822 Triglochin transplants planted. TABLE 6. Cover in September of 1979 and 1980 and survival in September 1980 for sprig and plug type Puccinellia transplants for six fertilizer treatments at Yerlavos; planted May 1979. Amount (g) Cover (cm2)a Survival per (%)b Treatment transplant Sep 1979 Sep 1980 Sep 1980 N P Sprig Plug Sprig Plug Sprig Plug Control 0 0 34 91 46 174 65 95 Ammonium sulfatec 2.8 1.2 249 128 338 292 78 95 Mag Amp + Osmocote 3 2.8 4.1 77 260 234 533 50 83 Ammonium sulfatec 2.8 4.1 154 188 177 275 40 70 Ammonium sulfate 2.8 0 79 108 135 264 75 95 Concd superphosphate 0 1.2 62 107 145 ' 206 75 78 Avg. over treatment 109 147 152 291 64 86 a Cover was ca. 5 cm2 for sprigs and ca. 25 cm2 for plugs at planting. Standard error of difference between equally replicated transplant type means and among fertilizer treatment means: 40 for September 1979, 57 for September 1980, n=10. b There were 40 transplants of each transplant type per fertilizer treatment at planting. c Source of P was concentrated superphosphate. 393 area (12 M2), and was an exception to the relatively uniform topography of the experimental site, it served to emphasize the importance of adequate drainage for plantings of Puccinellia. When cover is averaged over treatment for the period from September,1979 to September 1980, the cover value for plugs increased two fold whereas that for sprigs increased by about 40%. Average survival over this same period of time was 22% higher for plugs than for sprigs. Sixteen months after planting, cover for plugs was significantly higher in the Mag Amp + Osmocote 3 (estimated to last for 3 months) treatment and for sprigs it was significantly higher in the ammonium sulfate + concentrated superphosphate treatment (2.8 g N + 1.2 g P). The controls achieved only 14 and 33% of the cover of the best treatments for sprigs and plugs, respectively, over this 16-month period. Response to Fertilization Kerlavos Analysis of variance of cover and dry weight data of plug type transplants of Puccinellia on a disturbed site at Kerlavos indicated a significant response to fertilizer materials (Tables 7, 8). one year after planting the cover of plugs in all three fertilizer treatments containing both nitrogen and phosphorus was significantly greater than that of plants in those treatments which provided only nitrogen or phosphorus or neither (Table 7). These results emphasize the requirement for fertilizer materials on those disturbed sites which substrate samples indicated contained amounts of nitrogen and phosphorus which were too low for good initial growth of transplants. The dry weight of aboveground plant samples from the Mag Amp + Osmocote 3 slow release fertilizer treatment was significantly greater than that of plants from any other treatment at 4 and 16 months after planting (Table 8). Although cover in the Mag Amp + Osmocote 3 treatment was not significantly different from that in the two conventional ammonium sulfate + concentrated superphosphate treatments 1 year after planting, by 16 months after planting, cover of plants in this Mag Amp + Osmocote 3 treatment was significantly greater than that of those in any other treatment. Cover of plants in this treatment was about twice that of transplants in the second best treatment. The center row of transplants (plugs) in Figure 33 is the Mag Amp + Osmocote 3 treatment. These data indicate the advantage of a slow release over a conventional fertilizer material on this disturbed site for a relatively long period (16 months). We excavated the belowground portion of several healthy transplants in the Mag Amp + Osmocote 3 treatment and noted that the fertilizer material still present below the transplant could be identified after 4 months (Fig. 35). Cover in the control plants remained significantly below that of plants in those three treatments which provided both nitrogen and phosphorus 16 months after planting. Cover in these control plants was 394 TABLE 7. Cover of plug-type Puccinellia transplants on three sampling dates for six fertilizer treatments at Kerlavos; planted May 1979. Amount (g) per Cover (cm2)a,b Treatment transplant N P May 1980 Sep 1980c May 1981 Control 0 0 66 174 388 Ammonium sulfated 2.8 1.2 1. 22 292 638 Mag Amp + Osmocote 3 2.8 4.1 209 533 819 Ammonium sulfated 2.8 4.1 187 275 687 Ammonium sulfate 2.8 0 73 264 481 Concd superphosphate 0 1.2 106 206 501 Avg. over treatment 127 291 586 a Cover of a plug at planting was ca. 25 cm2. b Standard error of difference among equally replicated fertilizer treatment means: 48 for May 1980, 49 for September 1980, 107 for May 1981; n--10. Comparison of initial fertilizer treatments for May 1981 may not be entirely appropriate because of the refertilization. C Refertilized with Mag Amp + Osmocote 3 (2.8 g N + 4.1 g P per transplant) in September 1980 after data collection. d Source of P was concentrated superphosphate. 395 TABLE B., Aboveground dry weight of plug type Puccinellia transplants in September 1979 and 1980 for six fertilizer treatments at Kerlavos; planted May 1979. Amount (g) Aboveground per dry wt .transplant (g)a Treatment N P Sep 1979 Sep 1980 Control 0 V 1.9 10.7 Ammonium sulfateb 2.8 1.2 3.8 23.3 Mag Amp + Osmocote 3 2.8 4.1 10.1 52.5 Ammonium sulfateb 2.8 4.1 4.5 22.1 Ammonium sulfate 2.8 0 3.3 14.6 Concd superphosphate 0 1.2 3.1 15.5 sac 3.0 6.2 a Aboveground dry weight at planting was less than 1 g. b Source of P was concentrated su.perphosphate. I standard error of difference among equally replicated fertilizer treatment means, n=3. FIGURE 35. Excavated Puccinellia transplant 4 months after planting showing new roots (white) and slow release fertilizer material still in place. 396 significantly below -that of those same three fertilizer treatments 2 years after planting even though these initial control plants were fertilized with Mag Amp + Osmocote 3 in September 1980. Although comparison of the original fertilizer treatments is confounded and may not be entirely appropriate in May 1981 since all transplants were refertilized. in September 1980, it is interesting to note that the same three fertilizer treatments with both nitrogen and phosphorus continued to have cover values which were significantly higher than those of plants in any other treatment. Cover of transplants in the best fertilizer treatment achieved an average radial spread of about 10 cm annually (Fig. 36). At this rate of spread, these Puccinellia plants would achieve complete substrate cover in about 3 years after planting (Fig. 37). lk" FIGURE 36. A 2-year old Puccinellia transplant with an average diameter of ca. 60 cm and cover of ca. 2,800 cm2 or 112 times that of the transplant at planting. 397 4 FIGURE 37. Several 2-year old Puccinellia transplants that were planted 0.5 m apart. The substrate should be completely covered by these plants by May 1982. Ile Grande Analysis of variance of cover data of Halimione and Puccinellia transplants 1 year after planting on a disturbed site at Ile Grande indicated a significant response to fertilizer materials (Table 9). Best growth as measured by cover was achieved by both species in the Mag Amp + Osmocote 3 treatment (Table 9, Figs. 38, 39). Cover of Halimione transplants in this treatment was significantly higher than that of transplants in any other treatment except for the Osmocote 8-9 (estimated to last 8 to 9 months) + concentrated superphosphate treatment. The cover data for Puccinellia transplants indicate the advantage of slow release over conventional fertilizer materials at this particular site. Apparently leaching of the conventional fertilizer materials was a problem because of the coarse sandy substrate in this planting. Significantly greater cover of Puccinellia was produced by the slow release fertilizer treatments than by ammonium sulfate + concentrated superphosphate except where the rate of ammonium sulfate was doubled (5.6 g N per transplant). Differences among, the cover values of the Triglochin transplants are meaningless except to document the poor growth by this species under all experimental treatments. The response by Triglochin in the particular experiment is repr esentative of its response at several experimental sites and is the reason for our decision to delete it from 398 TABLE 9. Cover and survival of transplants of three species in May 1981 for nine fertilizer treatments at Ile Grande, planted May 1980. Amount (g) per Cover (cm2)a Survival (%)b transplant by speciesc by species' Treatment N P H P T H P T Control 0 0 326 319 26 100 100 87 Ammonium sulfate 2.8 0 108 305 14 60 100 73 Ammonium sulfated 2.8 1.2 128 302 25 67 100 93 Ammonium sulfated 5.6 1.2 100 460 12 100 100 93 Mag Amp + Osmocote 3 2.8 4.1 535 725 23 100 100 60 Osmocote 3 2.8 1.2 277 556 14 67 100 93 Osmocote 8-9 2.8 0.4 228 449 31 73 100 73 Osmocote 8-9 5.6 0.8 272 578 9 73 100 roo Osmocote 8-9e 2.8 1.2 366 611 21 60 100 93 S-f 103 93 7 d a Cover of a plug transplant for Puccinellia and Triglochin was ca. 25 cm? at planting; n=14 for Puccinellia, n=8 for Triglochin. Sprigs of Halimione had quite variable cover at planting, but generally less than 50 cm2; n=8. b There were 15 transplants per treatment for each species. C H = Halimione, P = Puccinellia, T = Triglochin. d Source of P was concentrated superphosphate. e Source of additional P was concentrated superphosphate. f Standard error of difference among equally replicated treatment means. 399 @7 4'1 4111W 10' FIGURE 38. Experimental planting at Ile Grande to determine response to fertilizer by four species, from right of stake: Puccinellia, Triglochin, Halimione, and Juncus just after planting in May 1980. 4 W*W. 0@ NIT, IPA! FIGURE 39. Same experimental planting as shown in Figure 39 in May 1981, 1 year after planting. First seven plants in foreground were refertilized with same fertilizer materials as in initial treatments (Table 9) in September 1980, 4 months after planting. 400 further consideration as a desirable species in our rehabilitation efforts. The relatively high survival of all three species indicates a marked improvement in our selection and handling of transplants as well as the selection of a favorable planting site. These survival percentages were based on 135 transplants per species in this particular experiment. The high survival of Puccinellia transplants coupled with the relatively high cover values as compared to those of the other two species indicates that our emphasis on this species is justified. Analysis of variance of cover data indicates that the Osmocote 8-9 slow release fertilizer material maintained the original cover of Spartina transplants at planting with very little growth through the first year (Table 10). Transplants in the control and ammonium sulfate + concentrated superphosphate treatments decreased in cover over the first year. Although growth was not good, survival of these and other 1980 Spartina transplants was consistently above 80% (Table 5, Figs. 40, 41). Growth of this species has been very slow in all our experimental plots but because it can occupy lower elevations than most of the other species, we plan to continue to experiment with it on a limited scale. TABLE 10. Cover and survival of plug type Spartina transplants in May 1981 for three fertilizer treatments at Ile Grande; planted May 1980. Amount (g) per transplant covera Survivalb Treatment N P (cm2) Control 0 0 11 93 Ammonium sulfatec 2.8 1.2 15 100 Osmocote 8-9d 2.8 1.2 29 80 a Cover of a plug transplant at planting was ca. 2 5 cm2. Standard error of difference among equally replicated treatment means = 6.1, n=6. b There were 15 transplants per treatment planted. c Source of P was concentrated superphosphate. d Source of additional P was concentrated superphosphate. 401 Opt, 4, FIGURE 40. Experimental plantings of Spartina at Ile Grande to determine transplant response to fertilizer materials just after planting in May 1980. ';7- memo law A- -lit FIGURE 41. Same experimental planting as shown in Figure 40 in May 1981, 1 year after planting. 402 Response to Refertilization Analysis of variance indicated that refertilized Puccinellia transplants produced signif icantly more cover than those not refertilized (Table 11). Without refertilization transplant cover increased by 2.1 times from September 1980 to May 1981 whereas with one ref ertilization in September 1980 cover increased 2.9 times by May 1981. Those plants refertilized tiwce (May and September 1980) increased their cover 3.8 times by May 1981 or by 1.9 times between May and September 1980 and by 1.9 times between September 1980 and May 1981. One refertilization increased cover 1.4 times that of the plants which were not refertilized over an 8-month period and two refertilizations increased cover 1.8 times that of the unrefertilized plants over a period of 1 year. TABLE 11. Cover of Puccinellia transplants at two sampling dates for three fertilizer treatmentsa at Kerlavos, planted May 1979. Cover (cm2)b Treatment Sep 1980 May 1981 Not fertilized 222 460 Refertilized Sep 1980C 221 645 Refertilized May 1980d 428 833 and again Sep 1980c a All treatments were fertilized in May 1979 at planting. b Standard error of difference among equally replicated treatment means = 115, n=11- C Refertilized with Mag Amp + Osmocote 3 (2-8 g N, 4.1 g P per transplant). d Refertilized with Osmocote 8-9 + P (2.8 g N + 1.2 g P per transplant). Response to Fresh Oil In the spring of 1980 oil from the Tanio reached the estuary at Kerlavos. Although it was observed on several of our 1979 transplants, we could not document any adverse effects. We decided to take advantage of the opportunity to plant in some of the fresh oil deposits along a creek bank. The marsh surface of the planting site had been removed in a cleanup of Amoco Cadiz oil earlier. The fresh Tanio oil was a superficial layer on the substrate which did not appear to penetrate into the substrate. In May 1980 transplants of Halimione and 403 Puccinellia were planted at one of these freshly oiled sites (Fig. 42). Cover and survival data determined 1 year later indicate no noticeable effect of the oil on either species (Table 12, Fig. 43) as compared to these data from transplants at unoiled sites which were planted the same year (Table 9). As in many other experiments, cover data indicated a significant transplant response to fertilizer materials with best growth realized in the Osmocote slow release treatment. TABLE 12. Cover and survival of Halimione (sprigs) and Puccinellia (plugs) transplants in May 1981 at a site oiled by the Tanio in the spring 1980; planted May 1980. May 1981 Amount (g) Halimione Puccinellia per transplant Cover Survival Cover Survival Treatment N P (CM2)a M (CM2)a M Control 0 0 86 67 381 100 Ammonium sulfateb 2.8 1.2 258 100 493 67 Osmocote 8-9c 2.8 1.2 631 67 550 100 a Cover of Halimione sprigs was quite variable at planting, but was generally less than 50 cm2; that of a Puccinellia plug was ca. 25 cm2. Standard error of difference among equally replicated treatment means; 100 for Halimione, 160 for Puccinellia, n=3. b Source of P was concentrated superphosphate. c Source of additional P was concentrated superphosphate. Creek Bank Plantings Creek banks with no vegetation cover are one of our top priority .planting sites. Preliminary plantings of Halimione made in may 1980 (Fig. 44) achieved over 90% survival and good growth by the following May (Fig. 45). Puccinellia plantings have also achieved good survival and growth over this period of time (Figs. 46, 47). About half of the over 4,900 May 1981 transplants were planted along creek banks (Fig. 20). A similar proportion of the overall planting effort is planned for creek bank sites in May 1982. 404 44 IN 7 FIGURE 42. Planting Puccinellia and Halimione in fresh oil from the Tanio in the estuary at Kerlavos in May 1980. rt Wf ZIP FIGURE 43. Same experimental planting as shown in Figure 42 in May 1981, 1 year after planting: Puccinellia on left, Halimione on right. 405 ik FIGURE 44. Preliminary planting of.Halimione sprigs along a creek bank at Ile Grande in May 1980. v W1 v t FIGURE 45. Same site as shown in Figure 44 in May 1981, 1 year after planting. 406 X, T7. M. TP' J FIGURE 46. Experimental planting of Puccinellia along a creek bank at Ile Grande in May 1980 just after planting. Az 7S if FIGURE 47. Same site as shown in Figure 46 in May 1981, 1 year after planting. 407 Transplant Time Requirement It is difficult to determine the time involved in the transplanting operations when experimental plantings are being established. In May 1981 we kept records of the time required for four persons to dig and transplant sprigs of Halimione and plugs of Puccinellia. . Halimione plants were dug, separated into sprigs and put into plastic bags for transport to the planting site at the rate of about 180 per person hour. Puccinellia plants were dug, cut into plugs and put into a container for transport to the planting site at the rate of about 75 per person hour. These rates indicate that Halimione sprigs can be obtained about 2.4 times faster than Puccinellia plugs. The planting operation includes opening the transplant hole with a soil auger, inserting the appropriate amount of fertilizer, and inserting the transplant and firming the substrate around it. Both types of transplants can be planted at the rate of about 40 per person hour. These time requirements for digging and planting make no allowance for travel, supplies, and equipment, which must also be considered in the total cost of a planting operation. Based on our digging and planting time requirements only, the time required to plant 1 ha of Halimione on a 0.5 m spacing (40,000 transplants) would be about 1,220 person hours (220 person hours to dig sprigs + 1,000 person hours to plant). The time required to plant 1 ha of Puccinellia on a 0.5 m spacing would be about 1,530 person hours (530 person hours to dig + 1,000 person hours to plant). These cost estimates indicate that it would take four persons working 8 hour days about 38 days to plant 1 ha of Halimione on a 0.5 m spacing and about 48 days to do the same using Puccinellia. Recovery of Transplant Source Sites From the beginning of our restoration efforts we were aware of the potential for impact to the natural marsh in digging transplants for the plantings. Consequently, we confined our digging of plants in the natural marsh to areas adjacent to narrow drainageways (Fig. 6) or to small areas (0-25 m2) in the marsh. All Puccinellia transplant source sites were replanted and those areas that were dug in 1979 and 1980 were almost completely revegetated by May 1981'(Figs. 48, 49). In a further attempt to lessen the pressure for obtaining transplants from the natural marsh, we have initiated nursery areas for Halimione and Puccinellia. These combined actions will help keep impact to the natural marsh to a minimum and serve as a model for others who may engage in similar activities in the future. Nursery Plantings The Puccinellia nursery area at Kerlavos was established in May 1979 and now contains about 300 plants that can be dug and separated into transplants. Although the plants vary in size, the average cover 408 M), '14 'ern w "'N *N FIGURE 48. Site where Puccinellia transplants were dug in May 1980. The area was replanted and was becoming rapidly revegetated in September 1980, 4 months after digging. f-, W-j FIGURE 49. Same site as shown in Figure 48 in May 1981, 1 year after digging. Vegetation cover is almost complete. 409 is about 540 cm2 or over 20 times that of a plug type transplant. To determine the actual number of plugs that could be obtained from a sample of plants, we dug 11 plants from the row nearest the estuary in May 1981 (Fig. 50). One of the largest plants yielded 50 plugs (Figs. 51, 52, 53), but the average number of plugs per plant dug was 20, which agreed well with what we predicted based on the average cover. Since the nursery area contains about 300 plants, we can predict that it could have yielded a minimum of 6,000 plug type transplants in May 1981. It seems reasonable to assume that cover will increase from 50 to 100% by our next major planting effort in May 1982. This assumption translates into a conservative estimate of about 10,000 plug type transplants or enough to plant 0.25 ha on a 0.5 m spacing. The Halimione nursery area at Ile Grande was established in May 1980 and added to in May 1981. It contains about 200 plants that can be dug and separated into transplants (Fig. 54). In May 1981 we dug a sample of seven plants to determine the average number of sprig type Halimione transplants that could be obtained per plant dug. We obtained an average of five sprigs from each plant dug (Fig. 55). Based on 200 plants in the nursery area, we estimate that there were about 1,000 Halimione transplants available in May 1981. We estimate that the increase in cover by May 1982 will result in about 1,500 to 2,000 Halimione sprigs available for digging at that time which would plant about 0.05 ha on a 0.5 m spacing. C =77 1b, Vw-- + gar" A S, FIGURE 50. Row of 2-year old Puccinellia transplants nearest the estuary in the nursery area at Kerlavos, May 1981. 410 V-1, 4.t ,7 A.4, -7i't "fig @4 FIGURE 51. Sample Puccinellia plant that was dug for transplants from the same row of plants shown in Figure 50, May 1981. AW2 OOV), Mrm FIGURE 52. Cutting the same plant shown in Figure 51 into plug type transplants. 411 Al /f FIGURE 53. Plant shown in Figure 51 yielded 50 plug type transplants. #7 f.7, � v J@ 0@,, -'k-I FIGURE 54. A 1-year old Halimione transplant in the nursery area at Ile Grande. 412 IiAIL 71.1 FIGURE 55. Halimione sprigs being dug from the nursery area at Ile Grande. Invasion of Plantings by Other Plants Observations at Ile Grande and Kerlavos indicate that other marsh plants invade our experimental plantings more rapidly than they colonize areas that still lack vegetation cover as a result of cleanup operations. In. one of our May 1979 experimental plantings of Puccinellia at Kerlavos (Figs. 56, 57, 58), 97% of the transplants in the 60 m' area had been invaded by at least one other species by May 1981 (Fig. 59). Of these transplants which had been invaded, 66% were invaded by two or more other species. The most abundant invader was an annual species of Salicornia which was present in 94% of the transplants sampled. Other invading genera in the order of their percentage of presence per transplant sampled were Cochleria (49%), Halimione (24%), Spergularia (10%), and Armeria (1%). 413 IN K ;-C @, wm@ S* 4 7@ T, FIGURE 56. Experimental planting of Puccinellia at Kerlavos in may 1979 just after planting. '7@ AVOW- Ad J@, z AlOb- 111L 4W Amok., Jftiw 7 -Ok '�r ar -2% 1' FIGURE 57. Same experimental planting as shown in Figure 56 in May 1980, 1 year after planting. 414 -4t4 !-@ 7--Z" tit FIGURE 58. Same experimental planting as shown in Figure 56 in May 1981, 2 years after planting. 4kr A FIGURE 59. A 2-year old Puccinellia transplant from the experimental planting shown in Figure 56 with two invading marsh plants: Cochleria (white flowers) and Salicornia (left center near cluster of.white flowers). 415 ACKNOWLEDGEMENTS Cooperation with our French colleagues (Madames Le Campion-Alsumard, Plante-Cuny, and Vacelet from Marseille and Monsieur Levasseur and Mademoiselle Jory from Rennes) has been invaluable. In fact, our work would have just about been impossible without their help. They have gone out of their way to help us while we were in France, such as arranging a meeting with the Mayor of Pleumeur-Bodou, making observations on our experimental plots during the interim of our visits, and providing laboratory facilities for processing samples. Presently we are cooperating with Monsieur Levasseur on nursery production of transplants and monitoring of our plantings until November 1982. He has also agreed to serve as the major professor for a graduate student from the United States who is a candidate for a Fulbright Scholarship. Among his tasks, this student would follow our plantings and document the invasion of these plantings by other marsh plants subsequent to our active involvement in the project. In summary, the association with our French colleagues has been very beneficial from our standpoint and we cannot overemphasize the vital part they have played in making our research proceed smoothly. We thank Amoco Oil for providing the funds for our research through NOAA Contract No. NA79RAC00018. We thank NOAA personnel especially Drs. W.N. Hess and D.A. Wolfe for providing us the opportunity to conduct this very timely and environmentally beneficial research and for their cooperation throughout the period of the project. Last, special thanks for technical assistance in field work and in data analysis go to Messrs. C.L. Campbell and L.L. Hobbs who made the overall research effort a success. SUMMARY Experimental plantings of Halimione portulacoides, Juncus maritimus, Puccinellia maritima, �2artina maritima, and Triglochin maritima have been made at Ile Grande and Kerlavos, France in an attempt to rehabilitate salt marsh that was impacted by the Amoco Cadiz oil spill and subsequent cleanup operations. over 61 experimental plantings including over 11,000 transplants have been established to test two types of transplants, conventional and slow release fertilizer materials over a wide range of substrate and elevation conditions and to develop nursery areas. Spartina transplants survived at lower elevations than those of any other species tested, but the best growth of transplants of all species tested occurred within + 0.3 m of the elevation of the natural marsh in the vicinity. Survival and growth data indicate that transplants of Puccinellia with a core of root and substrate material intact (plugs) were superior to those transplants with roots only (sprigs). 416 Although there was considerable variation in response to fertilizer materials and rates, both nitrogen and phosphorus were required for good transplant growth on the disturbed sites tested. Slow release fertilizer materials produced better growth over a wide range of substrate types than did the conventional, more soluble fertilizer materials. Higher survival and better growth were obtained with Halimione and Puccinellia transplants than with those of the other three species tested. Aboveground growth of the best experimental plantings of Puccinellia spread radially at the rate of about 10 cm. annually. At this rate of spread, these experimental plantings would achieve complete substrate cover in about 3 years after planting. Refertilization at various periods after planting produced a significant increase in cover. Halimione sprigs were dug at the rate of about 180 per person hour and plugs of Puccinellia at the rate of about 75 per person hour. Transplants of both species were planted and fertilized at the rate of about 40 per person hour. Sites in the natural marsh from which Puccinellia transplants were dug, were replanted and became almost completely revegetated within 1 year. Nursery areas were established for both Halimione and Puccinellia and estimates indicated that in May 1981 they contained about 6,000 transplants of Puccinellia and 1,000 of Halimione. Preliminary data indicate that other marsh plants invade our plantings more rapidly than they invade unplanted disturbed sites. 417 LITERATURE CITED Baker, J. M., 1971a, The effects of a single oil spillage: in E.B. Cowell (ed.), The Ecological Effects of oil Pollution on Littoral Communities, pp. 16-20, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971b, Successive spillages: in E.B. Cowell (ed.), The Ecological Effects of Oil Pollution on Littoral Communities, pp. 21-32, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971c, Refinery effluent: in E.B. Cowell (ed.), The Ecological Effects of oil Pollution on Littoral Communities, pp. 33-43, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971d, Seasonal effects: in E.B. Cowell (ed.), The Ecological Effects of Oil Pollution on Littoral Communities, pp. 44-51, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971e, Effects of cleaning: in E.B. Cowell (ed.), The Ecological Effects of Oil Pollution on Littoral Communities, pp. 52-57, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971f, Oil and salt marsh soil: in E.B. Cowell (ed.), The Ecological Effects of Oil Pollution on Littoral Communities, pp. 62-71, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971g, Growth stimulation following oil pollution: in E.B. Cowell (ed.), The Ecological Effects of Oil Pollution on Littoral Communities, pp. 72-77, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971h, Comparative toxicity of oils, oil fractions and emulsifiers: in E.B. Cowell (ed.), The Ecological Effects of oil Pollution on Littoral Communities, pp. 78-87, Applied Science Publ., Bristol, England, 250 pp. Baker, J. M., 1971i, The effects of oils on plant physiology: in E.B. Cowell (ed.), The Ecological Effects of Oil Pollution on Littoral Communities, pp. 88-98, Applied Science Publ., Bristol, England, 250 pp. Cowell, E. B., 1969, The effects of oil pollution on salt-marsh communities in Pembrokeshire and Cornwall: J. Appl. Ecol., Vol. 6, pp. 133-142. Cowell, E. B. and J. M. Baker, 1969, Recovery of a salt marsh in Pembrokeshire, Southwest Wales, from pollution by crude oil: Biol. Conserv., Vol. 1, pp. 291-295. Garbisch, E. W., P. B. Woller, and R. J. McCallum, 1975, Salt marsh establishment and development: U.S. Army, Coastal Engineering Research Center, Fort Belvoir, Virginia, Tech. Memo. 52, 110 pp. 418 Ranwell, D. S., 1968, Extent of damage to coastal habitats due to the Torrey Canyon incident: in J.D. Carth and D.R. Arthue (eds.), The Biological Effects of Oil Pollution on Littoral Communities, pp. 39-47, Field Studies Council, London, England, 198 pp. Seneca, E. D., S. W. Broome, W. W. Woodhouse, Jr., L. M. Cammen, and J. T. Lyon, 111, 1976, Establishing Spartina alterniflora marsh in North Carolina: Environ. Conserv., Fol. 2, pp. 185-189. Stebbings, R. E., 1970, Recovery of salt marsh in Brittany sixteen months after heavy pollution by oil: Environ. Poll., Vol. 1, pp. 163-167. Woodhouse, W. W., Jr., E. D. Seneca, and S. W. Broome, 1974, Propagation of Spartina alterniflora for substrate stabilization and salt marsh development: U.S. Army, Coastal Engineering Research Center, Fort Belvoir, Virginia, Tech. Memo. 46, 155 pp. 419 ETUDES MICROBIOLOGIQUES ET MICROPHYTIQUES DANS LES SEDIMENTS DES MARAIS MARITIMES DE MILE GRANDE A LA SUITE DE LA POLLUTION PAR L'AMOCO CADIZ par Therese Le Campion-Alsumard, Marie-Reine Plante-Cuny et Eveline Vacelet Station Marine d'Endoume et Centre d'Oceanographie 13007 - Marseille, France. PRESENTATION SOMMAIRE DES BIOTOPES ET STATIONS ETUDIES Trois des biotopes caracteristiques des marais maritimes - schorres, chenaux de schorre, et baute-slikke - ont ete retenus pour cette etude dans deux sites, differant treS nettement quant A Ilimportance de la pol- lution par le pEtrole de l'Amoco Cadiz (Fig. 1A - site Sud, Sud-Ouest tres pollu6 ; site Est, Nord-Est peu pollug). Le bloc diagramme (Fig. 1B) montre la difference de niveau altitu- dinal entre les 3 biotopes, entrainant evidemment des differences de durSe d'immersion. Les Schorres Les schorres de l'Ile Grande sont des prgs-salgs A Juncus maritimus et Halimone portulacoides qui presentent une surface plus ou moins tabu- laire et un reseau de drainage variable (Fig. 1A, scborre DI mieux drainE que le schorre Al). Les deux stations de reference pour le "biotope scborre" sont BI et C1 (site Est, Nord-Est, Fig. 1A) faiblement atteintes par la pollution du fait de la mise en place dq'un barrage, de protection sous le pont reliant I'Ile Grande A la terre. Les schorres tres pollues, Al et DI, sont donc situes dans la par- tie Sud, Sud-ouest. Cette partie du marais servit pendant un certain temps de zone de stockage d'hydrocarbures issus du nettoyage des plages et ro- chers. 421 Los Chenaux Ces schorres sont draines par des chenaux presque constamment immer- ges, dont le sediment est une vase fluide (station de reference C2 ; sta- tion tres polluge A2 - Fig. ]A et B). Le Biotope Haute-Slikke Le bintope de vase sableuse intertidale ou haute-slikke, de part et d'autre du chenal central, a et ftudie en A3 pour le site pollue et on B3 pour le site peu pollug. Cette derniere station fut remplacge A par- tir de mars 1980 par E3 (meme type de biotope) lorsque les travaux de sur- crousement du cbenal central e1imin;erent la station B. FIGURE I (OPPOSITE) A Sch&ma de localisation des stations Stations de reference = CIP B,, schorres (Est, Nord-Est du pont) C 21 chenal B 3P E31 haute-slikke Stations tres polluees = Alt D,, schorres (Sud, Sud-Ouest du pont) A 2' chenal A 39 baute-slikke B Bloc diagramme representant les 3 types de biotopes et le niveau des hautes-mers. 422 0 20 40 In _B1 B C 0 E3 A -ILI A A3 SOIL OF v S0,0RRE:A1, D1,C1,B1 IV IV IV 'v __j MUD OR SANDY @/. Ej MUD: ALC2, A3, B3, E3 SIVAING SITES REFERENCE STATIONS C1, B1 SCHORRES (SALT MEADOWS) C2 TIDAL CREEK B3, E3 SLIKKE (MUD SLOPE) POLLUTED STATIONS Al, DI SCHORRES (SALT MEADOWS) A2 TIDAL CREEK A3 SLIKKE (PILID SLOPE) E YL iv 600 chenal tidal creek HIAGVE STHWL HMMME @A' ID NTHWL Lhaute-slikke.- schorre higher mud slope sail meadow 423 ETUDES REALISEES DANS CES BIOTOPES ET METHODES UTILISEES Neuf missions d'echantillonnage pour I'etude des hydrocarbures et des peuplements bacteriens et microphytiques ont ete realisges entre de- cembre 1978 et novembre 1980 (une dernii_ere mission est programmee en no- vembre 1981, ce qui donnera un "suivi" de 4 ans). H Divers parametres physiques et chimiques (tempgratures, salinites, p , Eh) ont W mesurgs dans eau et les sediments. Les Hydrocarbures Les hydrocarbures (HC) ont ete doses et leur composition analyse dans des fractions de carottes a differents niveaux (Tab. 1, 2, 3). Les concentrations en hydrocarbures totaux (HCT) exprimees en g. kg-I de sediment see sont definies comme etant la fraction FA apras les traitements suivants appliques aux echantillons de sediments : - extraction de la matiere organique au toluene-mgthanol sur echan- tillon humide (Farrington et Tripp, 1975). - glimination du soufre (pesge avant saponification -4 poids AV- SP). - saponification A la potasse (pesge apres saponification --+ poids AP-SP). - fractionnement du produit de la saponification par la mthode dite au "Sep-pak"(micro-colonne de silice a compression radiale) (Giusti et al., 1979). - glution par trois solvants successifs 10 hexane fraction FA =.HCT 2* chloroforme fraction FB = fraction polaire 3' methanol fraction FC L'extrait a I'hexane (FA = HCT) est ensuite analyse en chromatogra- phie liquide haute pression (HPLC reverse-phase) puis en chromatographie gazeuse capillaire (CPG) et infra-rouge lorsque la concentration est suf- fisante. Une analyse plus fine par spectrometrie de masse couplde avec.la CPG est effectuge si necessaire. Etude Bacterienne Pour I'etude bacterienne, 1'echantillonnage des sediments etit effectuepar carottages. Les fractions eudiges eaient dans les schorres (carotte de 60 cm environ) - couche de surface (su) - rhizosphee (rh) - couches plus profondes (couche argileuse : ca ; couche sableuse: CS0q) 424 dans les chenaux (carottes de 30 A 40.cm) - couche de surface (su) - zone re'duite (zr) - couches plus profondes (ca ou cs) dans les slikkes (carottes de 10 a 40 cm) - memes couches que les chenaux. L'etude bacterienne de sous-echantillons comprenait - denombrement de la microflore heterotrophe par MPN sur eau de mer pep- tonge A 5 g.1-1 ; - estimation de l'activitg bacterienne sur les mames ensemencements etablissement d'une courbe d'activite qui depend de la presence de sou- ches a croissance rapide ; - denombrement des germes capables de degrader les hydrocarbures par MPN sur milieu mineral contenant du petrole "Arabian light" comme source de carbone ; - enzymologie : mise en evidence des differentes hydrolases presentes et estimation comparative de leur activite par la methode APIZYM. A la surface des carottes, activite enzymatique est due A 1'ensem-- ble du peuplement "bacteries + microphytes". Dans les couches plus profon- des, seule intervient Vactivite bacterienne. Les peuplements microphytiques Les peuplements microphytiques etaient etudigs sur des carottes plus courtes (3 premiers centimatres d'epaisseur). Aspect Quantitatif L'aspect quantitatif du peuplement est essentiellement apprghendg par estimation d'un indice chlorophyllien de biomasse que nous abrage- rons en ICB : extraction a acetone et mesures (avant et apres acidifi- cation des extraits) des concentrations en chlorophylle a (Chla ou ICB) et en produits de degradation de la chlorophylle (phgopigments = Pheo.), methode de Lorenzen (1967) modif iee par Plante-Cuny 0974) pour les sediments. Resultats exprimes enpg.g-I de sediment sec. Rapport Chla/Chla + Pheo : indice de vitalitg des peuplements s'il y a preponderance de la chlorophylle (rapport > 0,5). Aspect Qualitatif Lq'aspect qualitatif du peuplement consiste en une etude ecotaxino- mique des principales especes de diatomees et cyanophycees presentes. Note Il est evident que dans la presente synthease tous les resultats obtenus dans les etudes microbiologiques et microphytiques n'ont pu eZtre pris en compte. Ils feront l'objet de rapports separes. (Voir aussi Vacelet et al., et Plante-Cuny et al., 1981). 425 SYNTHESE DES PRINCIPAUX RESULTATS Les resultats concernant analyse fine des hydrocarbures et de leur eventuelle degradation font objet d'un rapport separg E. Des resultats succints seront donnes ici seulement pour servir A interpretation des phenomenes biologiques. Differents Degrgs de Pollution au Debut des Observations 11 faut noter ici Vabsence de "point zero" concernant 1'etat des marais de I'Ile Grande avant l'echouage de l'Amoco Cadiz en mars 1978. Aucune etude prealable n'existait sur ce site et il nous a ete impossible de trouver dans une region avoisinante un marais maritime de meme type, indemne, pouvant servir de reference. De sorte que, les stations choisies presentent en fait differents degrgs de pollution en decembre 1978, au debut de nos observations (HCT en surface, exprimes en g. kg I de sediment sec). Les valeurs ci-dessous, et notamment celles de Al et DI,sont tras glevees en comparaison des valeurs donnees par Marchand (1981) pour des sediments profonds (15 A 100 metres) a la suite du naufrage du "Bohlen". Rappelons que le site de Vile Grande tait un lieu de stockage des hy- drocarbures apres nettoyage d1autres sites. - schorres encore couverts de mazout en decembre 1978 Al : 32,97 ; DI : 94,68 - schorre avec traces de pollution CI : 4,17 - schorre apparemment indemne BI : 1,9 - chenaux drainant les schorres tres pollue, A2 : 7,69 ' apparemment indemne, C2 : 3,26 - haute-slikke bordant le chenal central visiblement tres polluge, A3 : 5,56 apparemment indemne, B3 : 0,50 Les concentrations evaluqes en A2 et A3 ont td mesurges sur des echantillons provenant de extreme pellicule superficielle du sedi- ment (moins de I cm d'epaisseur) dejA colonisee par des microphytes. Evolution des Concentrations en Hydrocarbures et des Peuplements Bacteriens et Microphytiques A Partir de Decembre 1978 Dans chacun des 3 biotopes - schorres, chenaux et slikkes - seront faites des comparaisons entre les stations tres pollueges A differents . Etudes realisedes par Henri Dou, Ggrard Giusti et Gilbert Mille. Laboratoire de Chimie Organique A et LA 126 CNRS - Faculte des Sciences de Saint Jereme - 13397 Marseille cedex 13. 426 degrgs dune part, et entre les stations tris polluees et les stations peu polluees d'autre part. On expose dans chaque cas I'evolution des concentrations en HCT (g.kg-1), et de leur eventuelle degradation evolution de I'activitg bacterienne, du nombre de germes degradant les HC, de l'activitg enzyma- tique ; - 1'evolut--on quantitative et qualitative des peuplements micro- phytiques. Evolution dans les schorres Schorres tres pollu6s, DI et Al. Schorre DI. En surface, la station DI (Tab. 1) presente, presque deux ans apres 1'echouage, la mame concentration en HCT (94,51) qu'en decembre 1978. Il existe dans ce site des zones encore plus polluges (prelevement intention- nel dans une tache d'hydrocarbures en mai 1980 : 230,60). Il semble y avoir, en ces points, une accumulation en surface par drainage du schorre environnant, lui-mqame encore visiblement tres pollue en 1980. Le rapport AV/AP (Tab. 2). indicateur presume d'une biodegradation, augmente legrement en 1979 ainsi que l'activitd bacterienne : un debut de degradation aurait eu lieu entre 1978 et 1979. On note parallalement une augmentation du nombre de germes degradant les HC jusqu'en novembre 1979 (104 a 107 germes.ml-I de sediment) suivi d'une regression en 1980 (Fig. 2). L'activite enzymatique de ensemble de la microflore a etg impor- tante en decembre 1978, s,'est prolongee jusquen 1979 ce qui suggere la possibilite d'un effet favorisant des HC. Cette activitg regresse en 1980 indiquant peut-etre un appauvrissement de la microflore bacterienne (rg- gression de chymotrypsine, trypsine et hydrolases des glucides) (Fig. 5). Les microphytes (Fig. 8kl)totalement e1imines et encore absents en novembre 1978, ont recoloni'se peu A peu la surface du sol et ont pre- sente un maximum non ne-gligeable en juillet 1979 (50 pg Chla.g-1). Ensui- te, la preponderance au printemps 80 des pheopigments indique un etat peu florissant de la population. Une reprise est amorcge en novembre 1980 (40,ug Chla.g-1). II est possible que cette population vegetale comportant un fort pourcentage de cyanophycees reputees riches en hydrocarbures natu- rels en C17 (Han et Calvin, 1969 ; Saliot, 1981) contribue A l'augmenta- tion observee du rapport C17/Pr (Tab. 3) ce qui.rend difficiles les inter- pretations quant a 1'etat de degradation des HC. DI : Dans la rhizosphare (cf. schemas des carottes, Tab. 1), on note dans le temps une diminution (de 8,13 a 0,23) de la quantitg d'HCT et une augmentation du rapport AV/AP (1,20 a 2,06) pouvant indiquer une forte biodegradation. En effet, la concentration en germes degradant les HC croit de 102 a 106 germes.ml-I jusqu'en 1980 (Fig. 2). Quant A acti- vitg enzymatique, elle est fluctuante et plus faible en gegral en 1980 (Fig. 5). 427 DI : Dans les couches plus profondes (30 cm), les concentrations en HCT, faibles au depart (0,48) diminuent en 1979 (0,10). Le nombre des germes degradant les HC a cru jusqu'en avril 1980 puis a diminug (Fig.2). L'activitg enzymatique regresse depuis decembre 1978. Tous les groupes d'hydrolases sont concerngs (Fig. 5). Schorre Al. contrairement au schorre precedent, oa persistent de fortes concentrations en HCT, les valeurs passent de 32,97 A 18,84 de 1978 A 1979, mais la biodegradation paralt peu importante (AV/AP 1). En 1980, la concentration tombe a 14,98. Les resultats concernant la rhi- zosphere en 1979 tendraient A prouver qu'il y a eu percolation. L'analyse montre en 1980, absence d'alcanes lingaires et la persistance des HC saturgs ramifigs et aromatiques. On observe egalement que la densite des germes degradants les HC augmente jusqu'en novembre 1979 et regresse en 1980 (Fig. 2), soit faute de substrat (alcanes lineaires), soit par sui- te d'un phenomene climatique general (voir BI et CI). L'activitg enzymatique en Al en surface, est comparable A celle de DI (accroissement en 1979, regression en 1980) (Fig. 5). Les microphytes ont et 1imines sur le sol Al par I'arrivee du mazout et etaient encore absents en decembre 1978, comme en DI (Fig. 8A). Par contre, une tren 1egere colonisation seulement etait amorcge en 1979, suivie d'une diminution en 1980 (quelquesii Chla.g I seulement). Les pigments degrades sont dominants en toutes saisons, traduisant le peu de vitalite de la population (Chla/Chla + Pheo. : 0,2 en moyenne). Al : Dans la rhizosphere du schorre Al, il y a augmentation de 0,47 A 3,68 des concentrations en HCT indiquant probablement une perco- lation en 1979, accompagnee d'une degradation importante (AV/AP : 1,47 et 1,26, Tab. 2). Ensuite, la pollution diminue en 1980. La microflore degradant les HC etait en densitg maximale en 1979, puis a regresse for- tement en 1980, davantage qu'en surface (Fig. 2). L'activitg enzymatique est en augmentation depuis 1978, la pollu- tion etant proportionnellement moins forte que dans le schorre DI (Fig. 5). Conclusion sur les schorres tres pollues. Ces deux biotopes, tres pollues au depart, ont reagi differemment puisque le plus pollue (DI) n'est pas le moins recolonisg par les micro- phytes. Il faut sans doute y voir influence benefique d'une plus forte humectation : DI, mieux draing donc plus souvent inode, est plus rapi- dement recolonise du fait de apport de particules argileuses favorisant la fixation des microphytes. Dans les rhizospheres; la degradation paraeit se faire convenable- ment, meme en cas de percolation. C'est donc A la surface des schorres les plus leves (ou les plus souvent exondes) que se restaure le plus lentement le peuplement micro- biologique. 428 Schorres peu pollugs, CI et B1. Schorre C1. En surface, sur le schorre C1, moins pollue (4,17 en 1978) les con- centrations en HCT ont rapidement diminug (0,54 en 1980). Les germes de- gradant les HC sont restes en nombre stable (104 a 105 germes.ml-I de sediment (Fig. 2). L'activitg enzymatique a ete constamment levge. La degradation semble donc s'effectuer normalement (Fig. 5). Dans la rhizosphere, il y a forte diminution des HCT avec le temps et forte degradation (AV/AP = 1,25). Les germes degradant les HC se sont maintenus en nombre stable (10 a 103 germes.ml-1). L'activitg enzymatique a decru en 1980. Dans les couches les plus profondes, les concentrations en HCT sont tri-s faibles en 1980 (0,05) et I'activitg enzymatique constante. Schorre BI. En surface, le deuxime schorre de reference, peu pollug (1,90 en decembre 1978) n'a pratiquement pas montrg de variations jusqu'en 1979 (1,75). Le nombre de germes degradant les HC-et les indices de biodegra- dation (1,34) tendent A prouver qu'il y aurait eu forte degradation (Fig. 2). Les HCT doses en 1979 n'ayant pas les caracteres d'HC degrades (C17/Pr = 4,43), nous emettons les hypotheses que de faibles apports chro- niques d'HC se produisent en cette station, ou plutet que les microphytes, tres abondants sur ce site, seraient responsables d'apports d'HC biogenes (Predominance d'imparitg > 1). Contrairement aux faits observes sur les schorres tre6s pollues, les microphytes n'ont pas ete ici limines au depart et ne paraissent pas af- fectes par une faible pollution, tout au moins en 1978 et 1979. Une densi- te maximale du peuplement est observege en juillet 1979 avec 130,Pg Chla. 1 (Fig. 8A) et un rapport Chla/Chla + Ph9o. de 0,75 indice de bon f6nc- tionnement de la population. Tout au long de l'annge, et contrairement aux schorres pollues, cet indice est toujours superieur a 0,5. Le peuplement de cyanophycees et diatomees est toute annge bien diversifig et les especes caracteristi- ques des marais maritimes y sont presentes. En 1980, cependant, un ICB moyen de 50)lg Chla.g-I est plus faible que celui des annges precedentes, mais comparable A celui du schorre C1. On ne peut, dans ces deux stations, exclure influence nefaste eventuelle d'un facteur climatique en 1980. Dans la rhizosphere BI, l'activiteg bacterienne et l'activitg enzy- matique ont etg importantes et les concentrations en HCT, faibles au de- part, ont diminug encore. Conclusions sur le biotope "schorre". Dans les stations tres polluees au depart 429 I* Une depollution se produit peu A Peu, sauf si de nouveaux apports par drainage (?) maintiennent des taux de concentrations eleves. 2* La degradation des HC est toujours plus faible, en surface et dans la rhizosphere, que dans les schorres peu pollues. Cette degradation est, dans les sites peu humectes, pratiquement stoppee. 30 Les concentrations en germes degradant les HC sont toujours supg- rieures de I A 2 ordres de grandeur a celles des schorres peu pollugs. 40 Par contre, l'activite enzymatique est 2 a 3 fois plus faible que dans les schorres peu pollues. 50 Apres avoir totalement disparu, les microphytes recolonisent tres,lentement les schorres "asphaltes", un peu plus rapidement les schor- res pollues mais plus humectes. Par contraste, les schorres peu pollugs sont tres florissants (ICB 10 A 20 fois supgrieur). Evolution dans les Chenaux Chenal tres pollue, A2. En surface, les concentrations en HCT ont diminug de 10,78 A 2,57. Les indices adequats prouvent u'une forte degradation a eu lieu. Les chromatogrammes des HC restant en 1980 ne presentent plus les pics des alcanes lineaires. La degradation rapide parait terminge et les, autres fractions resteront probablement en 1'etat. Les germes degradant les HC ont augmentd en nombre jusqu'en 1979 (106) puis ont decru en 1980 (104 germes.ml-1, Fig. 3), ce qui corrobore 1'hypothese precdente de la non poursuite de la degradation mais n'exclut pas I'hypothase du facteur cli- matique. L'activitg enzymatique est plus faible que dans la station non polluge (Fig. 6). Les microphytes ont recolonise tries rapidement la surface de la vase et ont atteint en'novembre 1980, un maximum de 588, Chla.g-1, maximum absolu dans toutes les stations etudiges (Fig. 8B). En 1978, cette microflore etait paucispecifique (Phormidium et Amphiprora aZata). En 1980, la diversite specifique d'une vase normale dequivalente (Carter, 1933) est retrouvee. Pourtant un raclage effectue en mai 1980 montre encore une concentration d'HC de 14,20 g.kg-1. Les chromatogrammes montrent I'absence totale d'alcanes-lingaires (Tab. 3). Dans la zone reduite (Tab. I : zr), les concentrations en HCT ayant augmentg jusqu'en 1980, on peut evoquer la possibilite d'une perco- lation. A ce mgme niveau, en 1980, tous les alcanes linegaires sont degra- des. La concentration mesuree concerne donc des HC plus resistants. On observe en outre que le nombre de gemes degradant les HC decroeqit forte- ment (10 germes.ml-1). L'activitg enzymatique decroit galement. Chenal peu pollue, C2. En surface, les concentrations en HCT sont passees de 3,26 A 0,70. La biodegradation est importante (AV/AP glevg). Le nombre de germes degra- dant les HC est stable (103 a 104 germes.ml-1). 430 L'activit6 enzymatique dans cette station "propre" est aussi impor- tante que dans les schorres peu pollu6s. Elle est en nette augmentation en 1980, ce qui est peut-@tre a relier aux travaux d'am6nagement effectugs a cet endroit (voir ci-dessous). Les microphytes ont manifestg un maximum de d6veloppement en 6t6 1979 suivi d'une d6croissance. Les variations saisonni@res paraissent donc tr;_@s diff6rentes dans les deux stations de type "chenaux", mais le chenal oa se trouve la station C2 a 6t6 obtur6 momentangment, en mai 80, par des travaux de surcreusement du chenal central. La composition des peuplements de microphytes a 6t6 nettement perturb6e et s'est appauvrie en esp@ces et en individus. La zone r6duite et la couche sableuse sous-jacente sont fortement d6polluges (0,09 et 0,05). Le nombre de germes d6gradant les HC est sta- ble. Dans la zone r6duite, une diminution de Vactivit6 enzymatique se poursuit depuis d6cembre 1978, indice possible de la restauration doun 6tat d'origine (disparition des lipases, augmentation des activit6s estd- rases, aminopeptidases, chymotripsines et glucidases). Conclusions sur le biotope "chenal de schorre". 10 Dans le cas de pollution forte, la biodggradationapr6s avoir 6t6 active, s'est ralentie ou arrZtEe. Les HC encore pr6sents ont peu de chances d'@tre d6grad6s rapidement. Dans le cas de pollution faible, la d6gradation a 6t6 presque compline. 20 ParaM-lement, le nombre de germes d6gradant les HC a augment6 puis diminug dans le cas de forte pollution. Il est stable ailleurs. 3' L'activitg enzymatique est moins importante, en surface, en cas de pollution. Elle est toujours plus faible en zone r6duite. 40 Les microphytes se d6veloppent en surface de fagon luxuriante dans les chenaux pollu6s. Les concentrations en Chla sont en 1980 nette- ment sup6rieures A celles du chenal non pollug. Evolution dans les slikkes Slikke tri@s pollu6e, A3. En surface, a Vendroit pr6cis o@i nous avons situg la station A3, on peut dire que les concentrations en HCT sont pass6es dans la pelli- cule superficielle (ps = quelques millim@@tres d'6paisseur) de 5,56 A 0,27 g.kg-I entre d6cembre 1978 et mai 1980 (Tab. 1). Dans la couche sous-jacente (su : 2 A 3 cm d'6paisseur) les conceir trations sont pass6es de 24,95 a 0,60 entre 1979 et 1980. Il y a donc eu d6pollution en surface. Dans le mame temps, les indices de biod6gradation ont augment4 jusqu'en 1979 (Tab. 2) et le nombre de germes d6gradant les HC 6galement (Fig. 4). Dans la couche sableuse, par contre, les concentrations en HC ont augment6 de 0,65 A 2,40 en 1979, puis diminu6 en 1980 (0,50). Les indi- ces de biod6gradation sont faibles. L'activit6 enzymatique est r6duite (Fig. 7). Il semble y avoir eu percolation, surtout en novembre 1979. 431 Devant les difficultes d'interpretation des resultats dans un tel biotope (problemes d'echantillonnages), des prelevements par raclages ont ete faits en mai 1980. Ils ont permis d'effectuer des dosages A partir d' un materiel abondant et de differencier, d'une part la pellicule superfi- cielle constituge essentiellement de particules argileuses compactees par les microphytes, et, d'autre part, la couche sableuse visiblement encore tri-s polluge. Les resultats sont eloquents : 0,27 dans le premier cas, 15 g.kS-1 dans le second. Ce sont des hydrocarbures d'origine "Arabian light" dont les alcanes lineaires certes sont degrades, mais dont les autres consti- tuants sont toujours presents.(Tab-4,S)- Notre premiare hypothase du "piggeage" du petrole sous la matte vegetale se trouve confirmee. En effet, les filaments de cyanophycees et les diatomees compactant les particules argileuses avaient treas rapidement colonisg ces vases im- prenees d'HC puisqu'en decembre 1978, on observait un ICB de 140 Ig Chela. 9 1 (Fig. 8C), et tres peu de pheopigments (rapport Chla/Chla + Pheo. de 0,96 , le plus elevg de cette etude) indiquant une population jeune. Ce peuplement se revelait paucispecifique (Phormidium, deux espe-ces de Nitz- schia, Anphipleura_, Rhopalodia) mais se diversifiait tres rapidement. Les deux cycles annuels de 1979 et 1980 prq6sentent tous deux un maximum en automne ou en hiver, tout comme la vase du chenal pollug, avec des ICB presque aussi elevs et toujours tres peu de pheopigments. En 1980, le peuplement est treas riche et tri-es diversifie (presence de nombreuses espaces caracteristiques de ces milieux). La matte algale recouvre donc un sediment dans lequel une depollu- tion a eu lieu, tout au moins dans la partie superficielle, mais dont la couche sableuse est toujours impregnee d'HC dont la degradation parait treas ralentie. Cette matte semble toujours jouer un rele de frein A une depollution mecanique par le jeu des mardes. Slikkes peu--polluges, B3 et E3. La station B3 nous a paru devoir etre remplacee, en mars 1980, par une nouvelle station de rference (M), la slikke centrale ayant etg per- turbee par l'obturation momentange du pont, apres l'echouage du Tanio (mars 1980) puis par des travaux d'amenagement. Les concentrations en HCT dans ces stations sont faibles. La bio- degradation a ete importante. Le nombre de germes degradant les HC est stable (Fig. 4). L'activite enzymatique de surface est en augmentation en 1980 (Fig. 7). Le peuplement microphytique est florissant, particulierement en E3, ou un indice C17,/Pr de 4,47 en mai 1980 pourrait traduire, comme dans eles schorres non pollues B1 et CI, la presence d'un hydrocarbure biogeane par- ticulieirement abondant chez certaines cyanophycees (Fig. 8C). Contrairement a la slikke polluee, les couches sous-jacentes ici ne renferment pas d'hydrocarbures. 432 Conclusions sur le biotope "slikke". Ces conclusions so nt tres voisines de celles qui concernent les chenaux I* En cas de pollution grave, la biodegradation, active d'abord, lest ralentie ou meme arretee. 20 Le nombre de germes degradant les HC a diminug depuis 1978. Dans les stations peu polluges il est stable. 3* L'activite enzymatique paralt toujours freinee en surface en cas de pollution forte. 4* Les microphytes se sont developpes rapidement sur les sediments pollugs, piegeant des particules argileuses et constituant une 11matte al- gale" plus compacte que dans les chenaux, pellicule qui freine la depollu-- tion de ces sediments. 433 CONCLUSIONS Nous avons essaye au cours de cette etude de nous attacher e compren- dre les interrelations qui pouvaient exister entre 1'etat de degradation des hydrocarbures et 1'evolution des peuplements bacteriens et microphy- tiques des marais maritimes. La complexite de ces milieux et les problames d'echantillonnage qui en decoulent rendent parfois difficile la comprehen- sion du fonctionnement d'un tel gcosyst;me perturbe. En ce qui concerne ensemble des bio'topes. 10 Il apparalt tout d'abord que les marais maritimes de l'Ile Grande restent tres pollugs malgre une biodegradation,importante. 20 Les hydrocarbures presents actuellement A la surface des sediments ou dans des couches plus profondes (percolation ou "pifteage") sont A un stade tel (disparition totale des alcanes lineaires) que la degradation ne paralt pas se poursuivre actuellement. Ce ralentissement peut avoir plusieurs causes : persistance des seules fractions les plus resistantes des HC, toxicitg,pour le peuplement bactgriende certains produits de de- gradation, faceteur climatique defavorable a l'activitg bacterienne. 30 L'evolution de ces milieux s'est avqgr6e differente suivant le degre initial de pollution : - dans les stations tres polluges, les concentrations en germes de- gradant les hydrocarbures ont ete triz is glevges. Puis leur nombre a decru en 1980. L'activitg enzymatique a etq6 moins elevge en surface et dans les rhizosphi_res. dans les stations peu polluges, l'impact sur les peuplements micro- phytiques et bacteriens a ete peu perceptible, et la degradation a etqg plus poussee, mettant en evidence eVexistence probable d'un seuil de con- centration en HC en-dessous duquel la "restauration" est possible. En ce qui concerne chaque biotope en particulier, le deversement massif du petrole "Arabian light" n'a pas eu les memes consequences dans les sols du pre-sale, le plus souvent exondg (schorres) que dans les se- diments fins, le plus souvent immerges (chenaux et slikkes). L'aspect mi- crobiologique et aspect mecanique sont A prendre en compte dans les differences de depollution. Les vases intertidales Elles ont etg probablement plus vite nettoyees par le jeu des marges que la surface des schorres. Mais, par ailleurs, elles ont et tres rapi- dement recolonisges du fait de apport de particules argileuses favori- sant installation d'un peuplement paucispecifique de eyanophycees et de diatomees qui, par la suite, s'est diversifig. Le mazout restant slest trouve ainsi plus ou moins piegg sous cette "croate" algale et pourrait sly maintenir longtemps. 434 Les schorres Les schorres, par contre, ont gt6 moins rap_idement d6pollugs par les marges, certains pr6sentant me^me des zones d'accumulation. Apras avoir totalement disparu, les microphytes recolonisent tr@s lentement les sols, d'autant plus lentement qu'ils sont moins souvent immerggs. La colo- nisation est actuellement environ 10 fois inf6rieure A la normale sur le sol.des schorres A Juncus, observation qui concorde, avec les r9sultats obtenus par Levasseur et Seneca sur la flore macrophytique (voir contri- butions de ces auteurs). 435 TABLEAu 4 EVOUrTION DES CONCENTRATIONS EN HYDROCARBURES TOTAUX DANS LES SEDLMENTS DES KAPAIS KAkITIMES DE VILE CRANDE. Biot.opes Diff6rents HC tocaux g.kg-I do r6diment sec Ct Diveaux Stations daus la IGI IG2 IC6 Ics A carotte 12/78 3/79 11/79 5/80 su su Schorre su 32.97 18,84 14.98 AI Th 0,47 3,68 0,03 C& 0.04 hu 94,6B 94.51 230.60* rh 8.13 0.23 17.789k I I ca 0.48 0.10 0,16* su 1.90 1.75 Th 0.17 0.10 lu su 4,17 0.43 0,54 C rh 0.26 0.03 0.15 cs(14-32 cm) 0,05 Cl@enaux ps 7.69 14.20 A2 su 1q,78 (.,59 2,57 f ar 0.48 0.22 0.29 1.14 su ca 0.10 0.03 0.0a - 28-36 cm 0.08 su 3.26 1.41 0.70 C tr 0.77 0.09 2 Cs 0,23 0.05 A3 9 3 lfa@tk ps 5,56 0.27-15" A 24,95 3.45 0,60 Cs 0.65 2,40 0.50 su 0.89 zr 3 ca 0,09 su 0.52 0,56 zr 0,19 0.22 3 ca 0.16 ps pollicule superficielle. pr6l@v,i-.,.ont par raciave, Ivu p3rtio qu'lque's 'k Uttl' I@L@@LtX 11'a j-1-f- et@ I-r4.,IcvCc au lm-iid unis dans vne tache rh rl@i?ospli'rc aiin d',@i @imdivr I'kat d, d@grad;Uilm. c s couche sableuse sk F),t, nt eti; @ff,,I-ts &w. le central. zr xone )-@duite CA 436 TABLFAU 2 HYDROCARHURES DANS LUS SEjljI,U-',,n'S DES MARAIS MARUMS DE MILE CKANDE (POLIAMON PAR L'A4(lCO rADTZ) ANALYSE CHIMIQUE stations Poids humide des AV-SP AP-Sp AV/AP FB Hydroca rbu re a kchantillons I totaux (FA) g. g-kg-I gA871 g. ka- g. kg-i Icl IG6 ic 1 IC6 IGI IC6 lGi 1136 IG I XG6 Ic I IG6 12/73 11179 12/78 11/79 12/78 11/79 12/78 11/79 1 12/78 11179 12/78 11/79 Schorre Al r. u 13,4o 25.10 73,60 43,48 67,60 39,58 1.09 1,10 35,70 16.93 32,97 18,84 rlk 68,70 99,50 2,00 8.92 1.37 7.10 1.47 1.26 0.90 2,99 0,47 3.68 su 12,55 14,60 173,90 243,45 162,50 210,50 1.07 1,15 67,90 102.12 94,68 94.51 DI A 59,15 90.10 19,78 1,98 16,50 0,96 1,20 2,06 8,40 0,46 8,13 0,23 ca 24.25 74,80 1,89 0.24 1.48 0,19 1.26 1,26 1.00 0,06 0.48 0.10 81 su 13,3o lo,85 8.00 10.29 5,1S 8,98 1,34 1.14 4,00 5.56 1,90 1,75 rh 53,50 97,80 2.12 1.99 1,30 0,89 1.63 2,24 1,06 0,58 0,17 0.10 C, su 15,35 19,85 13,74 2.43 12,27 1,93 1.12 1,26 8,10 1,19 4,17 0,43 A 58,80 132,50 2.41 0,15 1,59 0,12 1,52 1,25 1,33 0,05 0.26 0,03 (lienaux de schorre ps 51.20 6jo 18,52 21,49 16,70 12,41 1.11 1,73 9.00 4,25 7,69 6,59 A2 zr 83,50 120,60 1.78 0.95 1.0 0,60 1,25 1,58 0.91 0,23 0,48 0,29 ca 23.10 139,20 0,63 0,17 0,31 0,07 2,03 2,43 0,21 0,02 0,10 0,03 su 10,50 15,90 14,07 6.32 8,21 4,28 1.70 1,48 5.00 2,17 3,26 1,41 C2 zr 39,80 41,20 3,o7 1,89 2,26 0,94 1,36 2,01 1,50 0.66 0,77 0.09 cs 18,65 76,90 0,60 0,92 0.45 0,40 1,33 2,3 0.22 0,25 0,23 0,05 -Haute-slikkp A3 ps 37,80- 1.05 11,48 16,91 11,44 12,91 1,00 1.31 5,88 a 5,56* 3,45 CS 106,50 7,43 6,66 1,12 3,54 2,40 B3 ps 7,50 6,18 2,66 2,32 1,34 0,52 ca 112,30 3,43 1,76 1,95 1,14 Oi19 Arab i an 100 94 g 1.07 32X 68% li ght AV-SP avnnt Gnpunification ru p.-irtie uuIwr(i,i(!ljv All-SI, zpr@Ni t.-Ip"llificaLion 9-wl,pirm c(,nLinIP-l,rcs. jmidq en r. par @.p, d,, O@J;v,.,nt bec rh JI.V/Al' : rapport zr zmw T,@,Iqilv I R f I ;,c 1. 1 or, 9 f-p-PA III 1 .11 IIC CI (7n IA frnrtlon t., (-luliwi heymip cd 1,. 1,(.l I l(ole ktil-ri ic i,,) I(. (,,a r,,[ i,n, mit Aur Ifi-q I I),)- !,- rarli',. In hiji, )4. jur rAhi (o,5 ct;i gl'lp,iiuh4 ur ii,vi I,,,,) 437 TAIILFAU 3 HYDROCAJUSURES DANS LES SEDIMENTS DES MAKAIS MARITIMES DE LIME GRANDE (POLLUTION I'Ak L'AMOCO CADIZ) ANALYSE CIJRO.IIATO(;KAI-111QOF DE LA FKACTION SATUREE c 17 / Pr c Ph Pr Ph PrEdominance d'imparities Stations IG I IG 6 IG8 XGI IG 6 IG 8 IG I IG 6 IG 8 IG I IG 6 IG 8 12/78 11/79 5/80 12/78 11/79 5/80 12/78 11/79 5/80 12/78 11/79 5/80 Schorre :u 1) 0.5 xx 0.77 0,58 xx 0.54 0,67 xx @1 )OC xx h 48 0.5 12 0.41 0,21 4.5 0.61 0.35 0,25 .1 xx 1,30 su 0.60 0,2 1.88 0,47 0,11 1.21 0.60 0.54 0,67 @I xx 0.91 D rh 1,24 1,5 2,14 0.93 1.92 1,40 0.75 0,66 0.64 tIII 1,40 1 ca 3.30 2 3 0.78 1.92 2.39 0,17 0,83 0.61 xx xx 1.12 su xx 4,43 xx 2.62 xx 0.54 xx 1.57 rh 0,21 1.31 1,75 A 7.86 3,2 2,37 xx 1,42 6 6,20 4.64 3.80 0.55 1,11 1.27 Iul 2,80 6.82 6,67 7.23 1,60 0.85 1.05 1.84 Chcna-ix de schorre ps 0.35 XX 0,13 xx 0.56 xx xx xx A2 ir 0,50 0,26 0,58 1.25 ca 1,00 2,83 3,50 2.89 2.20 0.67 1.03 1,17 su 0.13 0.19 @xx 0.39 0,44 xx 3.50 23.33 xx 0.99 1.84 xx C2 zr 0.13 1.25 0.26 2,5 2,28 4 1.06 1,48 cs 2.33 3,33 5.30 3,5 2 0.75 0.89 1.30 Slikke A su 4,10 xx 6.00 xx 1,25 xx 1,03 xx cs 0109 xx 0,06 .0.85 xx xx xx B ps 0.53 xx 0.53 1,42 0,94 1,04 1,77 cs 0.29 0114 4.72 1,60 E su 14,47 1,18 0.5 1.19 ca xx 6,5 xx 2,36 Arabian I(J,3(j 4,70 0,48 0.88 I i gilt 2 (r. 23 + c25C 27 VyCdcq!,iyjancv d'irapjrit@ c22 4 2C 24 *2c 26 4 c28 xx Jell ... .... lltl*(-IIL 1,)Ij!. )a pis't-n-- d'a)-em-!; ffn@,,tirvx. )I u 1, ;1Tt iF..jl,I Oil 1, 111 1 ep-s I, III i m; L rf-R r I'll 438 .12-14 Juin 79 2 - 7 oct 79 23-25 now 79 .1-2 avrif 80 19-20 mal So now so 1 3 A 3 1 3 0 @k Al . . . . . . 1 3 5 T-r- 0 Di I/T T Bi Cl 7/E L FIGURE 2. Nombre de germes (log) d6gradant les hydrocarbures diff6rentes 0 profondeurs dans les schorres trZ-is pollu6s (Al et DI) et moins pollu6s (BI et CI) des marais maritimes de l'Ile Grande. I couche d'hydrocarbure argile couche a microphytes coucbe r6duite rhizosph@@re sable 439 12 -14 juill 79) 2-7 oct 79 23-25 nav 79 1-2 avril 80 19-20 mal 80 24-25 nov 00 5 1 3 3 7 a I I. -6 1 3 6 7 A2 v I T -T B2 11 FIGURE 3. Nombre de germes (log) d6gradant les hydrocarbures a diff6rentes profondeurs dans les chenaux tr6s pollu6s (A2) et moins pollu6s (B2) - mgme 16gende que figure 2 -. 12 -14 juill 79 2-7 oct 79 23-25 ncw 79 .1-2 avril 80 19-20 mal 80 24-25 nov 80 1 3 5 1 1 3 5 -- 1 3 s I 1 2 a 7 1 3 7 --r 7r-r-r-T-T- A3 63 FIGURE 4. Nombre de germes (log) d6gradant les hydrocarbures A diff6rentes profondeurs dans les slikkes tr;-@s polluges (A3) et moins pollw- 6es (B3) - meme 16gende que figure 2) 440 11 dec 70 13-13 m6flk TO 7.10 Mal 79 12 -14 jull 79 2-7 oct 79 23-25 A@ 70 1-2 &vril 00 19-20 Mal 80 24-25 no- 60 j, ot 7 I At It ell elo It %4- Ib X-1 It 4 81 0 0 91 /* el ell I ell It CA It FIGURE 5. Activit6 enzymatique (cf. m6thode) de Vensemble des peuplements bact6riens et imicrophytiques dans les schorres tras pollugs (Al et DI) et moins pollu6s (BI et Cl). 11 dec TO 13-15 rnoF% 79 7-10 Mal 79 12 -14 jull 70 2-7 oct 79 .23-23 nov 79 .1-2 avril 80 19-20 Mal 80 24-25 nov 80 A2 's 82 lp C.^ Is I A FIGURE 6. Activitg enzymatique dans les chenaux tras pollugs (A2) et moins pollugs (B2 et C2). I I dec TO Q-15 moff 79 7-10 mol 74 12 -14 julH 70 2-7 oct, 70 23-25 nov 70 .1-2 avril 00 19-20 Mal Go 24-25 nov 80 v Q, Til 3 Bn FIGURE 7. Activitd enzymatique dans les slikkes tras pollu6es (A3) et moins polluhs (B3). FIGURE 8 Evolution temporelle des concentrations en chlorophylle a (trait plein) et en pheopigments (pointilles) a la surface des sediments (en ug-g- de sediment sec). A - Stations de schorres B C, peu pollues D A tres pollues B - Stations de chenaux C2 peu pollue A2 tres pollue C - Stations de slikkes B3` E3' peu polluees A3 tres polluee 444 Chi. 150- - - - - - - - MO. B1 Cl dop 50 - cm t=rs C/D 1980 ZE CM 50 D1 I A ___r_ __u_l I I I I 1 1 1980 1979 1980 445 - - - - - - - Ph6o. 100- cm vo cm 1979 1980 41, 588 Pg-g-l A 2 100- 50 - 1980 446 150 B3 3 1979 C-0. CCXL-.. ISO A3 447 REFERENCES BIBLIOGRAPHIQUES Atlas, R.M. et A. Bronner, 1981, Microbial hydrocarbon degradation within the intertidal zones impacted by the Amoco Cadiz oil spillage : in Amoco Cadiz. Fates and effects of the oil spill. Proc. Internat. Symposium. C.O.B. Brest (France) novembre 19-22, 1979, pp. 251-256. Butler, J.N. and E.M. Levy, 1978, Long term fate of petroleum hydrocar- bons after spills. Compositional changes and microbial degradation J. Fish. Res. Board Can., Vol. 35(5), pp. 604-605. Carter,, N., 1933, A comparative study of the alga flora of two salt mar- shes : J. Ecol., Vol. 21, pp. 128-208, 385-403. Colwell, R.R., A.L. Mills, J.D. Walker,, P. Garcia-Tello, et P.V. Campos, 1978,, Microbial ecology studies of the Metula spill in the straits of Magellan : J. Fish. Res. Board Can., Vol. 35(5), pp. 573-580. Ducreux, J., et M. Marchand, '1981, Evolution des hydrocarbures pr6sents dans les s6diments de l'Aber Wrac'h, d'avril 1978 A juin 1979 : in Amoco Cadiz. Fates and effects of the oil spill. Proc. Internat. Symposium. C.O.B. Brest (France) novembre 19-22, 1979, pp. 175-216. Farrington, J.W., and B.W. Tripp, 1975, A comparison of analysis methods for hydrocarbons in surface sediments: in T.M. Church (ed.), Marine Chemistry in the Coastal Environments, @@er. Chem. Soc. Symp. Series no 18, Washington, D.C., pp. 267-284. Fujisawa, H., M. Murakami, et T. Manabe, 1977, Ecological studies on hy- drocarbonsoxidizing bacteria in Japanese coastal waters. I. Some methods of enumeration of hydrocarbon oxidizing bacteria : Bull. Jap. Soc. Sc. Fish., Vol. 43(6), pp. 659-668. Fujisawa, H., M. Masatada et M. Takehiko, 1978, Ecological studies on hy- drocarbons oxidizing bacteria in the oil polluted areas caused by the Mizushima oil refinery accident (Seta Inland Sea) : Bull. Jap. Soc. Sc. Fish., Vol. 44(2), pp. 91-104. Giusti, G., E.J. Vincent, H.J.M. Dou, et R. Faure, 1979, Etude par RMN de la concentration et de la nature des hydrocarbures pr6sents dans les s6diments c5tiers superficiels de I'lle des Embiez : La Vie Marine, 1, pp. 24-29. Han, J. and M. Calvin, 1969, Hydrocarbon distribution of algae and bacte- ria and microbiological activity in sediments : Proc. nat. Acad. Sci. U.S.A., Vol. 64(2), pp. 436-443. Lorenzen, C.J., 1967, Determination of chlorophyll and Pheo-pigments Spectrophotometric equations : Limnol. Oceanogr., Vol. 12(2), pp. 343-346. 448 Marchand, M. et J. Roucachg, 1981, Crit6res de pollution par hydrocarbu- res dans les s6diments marins. Etude appliqu6e A la pollution du "Bohlen" : Oceanol. Acta, Vol. 4(2), pp. 171-183. Plante-Cuny, M.R., 1974, Evaluation par spectrophotom6trie des teneurs en chlorophylle a fonctionnelle et en ph6opigments des substrats meubles marins : Doc. Sci. Mission ORSTOM, Nosy-B6, Vol. 45, pp. 1-76. Plante-Cuny, M.R., T. Le Campion-Alsumard, et E. Vacelet, 1980, Influence de la pollution due a l'Amoco Cadiz sur les peuplements bact6riens et microphytiques des marais maritimes de l'Ile Grande, 2, Peuple- ments microphytiques : in Amoco Cadiz. Fates and effects of the oil spill, Proc. Internat. Symposium, C.O.B., Brest (France) novembre 19-22, 1979, pp. 429-442. Saliot, A., 1981, Natural hydrocarbons in sea water : in E.K. Duursma and R. Dawson (ed.), Marine Organic Chemistry, Amster-d-am, pp. 327-374. Thompson, S. and G. E glinton, 1979, The'presence of pollutant hydrocarbons in estuarine epipelic diatom populations, II, Diatom slimes : Estua- rine and Coastal Marine Science, Vol. 8, pp. 75-86. Traxler, R.W., et J.H. Vandermeulen, 1981, Hydrocarbon -utilizing micro- bial potential in marsh, mudflat and sandy sediments from North Brittany : in Amoco Cadiz. Fates and effects of the oil spill. Proc. Internat. Symposium. C.O.B., Brest (France) november 19-22, 1979, pp. 243-249. Vacelet, E., T. Le Campion-Alsumard, et M.R. Plante-Cuny, 1981, Influence de la pollution due i I'Amoco Cadiz sur les peuplements bact6riens et microphytiques des marais maritimes de l'Ile Grande, 1, Peuple- ments bact6riens : in Amoco Cadiz. Fates and effects of the oil spill. Proc. Internat. Symposium. C.O.B., Brest (France) november 19-22, 1979, pp. 415-428. Ward, D.M., 1981, Note de synth6se. Microbial responses to Amoco Cadiz oil pollutants : in.,Amoco Cadiz. Fates and effects of the oil spill. Proc. Internat. S@Y_Mposium.'C.O.B., Brest (France) november 19-22, 1979, pp. 217-222. 449 1964-1982, COMPARAISON QUANTITATIVE DES POPULATIONS BENTHIQUES DE ST-EFFLAM ET DE ST-MICHEL-EN GREVE AVANT,PENDANT ET.DEPUIS LE NAUFRAGE DE L'AMOCO CADIZ par C. CHASSE et A. GUENOLE-BOUDER Laboratoire d'Ocdanographie Bioloqique, Institut d'Etudes Marines, Universite de Bretaone Occidentale, 6, avenue Le-Gorgeu, 29283 Brest cedex, France Resume La baie de Lannion, largement ouverte face a la progression des nappes de petrole de !'AMOCO CADIZ, fut particulierement souillee 60 millions de cadavres echoues furent denombres sur les deux plages du fond de la baie de St-Efflam et Locquemeau. Des etudes anterieures sur 1'estran de St-Efflam, representatif des nombreuses pl,agqes de sable fin de Bretaqone, ont servi de reference a ce travail. L'impact du petrole est tres variable sur les diverses especes d'une meme station. La partie Est de la plage, plus contaminee, montre une plus forte mortalite. Le haut et le bas de 1'estran sont plus affectes que la partie intermediaire. Deux processus semblent intervenir : - les nappes d'echouages en haut, - le petrole dissout ou en emulsion dans la masse d'eau en bas et dans tout l'infratidal. A la mortalite immediate s'ajoutent des mortalitds et des effets pathologiques 0qd long termeq. Certaines especes continuent a regresser meme en 1981 et les recrutements souvent inexistant en 1978 sq'amorcent en 1980, pour certains timidement encore en 1981. L'Est de la plage reste fortement touche bien que des signes de recouvrance certains apparaissent sur le reste de la plage mais les gros peuplements du bas de la plage a Soeten, Eneis, EchinocaAdium, Lut)Latia, MactAa coLtattina ne sont pas reapparus. 451 INTRODUCTION Les nappes d'emulsion petroliere de I'AMOCO CADIZ pouss6es i la surface de la Manche, le Jong de la cete, par les vents dOuest ont ete freinees en se heurtant sur les saillants successifs du rivage. Quatre zones principales d'obstacles se sont dressees face A leur progression vers I'Est. Ce sont les roches; et les ilots des abers et de la presqu"ile Ste-Marguerite, d'abord; le champ de roches de l"lle de Batz, de Santec, de Roscof ensuite; puis les roches de Primel et du Guersit; enfin, celles des rebords Est de la baie de Lannion avec l"He Grande. En chacun de ces lieux, et surtout dans les criques, les estuaires et les baies de sable fin les plus proches qui les precedent, le Petrole s'est abon- damment accumule en provoquant de lourdes mortalites donnant lieu a d'importants echouages de cadavres de poissons, d'oiseaux et de coquillages. La baie de Lannion, deja fortement souilke, en 1967, par le petrole du TORREY CANYON, a ete atteinte par les nappes de ]AMOCO CADIZ de le 5ejour apres le naufrage. Sur les deux grandes plages de sable fin du fond de la baie nous avons recense 60 millions d'indi- vidus morts, dont la moitie sur la Grande Greve (St-Efflam). Le cinquime seulement des cadavres etait accumule dans le spectaculaire et nausabond cordon d'choage du niveau des hautes mers; la fraction la plus importante, bien que plus discr&te, etait parpillee en nuages d la surface de Pensemble des plages. Par des transects de plages, realiss avec des gabarits metalliques d'1/4 de ml, par 5 equipes d'etudiants operant durant 3 jours, nous avons obtenu le ecompte suivant pour les principales especes ESPECES Nbre d'individus Poids en matiere Poids brut (cadavres) organique sche(t) t Echinocardiwr. 20.106 4 260 cordatum Cardium eduLe 16.106 5 70 Mactra coraZlina 14.106 4 0 Pharus Zegwen 5.10 6 10 100 Ensis siZiqua 1.106 2,5 25 Lutraria utraria 0,1.106 0,8 10 Dcnax vittatus 1 .1'06 0,04 0,6 TeqNina fabuqla 0,03.106 0,001 0,01 qTetLina tenuis 0,02.j06 0,001 0,01 SOMIqE 1 57,15.1q06 1 25,366 1 515,62 A cette mortaqlite initiale, brutale, s'est ajoutee une importante mortalite ulterieure plus discreke. L'observation des phenomenes dans leur ensemble geographique ne nous a pas permis de cerner cette mortalite durant les premiers mois. Deux campagnes trimestrielles, avec sculement 10 stations rgulirement suivies A St-Ef4qflam, ont pu ere effectues. Ce n'est qu'en janvier 1979, dans le cadre d'un contrat avec ]a NOOA, que nous avons pu mettre en place un reseau d'observations mensuelles mais qui ne couvrait 1'ensemble de la plage qu'en 6 mois. Les peuplements des plages de la Grande Greve (St-Michel et St-Efflam) et de Locque- meau ont q@tudis qualitativement depuis un sie cle par les chercheurs et les tudiants de ]a station biologique de Roscoff. Des tats de r60qfqrqences quantitatifs, itablis sous 0qforme de cartes d'isobiomasses, de 1965 d 1968 (C. Chass, 1972), donnaient, pour la plage de St-E20qf20qf20qlam, des points de comparaison utiles 2qd la fois pour*le suivi des principales esp6ces et pour 'impact des hydrocarbures de value lors du naufrage du TORREY CANYON. qen 1967. 452 PLAGE DE ST EFF LAM A WD I C E 40 5 O-E 4, 9 L 0 C A L IS A T 10 N 19 S U R V I 5 q%, )rag 0. B I GROB N 0 a Ix de 0 i 10 3 3 Rm hu fim E ISREST ..... .... zone etudie'e 3 Ln 2 r ATESULI 2 al 3 3 OURRnEMEZ 3 OUIMPER MCARSEAU :4 10 Extension maximale des nappes d'hydrocarbure 9 du 17 mar s au 28 avrill 1978. (D ,ap ras les ,nfo rmati ons de Is Marine-Nationale). L'ensemble de la nouvelle cartographie des peuplements de la plage de St-Eefefelarn a eke realise avec une centaine de stations quantitatives dont 65 ont ete etudiees d'une manire plus approfondie. Chaque station a fait l'objet de 3 prelevements de sediment effectues d la benne d main de 1/ 16 de M2 sur 20 cm de profondeur. La faune est recueillie sur place par tamisage sous I'eau sur maille de I mm, elle est determinee, comptee et pesee. Des cartes d'isobiomasses des principales espes ont ete dressees, comparables aux cartes anterieures realisees avant et apres le naufrage du TORREY CANYON. Dix stations caracteristiques des principaux peuplements. situes au bas et au centre de la plage, et suivies durant toutes les operations, ont permis d'etablir 1'evolution des especes dans le temps. LE MILIEU Sur lac6te Nord de Bretagne, s'ouvrant surla baiede Lannion, face aux vents dominants de secteur Nord-Ouest, d quelque 20 kin d I'Est de la baie de Morlaix, la (< Lieue de Greve >> est une vaste plage de 5 km' tapissee de sable fin de 100 d 130 emergeant presquie entie- rement aux grandes basses mers. Elie est profondement encaissee entre des collines e1evees, large de 2 kin au niveau des basses mers, elle atteint 4 kin au niveau des hautes mers, d'od son nom de < Lieue de Greve )>. 11 y a 1,6 km en movenne entre ces deux: niveaux. Une butte norme, Roc'al haz haute de 99 in. savance lege'rement en compartimentant falblement le fond de la baie, separant les localites de St-Efflam, d I'0qOuest, de St-Michel-en- Greve, d I'Est. Six ruisseaux issus des coteaux eleves qui bordent le fond de la baie coulent en convergeant, A basse mer, vers I'entree de la baie entre les pointes de Plestin, d I'Ouest, et de Beg-an-Fourri, A I'Est. Les trois ruissellements de I'anse orientale sont les plus importants. Par leur action de dessalure ils sont responsables de I'appauvrissement considerabie des peuplements des niveaux moyens de cette anse, liee, par ailleurs, A 1'expo- sition maximale aux houles des vents dominants du secteur Nord-Ouest. L'anse occi- dentale tie reoit que des apports d'eau douce tres limites (une leg salure estivale apparait). Elie est relativement abrite des vents de Nord-Ouest par la pointe de Plestin, et partiellement protegee des vents importants de Nord-Est par les pointes de Beg-an- Fourn. Locquemeau. et la ete Est de la baie de Lannion. Aussi, le sediment y est-il legerement plus fin, moins permeable et moins oxygen; c'est la zone la plus richement peuple. Elie presente un petit massif de roches metamorphiques noires, dures et tour- mentees : le < Rocher Rouge , qui d I km du fond de I'anse, couvre I ha et s'e1eve depuis les basses mers movennes j .usquIau niveau des pleines mers de vives-eaux. 11 offre un abri permettant le developpement de sediments 1egrement envases en arrire. Un maigre herbier de Zostera nana s'etend au niveau des basses mers de mortes-eaux; il ne modifie que tres peu le sediment. la biomasse des Zostera y tant feaible (250 d 600 g frais au in', moyenne 350). Notons son 1eger deplacement vers I'Ouest, depuis 1968. Le phenomene de sursalure estivale qui s'y produit, dfi d 1'evaporation du film d'eau qui n'a pas le temps de s'ecouler durant la basse mer, est insuffisant pour modifier les peuplements, A moins que I'on puisse lui attribuer la presence tres clairsemee de quelques Nereis diversicolor, Un aspect important particulier qi cette greve est I'accumulation croissante, d'annee en arinde, d'algues non fixees, d'ectocarpales brunes libres au. niveau des basses mers, mais surtout d'ra lactuca verte au debut du printemps et d I'automne, au-eIessus, du niveau de la mi-maree. Ces a1gues couvrent i basse mer, d'un revetement parfois continu, de trs larges ekendues de sable; elles sont animes d'un balancement pendulaire au gre des marees et des vents. Ces <4 marees vertes >> proviennent de 1'eutrophisation croissante de I'impiuvium des bassins de drainage des ruisseaux par les nitrates, eles phosphates, la potasse, les pesticides et les lisiers d'origine agricole. Elles jouent un rele trophique impor- tant notarnment par ]a liberation des spores et gametes et par le support qu'elles consti- tuent pour une riche faune d'Harpacticoides, de Foraminifres, de Cilies et d'Amphipodes (Dexamine spinosa). Elles sont partiellement consommes par les Talitres des hauts niveaux. La destruction des Amphipodes par le petrole explique peut-tre partiellement la particulire abondance des (4 marees vertes >) de 1978 et 19 79. 454 EFFETS DE LA MAREE NOIRE SUR LES PEUPLEMENTS SEDIMENTAIRES Sur les cartes suivantes sont reportes 4 etats successifs de distribution pour chaque espece : 1964-1968, 1979 , 1980 puis 1981 Sur les cartes A, on lit les zones de forte densite des principales especes. Nous comparons 4 periodes, les especes sont representees par leurs initiales Ba : Bathypmeia pitoza, swe et guiUamsonniana (notee tapectivememt P, eS) Ne : Ne/,Line 4qc4q&2qvquqLqt0qaqeqwqs 8qM : 8qMqothoe b4qAevicoqtniqz At : 0qkqtenico0qta maxina Ow : Owenia q4qwqsiqjo@uniqs Tt : Te4qZ4qUna tenui.q6 Tq4 : Teqttina q6abuta Do : Donax vittatu,q6 Oph : Aqc8qAocnida bqAachiata E4qmqsi,q5 qe2qm4qi,q5 et En,q5.qiqz qs4qi8qt0qiqquo- X 4 Mact8qta co8qta0qU-qina Phaqtu.q6 qZeqgumen L Lut8qAax-0qU 0qZu2qtqtaqA,4q& Les e,spq@ces q6tudiq6es re-agqissent de maniq@re diffq@rente cqomme en tq@moigne la courbe de 1'q6volution des biomasses et le tableau Notons qu'il s'aqqit lqa de 8qVq@volution moyenne de 10 stations. Des dq6placements des zones de peuplements visibles sur les cartes Wont pas pu q6tre pris en compte dans le tableau. Les cartes B a 0 prq6sentent les courbes d'isovaleurs en cal/m 2 des biomasses des princi pales espq@ces de 1 a macrofaune endog6q6e. Les facteurs de conversion sont : 1 cal = 1 a de mati6q@re or2qganique fra4qlche = 0,2 g de mati0qbr8qe organi que s4q@che . Les populations de quelques esp6q6ces nq'ont apparemment pas q@t6q6 touch0q6es , elles paraissent m8q6me sq'4q6tendre : par exemple, les deux polychetes errantes Nephty2q4 hombe04qtgii (Cartes B) et [email protected],q-qIi0qon mathi32qtdaqe (Carte C) ce dernier ayant tendance A coloniser maintenant le haut de la place. 455 Le crabe Ptatyonichus latipe (cartes D) dont les noyaux se decalent vers l'Ouest de la greve ou ils s'etendent en densite, et le bivalve Tettina tenuis (carte E), qui aprds une trds forte progression jusqu'en 1980, voit ses effectifs amorcer une diminution sur les noyaux Est et Ouest mais avec le maintien du noyau central fixe. D'autres especes par contre ont beaucoup regresse apres la catastrophe. En opposition a Tettina tenuis, Tettina fabuta (cartes F) qui occupait en 1968 tout le niveau de basse mer, continue -a regresser dans les tres bas niveaux ou elle disparalt actuellement a Vest. L'ophiure Acrocnida brachiata (cartes G) est de moins en moins presente. On assiste a une diminution du nombre des noyaux plus etales vers l'Ouest. Les polychetes sedentaires Arenicota marina (cartes H) et Owenia fusiformis (carte I) ont reqresse apres la maree noire, mais a partir de 1980 ; les deux especes se developpent a nouveau sur le cote Est de l'Estran,bien que pour Owenia on constate une 1egere diminution en 1981. Le bivalve Donax vittatus preponderant en 1968, a probablement bien diminue avant 1978 puis a completement disparu apres I' "AMOCOCADIZ". Apres une timide reapparition en 1980, le noyau central prend de limportance , s'etend vers le bas de la plage et le noyau de 1'Est se consolide (carte J). En 1979 on trouve les Amphipodes Bathypoeteia (carte K) et Urothoe (Carte L) en petite quantite juste au Nord-Ouest de la plage, ils ont beaucoup rearesse apres la maree noire en 1980, ils Wont pas recupere en biomasse mais se sont etendus. En 1981 une ld0qgCere reearession pour Bathypoeteia pourrait etre attribue a une mortalite estei-vale tandis qu'Uetothoe se deplace vers le centre et non l'Ouest de la plage. Les cartes M repredsentent des espeCzces non cartographides auparavant il semblerait qu'elles aient pris de Pimportance depuis 1968 : mais d partir de 1980, Mage44qZona papitticotnis et Gtyceta convotuta voient leurs effectifs diminuer fortement surtout a l'Est de la plaege. 456 En 1980, on a pu cartographier Caxdium edute et Netine citAatut qui avaient disparu apres la maree noire (carte N). Ils sont actuellement en regression. Spiophanes bombyx apparu en 1980 s'etend vers le centre et le bas de la plage (cartes 0) Par contre les trois autres especes apparues en 1980 Patadoneis Amata. Eteone Zava et CapitomaztuA (cartes 0) ont beaucoup diminue en 1981. Notons que Capitomaztuz minimuz est significatif lun milieu pollue (LE MOAL Y.et QUILLIEN-MONOT,1979) CONCLUSIONS I- PERTES DE BIOMASSE DIFFEREES La contamination des organismes n'a pas toujours ete immediatement mortelle. Dans les sediments restes longtemps contamines, certaines especes qui avaient bien survecues au printemps de la maree noire ont vu leurs populations s'effondrer en 6 mois, voire un an plus tard. Le tableau suivant, portant sur 10 stations de sable fin du bas de la plage de St Efflam en baie de Lannion, milieu bien representatif par sa nature et par son degre de contamination, montre que le taux de survie ultime pour les especes qui avaient'bien surve"cues est nettement plus faible que celui enregistre d la fin du printemps 1978 pris comme reference, soit apres la maree noire. ESPECES PRINTEMPS 1978 ETE 1978 1ER SEMESTRE 2me SEMESTRE 1ER SEMESTRE 1979 1980 1981 TELLINA FABULA 1 0,20 0,20 0,035 0109 TELLINA TENUIS 1 0,65 1,39 1 0,88 OWENIA FUSIFORMIS 1 0,75 0,32 0,34 0,36 ARENICOLA MARINA 1 0,72 1,67 3,30 3,11 NEPHTYS HOMBERGII 1 2 0,33 0,42 0,68 BIOMASSE TOTALE 1 0,77 1,13 1,02 0192q5 Le facteur multiplicatif est proche des valeurs ulterieures les plus faibles des dates variees rencontres dans ce tableau soit 0,09 ; 0,65 ; 0,32 ; 0,72 ; 0,33. 457 On peut estimer que les especes qui avaient bien survecues initialement accusent une mortalitd additionnelle raisonnable proche de : (,o9 + 0,65 + 0,32 + 0,72 + 0,33) /5 soit 0,4 La mortalite totale ultime etant 1,4 fois plus eleve que celle calculee pour la fin du printemps 1978. 2- DIVERSITE DES COMPORTEMENTS SPECIFIQUES Le comportement relatif des diverses especes est tres variable et assez imprevisible. En ce qui concerne les effets immediats pour une meme station certaines especes rdsistent parfaitement (Tettina tenuiz, Owenia eusieo,tmi,s) d'autres sont presque integralement detruites (Donax vittatuz, Cadium edute, Bathypoteia, Echinocatdium cotdatum, Phaqtuz tegumen, Ensiz enzis, Enie zUiqua, Macteta cotatetina, LuttaLia etutetakia. Seules les trois premieres especes de cette liste sont timidement reparues aujourd'hui. A plus long terme, les comportements sont aussi disparates TeMna tenuiz non affecte a prospere jusqu'en 1980 et amorce une diminution de meme pour Nephtyz hombetgii. TeUina eabuta, Owenia euzieoxmis, Aeniciofla matina peu affects initialement ont considerablement regresses en 1979, quelquefois tres tardivement bien que certains signes d'un retablissement certain apparaissent (Atenicoeta maetina) een 1981. Mothoe et Bathypoteia qui avaient initialement disparus sont reapparus mais seulement dans la partie la plus occidentale et sans encore atteindre les densites initiales. Depuis 1980 on assiste a la reapparition de certaines especes qui avaient completement disparu apres la maree noire, Donax vittatuz se consolide alors que Netine cittatutue, Enzi enziz et Catdium edute ont du mal a se reimplanter. Des espEces nouvelles pour la localite apparues en 1980 dont certaines seraient significatives d'une pollution residuelle, commencent a diminuer en 1981. On doit donc considerer que les peuplements presentent encore en fin 1981 surtout dans la partie Est de la plage un deseqilibre ecologique profond alors quA l'Ouest des signes d'une recouvrance avancee apparaissent. 458 3- EVOLUTION GLOBALE Il semble s'amorcer une derive qualitative generale des peuplements de sables fins bien calibres tres typiques a Donax vittatuz, Tettina eabuta, Echinocatdium coetdatum et grands Sof-enidae vers des peuplements plus banaliss de sables fins plus eutrophises qu'envases a ektekticota matina. La maree noire. West sans doute pas seule en cause ("marees vertes") mais elle a accelere cette evolution regressive des peuplements originels. Les parties hautes et surtout basses de 1'estran,plus que les niveaux medians,sont les plus touchees. Ceci coTncide avec une plus forte accumulation des echouages des nappes dans le haut de la plage et dans la moitie Est, suivie d'une persistance des hydrocarbures dans 1'epaisseur du sediment. Ceci est conforme a ce qui a de observe sur tout le littoral en matire d'accumulation d'hydrocarbures sous influence des vents d'Ouest dominants. La forte regression constate dans le basdelaplage, confirmee par la nature essentiellement infratidale de la qrande masse des cadavres retrouves dans les echouages, souligne un autre fait majeur en dehors des hauts de plage directement atteints par les nappes, 1'essentiel de la mortalite est a imputer au petrole disperse ou dissout au sein de la masse d'eau. Au niveau biomasse totale, on constate une chute importante en 1978 et 1979 par rapport A 1964-1968. Des affaiblissements des noyaux de peuplements de la partie Est par rapport a ceux de la partie Ouest, et du bas par rapport au haut, sont tres notables. Cette distribution se maintient en 1980 puis 1981 en s'appauvrissant encore. Notons que la proqression des Atenicotes se fait au depend despeces de petite taille plus productives. Il en resulte donc une baisse generale de 1 a fertil it de 1 'ensemble de la baie par rapport a 1968 de l'ordre de pres des deux tiers. Le retablissement encore tres incomplet des peuplements demanderait des etudes ulterieures. 459 BIBLIOGRAPHIE Beauchamp P, 1914. - Les geves de Roscoff. Le Chevalier ed. Paris. Chasse C1. 1:Hardy-Hajos M. T, Perrot Y, 1967. - Esquisse dun bilan des pertes biologines provoquecs par It mazout du ToRm-CANYoN sur It littoral du Trigor. Penn ar Bed, 6, 50, pp. 107-112. Chasse CL et coU, 1967. - La maree noire sur la te Nord du FiniWre. Penn ar Be 6,50, pp. 99-106 Chasse C1, 1972- - Econornie sedimentaire et biologique (production des estrans; meubles des obtes de la Bretagnee T1Wse d'iw Paris VL 1-293. Chasse CL 1978. - Bilm ecologique provisoize de l'impact de rekchouage de AMOCO CADIZ. Inventaire et evaluation de la mortafite des especm touchees. Public. CJVEXO. Chasse CL 1978. - Un indict malacologique pour mesurer rimpact ecologique de la rnaree noire de AMOCO CADIZ sur It littoraL Hatiothis PV : 9, no Z pp. 127-135. Chasse Cl- 1978. - Esquisse d7un bilm ecologique provisoire de rimpact de la maree noire de AMOCO CADIZ sur It fittoraL Public. CNEXO sie acres de colloqu. P, no 6, pp. 115-135. Congres AMOCO CADIZ Brest 7 juin 1978. Cbasse CL Morvan D, 197 - Six mois apres la maree noire de AMOCO CADIZ, Bilan pro- visoire de rimpact ecologique. Penn aeBeti PV, no 93, pp. 311-338. Chasse CL 1979. - Bdm ecologique de AMOCOI CADIZ. Evaluer pour dissuader. J. Peck Oc,ano. no 1, pp. 25-26. Toulmond A, 1964 - Les Amphipodes des me sableux intertidaux de Roscoff Apervus faums- tiques et ecologiques. Cak biol. Mar, 5, pp. 319-341 460 DATh 11164 AMOCO 1978 1979 1980 1981 Nb e e M2 2 1968 CAVIZ "RIL JUIN SEPT. I sem. 1 sem. SEPT. NOV. ITV. AVIU 1, AOUF ESPECES ""ca I /m TELLINA rABULA 157 113 29 15 9/1 S/1 11 /1 2/1 810,05 16 10,3 20 10,3 33 10,6 18 10,4 TELLINA TENUIS Soo 145 624 1104 SS4 163 448 174 810 1145 711 1128 660 1119 560 189 S80 184 630 198 590/93 DONAX VITTATUS 160 157 1/0 2/1 0/1 0/0 10 /1 11 /1,@ 70 11,8 40 12,4 37 /2,5 42 13 Lo I 576 440 294 340 98 143 144 106 136 184 120 us FO.RmIS 5 13 10 3.9 31 22 25 10, 9 11, 4' Idl a 10, OWENIA F I ARENICOLA :MAUNA, 5 1.9 219 i.19 114 31115 7129 9131 .8 /28 8124 10 132 7128- 6' 12 13 19 3 1 26 31 37 SI NEPHTHVS.J(0MBERGI 1 @ /@ /2 13. 13 5 14 6 a 6 15 9 17 0 0 1 0 0 24 24 S8 32 26 BATINPOREIA'.. . 10,7 /0 !0 0 /0 /0 /0'a /0j 10,6. 10,3 /.0,'2, 3 0@ 0 1 0 30 27- 28 29 30 UROTHOE 0,5 1-0,4 0,4 012. 0,2 0 0 0 035 5 CAPITO TO 10 21 80 46 18 MINUTUS 0,01 0,03 0, 16 0107 2 rTOTA L V7,a, 166 9 \_-O@ 146,2 1111'. h4 173 162 165,8' 133,03 125,86. 151,57 17 2 2 Variatidn dans le tenqis-du nombre dlindividus au met de la btomasse.ewcal./m pbur les principales espt%pe' de St-EFFIAM, suite I 116chouage de II AACO -CAD I Z . s de la plage (Le nopbre d'individus a(I m? et'la.biomasse sont des chiffres Ctablis d'apr4s une moyenne de 10 stations du bas et Centre de plage, ce qui repr6sente une surfs6e de 2 m2 au total) COURSE DE L EVOLUTION DES BIOMASSES Moyenne de 10 stations de bas de plage. Les nombres dlindividus au m 2 et de la biomasse en callm 2 pour tous les especes examines (44 especes differentes) sont disponibles sur demande de llauteur. 1000- BIOMASSE(cal/m 2) X + I Tt Tellina fenuis Tf Tellina fabula 500- Ov Donax viffatus Nh Nephtys hombergil Am Arenicola mor1no Ow Owenia fusiformis Both Bothyporeio Uro Urothoe Tt. 100- 50 -4 Ar. 0.W. 10 M I p h ,,,@N e p h. Tf. Bath. 0--- --1 U Tf Uro Don. 8st Uro. 1968 JANV. Both. MA179 JANV. 80 JANV. 81 7S A NIOCO mars) TEMPS (mois) CADIZ 462 4orp ologie de E F FLAM 1'es ran ------- EMPLACEMENT DES STATIONS DE PRELEVEMENTS. RUISSEAVX. k kn 11, 1) 2 NAPPES DE RUISSELLEMENT. 3 GRANDS RIPPLE-MARKS. 4 HERBIER DE ZOSTERA NANA 1979 5 HERBIER DE ZOSTERA NANA 1964-68 I KM G Salinitis C ST EFFLAM intersticielles en 10-3 IT 41 @KM L@_, K M xyg6nation en darcy. an cm. lz 10 0 0 .'7'b ri JV- 1. K M 1 KM _j 463 BIOMASSE TOTALE ej 31/M2 OMASS,f TOTALE Cal/m2 1964 68 1980 ID 0 0- NZ N I K 41 BidmAS'SE TOTALP BIOMASSE TOTALE en Cal/m2 en Cal/.M2 1979 1981 m 3 -209 10 lbo ,30 30 10 400 L -0 Pi I KM Noyaux de peuplements Noyaux de Peuplement, 1964-68/// 1980 0 w Du- w 0 I K M Ln Noyaux de peuplenbents Noyaux de Peuplement 1979 1981 rn @U, Do L no Riomasse en calIM2 tasse en cal/m2 Biorr NeDh tys lVephty5 hom be rgii hombergii 3 1964_68"// 1980 3 3- 3 3_ - 10 -3 N 1 KM Biomasse en a I m 2,/ Biomasse en cal/m2 Neph tys hom ergii 0- b @n =10 3 1979 \% 3 -1 1 3 1 3 B 3 0 N IV 1 K M IN- Biamasse en Cal/M2 Biomasse en cal/m2 Sigo lion Siga I ion Mathildoe mathildoe 3 1964.68,,/ m m 1980 0 Z- IV KM 4- ON -j 810masse en C cmasse en cal/m' Sigalion Sigulion m a t h i Id a e mu@hitdae 1981 M- 1979 La 3 Ib 0 1 KM 2 iomasse en cal/m2 E3iomasse en cal/m Platyorichus P&tyonichu,5 tatipe.6 lotipes -M m 1980 -rn m 3 L 0 IV 1 KM ON 00 Biomasse en Cal/m2, Biomasse en C@@ D P io ly onic hus I latyonichu,3 latipes 1979 @-3 0 off L 3- L .@y 0 46,4 1 KM Biomasse en cal/m2 Siomasse 2n cal/m2 re I @inq Tellina te 1) u i s tenuis -3 1964-6@ -M - -- 1980 3 50 3 k L 1 KM :Diomasse en Cal/m Te Ilino tenu is loo Zena i,3 1979 rmn 1981 io 0- lu 30 so- 36 3 10 3 3 3 3 30 No To- Z- kv 1 K M Biomasse en c 2 z", al/m Biomasse eb. cal/m2 I-elli na Tetllna fabutc fabula 1964-68,,, 1980 7 3 0) O)a 0 -3- 30- 0 3 3- 7, ly Z- IV - 1 KM Biomasse en cal/m 2 Bio asse en Cai 21 Tellino 7e lina fabula uga 1979 '--MM qj -M M 0.1 --7 iv I KA4 L Blomasse en cal/m' Biomasse en ca lj 2 m Acrocnida Acrocn;do I/ brachiata ,brachiata Nerels diversicap," -M m - 1964-6: -mm tor r @,6 1Q- k 3 0 lu 1 K M Biomasse en cal/m2 Biomasse en C I/M Acaocnida A cro cn i d a brachiato 1979 f-M m 1981 Mm- 3 k k 0 d 1 KM Biomesse en cal/11.1 /,n2 iomasse en cal Arenico to A r e n i c a to Marina 'marina 1964-6,: 1980,l/ 30 30 60- 30 k 1 KM Biamasse en cal/m Nomasse en Cal/M A renico to A/L en i co to Morino 10 maz in a 1981 1979 (30 30 3010 H 3 3 W/ 10 1 K M v BNomasse en calmy Biomasse en C81L,rr,' ;wenla Owenta fusiformIS fu s 1 fo r mis 1964.6: e 1980 6 30 _0 10 10 30 - 00- IF 3,0 1 KM Blomasse en cal/e7 iomasse en Owen,a Owenia fusiformi. lu,6iloAmi6 1979 1981 -M NIL 0) 3 --lo -30- 3 3 IV 1 KM Biomasse en cal/ 2 n cal m Donax Donox vittatus villolus 1964-68/ 1980 3 30 10- e 0 k .3 Z- 1 KM M.r r. 'm Cal/ Donox Donax Biomasse an cat/M2 vittatus -M m 1981 -Mm- 1979 3- k Pi 1 KM Biomasse en calFIr-F iomasse en cal/rr"l "l/ Bothyporeia Bothyporeia Q3 lo=- 1964.6: 1980 Q3 0.1 03 1 KM Ln Biomasse en cal/m2 Bilmasse en C But hyporeta 1979 - - ------ 1981 3 46 0 re,a G I K M Biomasse en calrmz Biomasse en cal/ril Urathoe Urothoe 1964-68 1980 OL3 Z- tv 1 KM giamasse en cal/rre- Hiamasse en C:@@ Uzo@hoe UtL)fhoe m 1979 cr NN 46 ID) 3 0;3 0.1 t 0 Ily 46 IV I KM Biomasse en cal/m2 Biomasse en cal/rr.7 @fogelono Glycera Pilpt Ilicorm c o n v o I ua 1980 1980 0.1 43.b 1 03 .03 OL3 (0 7 I KM 41 BiGmasse en Cal/ iyage@onu pa - Pig4gicozni-6 convoluia 4v 1981 Q3 Cl t k L .6 IKM Siomasse an cal/H' CaAdium e&tte Spiop hones NeAine ci)tAa- tutuz III ---I- I X\ b 0 m bqx 2 cat/m 093 1980 1980 01, 03 0 1 KM Z_ N co Krumm BlomassO an Cal Spiophanez Caadium edutp 90m9ix N in P- 4/Z autu lau 1981 1981 M M - 3 16 k r/F/, Q3 0 ft7 IV 1 KM iomasse en cal/m2 Biomasse en cal/m2 Paradoneis armata m capi tomastuz Q0 K Eleone fl ov., m p - - - - - - - - - 1980 1980 z ------- z 16 aol 0,03 3 Z- pi 1 KIM liomasse en Ca Biomasse en Ca I M/ ca'? i maj tu, lla4adonpi.6 a4maia m t n t n2u 1981 mm C4 1981 0, %% 1 @17111 0-- 01 A# 1.K M I I i I i I, @ 3 6668 0000116799