[From the U.S. Government Printing Office, www.gpo.gov]
The Lake Erie/Presque Isle Bay Fish Flesh Study 1987-1988 Prepared By: Erie County Department of Health December 1989 SH 174 L35 Ought 1989 F.PM ad Coordowtod no amt. of ("Woolloots 111211ort" coon I, @"Mtn sop@ 110.1 woolor 10"Woorm III.% coats low Muspow -10hE 87-PS.02 SH174.L35 1989 THE LAKE ERIE/PRESQUE ISLE BAY FISH FLESH STUDY 1987 - 1988 US Department of Commerce NOAA Coastal Services Center Library 2234 South Hobson Avenue Charleston, SC 29405-2413 Prepared by Erie County Department of Health Coastal I and Coordinated through Dept. of Environmental Resources Office of Resources Management Div. of Water Resources Management Div. of Coastal Zone Managealist Zone The preparation of this report was financed in part through the Pennsylvania Coastal Zone Management Program under provisions of the Coastal Zone Management Act of 1972, administered by the Division of Coastal Zone Management, Bureau of Water Resources Management, Pennsylvania Department of Environmental Resources Table of ConteDtS Page No. I. Introduction I II. Study Design 5 III. Quality Assurance/Quality Control 9 IV. Study Area 13 V. Selection of Target Species 16 VI. Equipment, Fish Collection and Processing 18 Field Procedures 23 Recommendations on Field Procedures 26 Laboratory Procedures 28 Discussion on Cblordane Results 28 Discussion of Lead.Results 31 VII. General Discussion 37 VIII. Recommendations 41 IX. Summary 43 Appendix A: FDA Fish "Action Levels" Appendix B: Sampling Locations Appendix C: Species Sampled and Common and Scientific Names for Fish Appendix D: Field Collection and Preparation Protocols Appendix E: Study Areas - Lengths and Weights of Fish, Collection Methods and Dates Appendix F: Sample Results - Organic and Inorganic, Percent Moisture and Percent Lipids Appendix G: Lead Results - Walleye Sections Appendix H: Preparing PCB Contaminated Fisb For Human Consumption Appendix I: Food and Drug Administration Guides - PCB's, Cblordane Appendix J: Quality Assurance Data - Comparison to Otber Agencies Appendix K: Memo Discussion on Testing Procedures References Acknowledgments This report was prepared by the Erie County Department of Health in Pennsylvania. The principle author is Robert J. Wellington. Mark Fedorchak and Douglas Ebert of the County Health Department provided invaluable assistance in the operation of the County's boat, fish collection, fish processing and related issues. The kind assistance and encouragement from Eric Conrad and others of the Division of Coastal Zone Management of the Pennsylvania Department of-Environmental Resources is appreciated. Special appreciation also goes to Robert Frey of the Bureau of Water Quality Management of the Department of Environmental Resources. His coordination of efforts and suggestions were most helpful. Also, thanks to Raymond Hasse of the Bureau wbo'belped in providing equipment and assisting in the collection and processing of some of the fish. Laboratory results were provided by the Pennsylvania Department of Environmental Resources, Bureau of Laboratories. We appreciate the help we received from Floyd Kefford, Vince White, Sam Harvey, Alan Bruzel, Dennis Neuin and all the personnel associated with the laboratory. Their patience in dealing with unexpected problems is appreciated. The Pennsylvania Fish Commission's assistance in providing us with the scientific collecting permits, as well as their suggestions on where to collect certain species of fish, is appreciated. The kind assistance of the Michigan Department of Health is appreciated. They graciously agreed to split a limited amount of samples with us as part of our QA/QC verification. Their assistance was most helpful. Special thanks also goes to Christine Sanfratello, our typist. We also acknowledge the help in the form of suggestions, literature searches, etc., provided for by other agencies, such as the U.S. Environmental Protecti on Agency, U.S. Food and Drug Administration, U.S. Fish and Wildlife Service and any and all others who contributed in any way to the production of this report. Abstract Fish were collected from Presque Isle Bay and the Pennsylvania waters of Lake Erie, by the Erie County Department'of Health. The fisb were analyzed at the Pennsylvania Department of Environmental Resources (DER) @laboratory in Harrisburg, Pennsylvania. They were tested for select organic and inorganic contaminates. The purpose ..of the study was to broaden the base of knowledge on fish contaminants in the.local area. This study was not intended to be the final word on the subject, and any choice on whether to eat or limit consumption of fisb is left up to the Individual. There is a great deal'of uncertainty as to exactly what might be considered "safe" to eat. Pennsylvania issues advisories against eating certain species of fish, @wben there is evidence to show that contaminant levels exceed the United States Food and Drug Administration (FDA) $faction level." 'Currently tbe'S'tate advises against eating carp and channel catfish from the Pennsylvania waters of Lake Erie. Recent controversy, however, bas developed witb respect to,tbe adequacy of the FDA's action level In adequately-protecting the health of sport fisbermen who may consume more than the "average" amount of fisb. An FDA market basket survey reportedly indicated that the average American consumes abou t five pounds of fish per year. Some sportsmen and their families may consume considerably more than the FDA estimate. Because organics, such as PCB's, accumulate over time in the human body, there are health concerns that need to be considered. In this study we have listed the FDA's action levels as points of comparison. However, because of the rather recent questions on the safety of such action levels, we do not mean to imply that just because certain fish tested apparently don't exceed the action level that they are ".safe" to eat. Such a determination is well beyond the scope of this project. Eighteen species of fish were collected and their edible fillets were tested for fourteen organic chemicals, including PCB's. They were also tested for eleven metals, percent lipids ("fat content") and percent moisture. Six species of fish did not reveal any amount of organic contamination above the detection level of the laboratory test. None detected (ND)-sbould not be equated with none present, because it is possible that some would be found if lower detection methods were used. Except for cblordane in five species, none of the eighteen species of fish analyzed by the Pennsylvania DER .exceeded any of the FDA action.levels. [Note i n a split sample with the State of Michigan (a large lake trout), Michigan did find.PCB's to be just slightly above the 2 ppm FDA action level.] As noted above, five species of fish bad values of "technical cblordane" above the FDA action level. They included carp and channel catfish (currently on Pennsylvania's advisory list), as well as lake trout, gizzard shad and a largemoutb bass. A significant question arose regarding the testing procedures for cblordane used by the State of Pennsylvania, compared to methods used by some other agencies. Depending on the methods used, cblordane results can be significantly different. For example, while the DER laboratory found 0.32 ppm cblordane in a largemouth bass (the FDA action level is 0.3 ppm), the Michigan Department of Public Health found none detected in the very same bass. Other data also showed that Pennsylvania might be overestimating the cblordane results, as compared to other agencies. An intensive review by the Pennsylvania DER laboratory of the chlordane testing methodology was undertaken. It is likely that, in the future, testing protocols will become more standardized so that inter-agency comparisons will be more meaningful. Of the eighteen species of fish tested none exceeded the FDA action level for mercury. Mercury currently is the only metal that has an established action level. Problems relating to quality assurance were discovered regarding lead analysis. The original lead results, as noted in this report, are apparently overestimates of the true values. A limited amount of followup laboratory work revealed that some of the originally reported lead results were up to about a factor of ten higher than the followup testing results using improved methodology. Refinements in laboratory methodology were undertaken, and future test results will reflect the changes. Additional fisb collections, already underway, along witb increased field and laboratory proficiency, sbould belp clarify the above issues. For example, for our "routine" water quality work for 1989, yellow percb, cbannel catfisb, walleye and Lake trout from Lake Erie bave been scbeduled and collections started. As new data on fisb from Erie County is generated, it sbould be available for the public. Interested parties may wisb to contact the Erie County Department of Healtb, 606 West Second Street, Erie, Pennsylvania, 16507, to cbeck for updates on the fisb analysis. i_2 - 7.7- P Weekday Ice Fishermen Presque Isle Bay Study Area "V1 zTh.is and other areas can get "very" crowded on weekends. ( T PLioto March 1989, R- J. Vellington) I. Introduction In October 1987 the Erie County Department of Health initiated a study of fish flesh contaminants in select species of fish from the Pennsylvania section of Lake Erie and Presque Isle Bay. The study was funded in part by the County of Erie and a grant from National Oceanic and . Atmospheric Administration (NOW through the Division of Coastal Zone Management, Commonwealth of Pennsylvania Department of Environmental Resources (see page 5). The purpose of the study was to determine the levels of certain organic chemicals and heavy metals in the edible portion (fillets) of fish (see figures 1, 2 and 3 for general study areas, pages 6-8). The Erie area is a center of attraction for sport fishermen, as well as providing a limited commercial fishery. The area offers unique diversification in sport fisheries. Both warm water and cold water fish are abundant. It is possible to fish for lake trout and salmon in the morning and for perch or walleye in the afternoon. Not only does the open lake provide excellent fisheries, but Presque Isle Bay offers a protected harbor for fishing even when Lake Erie is too rough to fish. In addition to the summer fishing, the bay also offers a good ice fishery for panfisb (see photo on previous page). Some winter catches in the bay may include an occasional rainbow trout, cobo salmon or northern pike, which adds interest to the ice fishing trip. Unfortunately, this protected harbor has received considerable amounts of pollution from the Erie area over the years. There was enough concern over the bay that the bay has been recommended to become the 43rd Area of Concern by the Science Advisory Board and the Water Quality Board of the International Joint Commission (IjC).2 An area of concern is an area designated where there is a major problem(s) and impairment of use(s) due to pollution e.g., closed bathing beaches, dredging restrictions, etc. Because of the high interest in the lake and bay fisheries and the general lack of information on fish contaminants from our specific study area, it was decided that more information was needed so that the public could make better educated decisions on the advisability of eating fish from these areas. This study, because of its limited resources, is not meant to be the final word on local fish contaminants. It is to serve to supplement existing data and provide the impetus for additional data collection efforts. The reader is cautioned that the numbers of fish collected are very small compared to the total population in the lake and bay and may not represent the "average" fish contamination levels of a particular species. Also, fish of different ages and/or sizes may and probably do contain different concentrations of contaminants. Likewise, fish from the eastern portion of the study area, for example, may or may not contain different contaminants and levels of contaminants from those to the west near the Ohio line. 2 The cboice of parameters analyzed for was based on some of the more commonly found contaminants, wbat the laboratory could run and wbat was affordable under the Coastal Zone Management grant. It is possible that otber contaminants not tested for could be in the fisb. It is also likely that some contaminants are present in concentrations below the laboratory's detection level. Tbe.term not detected (ND) sbould not be equated witb not present. Wbe.re Federal Food and Drug Administration (FDA) guidelines are available, we bave noted them (see Appendix A). They ar e used by some people as a general reference point but, again, caution must be exercised. The FDA action levels were not intended by the Federal government to serve as "safety" guidelines for localized populations wbo may be eating many more sport fisb caugbt than the "average" American. 3,4 It is likely that the FDA and U. S. Environmental Protection Agency (EPA) and/or otber cooperating agencies will furtber refine wbat are considered to be acceptable "safe" risk levels. Mucb needs to be evaluated, especially the buman bealtb effects of multiple contaminants in fisb tissue. We believe this study does provide a meaningful start on determining contaminants in fisb, but we must advise the reader that the decision on wbetber to eat certain fisb is an individual cboice. It is suggested that consumers of fisb keep tbemselves updated on new developments and be aware that "new" cbemical contaminants in fisb likely will 3 be discovered. "Acceptable" risk levels may be adjusted as more information is available. Qualified pbysicians sbould be consulted about the advisability of consuming fisb. Particularly the risk to cbildren, pregnant women and women of cbild-bearing age sbould be considered. Included in this report, as noted above, is a list of FDA action levels for informational purposes. It must be pointed out even if no tested-for-parameters for the specific fisb exceed FDA action levels, there is no guarantee those species of fisb are safe or are not safe to eat. On the otber band, literature indicates that eating fisb may bave beneficial effects. 5,6 4 II. Study Design The Fisb Flesb Contamination Study was initiated because of the public's concern over eating fisb from local waters. The Pennsylvania Department of Environmental Resources (DER) Division of Coastal Zone Management (CZM) provided an 80% grant for the study. The Erie County Department of Healtb provided the additional 20% witb botb monetary resources and in kind services. The Pennsylvania DER Bureau of Laboratories p rovided the fisb flesb analyses under contract to the County of Erie. Fisb were collected in accordance witb Pennsylvania Fisb Commission permits. The general collection metbods included the use of gill nets set in the lake and gill and boop nets set in the bay. Rainbow trout were collected from Trout Run by using dip nets. Fisb collected in Presque Isle Bay during the winter of 1988 were captured by book and line tbrougb the ice. Fisb were weigbed, measured and processed in the field. Fisb were frozen and sbipped to the Pennsylvania DER Laboratory in Harrisburg, Pennsylvania for analysis (see Appendix D for protocols). 5 Fi&ure I 0 N T A R 1 0 4or z Live 4&* dW N E W Y 0 R K -$OVA Presque Isle -3 State Park Erie Erie dW Untz 400 Cleveland P R N IN S Y L V A N I A 0 R 1 0 (not to scale) Figure 2 L K E I E x E. N SHADES BEACH REAS STUDY A 101 Gael PRESQUE ISLE BAY If A ij TROUT' RUN x STUDY AREA 9,1 ;@ T ABOUT 10 MILES Ek 1. if SOUTHWES, 4,; o""WLI" wh;1: J., iihl"ow k I-N" 1.11 A9.1 -NA As Ott a got 0.1 -Awllu. As Am. q saiv Apn3s A v q z a s I z n s a I a rc vrajV Spn3g r r rov. 41ic III. Quality Assurance/Quality Control (QA/QC) Quality control assures that test results are consistent and reproducible. Duplicate samples of the same test tbeoretically sbould be very close to eacb otber. Blank samples are samples that sbould not contain measurable amounts of the contaminants to be tested. They belp assure that there is no outside interference or contaminants being introduced into the testing process or errors in the interpretation of results. Spike samples belp determine the percent recovery of a particular element or substance in a particular test. In a spike sample, a known amount of contaminant is injected into the sample. The actual recovery of a spiked blank sample sbould be pr oportionately very close to the amount put in. If there are significant differences in concentrations one way or the otber, it could be inferred that eitber the laboratory could not find a particular substance even if it were tbere, or was perbaps finding mucb more than was tbougbt to be present. Quality assurance demands that not only are the results consistent and reproducible, but that indeed they are reasonably correct and that otber laboratories sbould be finding approximately the same levels of and types of contaminants. For some types of samples, a laboratory can obtain a certified standard, that is, a standard known to be witbin a certain tolerance. However, for some parameters it is difficult, if not impossible, for a supply bouse to certify the amounts of contamination in a commercially 9 available fish tissue. Because of the ubiquitous nature of some organic chemicals, it is difficult to find fish with absolutely no contaminants. Also because certain contaminants may be more concentrated in fat than in muscles and that fat is not uniformly distributed in fish, getting a perfect bomogenate" is, at best, very difficult if not impossible. In the absence of known samples for many parameters in fish, then "round robin" analysis of a particular sample by various laboratories may be the only reasonable substitute for the situation where there are no certified quantities available. "Round robins" involve having various laboratories participating in the analysis and recording their individual answers. Later the results can be prepared and a consensus may be taken and/or standard deviations determined, etc. Initially we bad hoped that there would have been more of a QA protocol prior to the running of any test for our project. However, because of monetary and time constraintst the State DER laboratory agreed upon the QA/QC program which would work in QA/QC along with routine testing of our CZM fish. The--program specified for every four fish tested, there would. be a spike and a duplicate sample (see Appendix J) and that the DER would obtain and check some reference samples. An actual lake trout sample of "known' chemical contaminants was obtained for the purpose. It was provided by the U.S. Fish and Wildlife Service and labeled "85-C." 10 This cbeck 85 -C came as the result of many round robin tests of the fisb. Wben our laboratory tested this sample, it appeared clear to us tbat there were differences in the cblordane results, that is, the State DER lab was sbowing considerably more cblordane than some otber laboratories (see Appendix J). Because of this difference and the fact that in the past the State lab found more apparent cblordane that otber laboratories, the Erie Count y Department of Healtb formally requested that the State DER laboratory try to resolve the question as to wbetber they were correct witb respect to the cblordane results, or at least witbin an acceptable window of accuracy (see Appendix J). In discussing the cblordane test results, it was suggested by an outside agency tbat, as the State of Micbigan bad considerable experience and expertise working witb cblordane and otber contaminants, perbaps they could belp us. The Micbigan Department of Public Healtb CEHS, Division of Laboratory Services was contacted and they agreed to test tbree fisb samples for us. We cbose the rigbt fillets from a composite of cbannel catfisb (CZM 32, 0692285), a largemoutb bass (CZM 18, 0692288) and a large lake trout (CZM 6A, 0692264). The Micbigan Department of Healtb found considerably less cblordane in the same fisb than did the Pennsylvania DER (see Appendix J). It was also clear that wbile Micbigan was looking at certain isomers and components of cblordane, the Pennsylvania State DER laboratory was looking for "technical cblordane," rather than adding the sums of selected cblordane components as Michigan was doing. It should be noted in comparing the data on DDE' both laboratories showed the single-peak component DDE test results to be virtuallyidentical (see Appendix J). This indicated good QA/QC for DDE but pointed out the question of quality assurance regarding the multiple components of cblordane. Subsequent to comparing the testing results of the three fish between the two laboratories, Michigan sent their extract from their right half of the Pennsylvania lake trout in question back to Pennsylvania. The Pennsylvania State DER laboratory tested Michigan's extract again and found virtually identical DDE results and found a PCB isomer; but in looking at two cblordane isomers, alpha and gamma, again Pennsylvania found more than Michigan, which raised a series of discussions concerning test procedures. The resolution of these concerns was determined to be beyond the scope of this project. EPA and others are currently working on a revised protocol for conducting chlordane testing in fish samples. 12 IV. Study Area The study area included open Lake Erie waters of Pennsylvania east of Presque Isle State Park, Presque Isle Bay and Presque Isle State Park waters, and Trout Run, a small tributary stream which enters Lake Erie about ten miles west of the City of Erie (see figures 1, 2 and 3). The general locations of the catches were primarily in Presque Isle Bay or Lake Erie to the east of Presque Isle State Park. The decision on where to collect fish was due, in part, to the fact that most of the major Pennsylvania discharges in the area enter the bay or Lake Erie to the east of the bay. Also, because Presque Isle State Park provides natural protection from westerly winds, it was determined to be somewhat safer for the collectors to set nets to the east of the park. Had we been able to have more samples tested, we also would have attempted to collect fish from the area near the Ohio line and in deeper offshore waters of Lake Erie. Although most collection locations are designated by latitude and longitude, it must be pointed out that these are general locations as determined by the Loran C navigation system and/or coordinants on lake or bay charts. For example, because more than one net was sometimes set in one area, the far end of one of the nets might be a distance from the starting point. Also, when we collected fish through the ice, it was necessary to move around in the general area of a given latitude and longitude until a sufficient number of fish were collected. 13 Most of the fish collected in the bay (see study area A, page 8) were collected at various water depths between the mouth of Cascade Creek and the Erie Yacht Club. This area is on the south side (City of Erie side) of the bay, and the area has received considerable amounts of waste water over the years from Cascade Creek.7 Because of the physical nature of the bay off the Cascade Creek sampling area including the creek mouth, weed beds, rock bars and some deeper water areas, we were able to catch a wide diversity of fish species. Some fish were collected in the lagoons on Presque Isle State Park (see study area B, page 8). This is a popular sport fishing area. It is somewhat removed from any direct waste discharges but is subject to flows into and from the bay, depending on wind direction and water levels. It is suspected there may be fish migration between the bay and the lagoons during certain times of the year. During the winter of 1988 yellow perch and bluegills were collected in study area C west of the Erie Yacht Club and east of the western end of the bay. These fish were collected by book and line. Rainbow trout were netted from Trout Run. These were Lake Erie fish that ascended the creek for spawning (see study area D, page 7). Fish were also collected off the area of Shades Beach, which is located about eight miles east of Erie. This is a 14 popular spot for fisbing for smallmoutb bass and walleye (see study area E, page 7). Some yellow percb were collected between Sbades Beacb and Sborewood Beacb to the east. 15 V. Selection of Target Species Fish were collected according to several criteria including their importance in the sport and/or commercial catch, their level on the food chain and their availability. It will be apparent to the reader that cobo salmon, a popular sport fish, were not sampled in this study. There is an ongoing cobo sampling program as part of the Great Lakes International Surveillance Plan (GLISP). Sampling cobo in this study would have duplicated that program. In general, in our CZM study larger sizes of fish were used when available, as we believed that the larger fish would be older and would have been exposed to contaminants for a longer period of time. However, IJC reported that in one study 5+ year old walleyes "contained lower PCB levels than both age 1 and age 4+ fish." The reason for this was not clear. It was suggested that it may have been due to bias because of small sample size.8 Originally we bad hoped to look at various sizes of fish of the same species but, because of limited time and money, we generally were not able to do this. We did collect two sizes of rainbow trout and one very large and one smaller lake trout. We delibera tely sampled terminal predator fish, such as northern pike and a muskellunge, as well as bottom feeders, such as channel catfish, carp and brown bullbeads. We believed terminal predators might show the results of 16 blomagnification of contaminants, and bottom feeders would be in more direct contact witb contaminated sediments and food cbain organisms that live on or near the bottom sediments. .* We also looked at yellow percb in Presque Isle Bay during different seasons of the year. This sampling was conducted because it bad been suggested that body burdens of contaminants in fisb may be different at different seasons, depending on factors sucb as feeding babits and/or pre- and post spawning conditions (see Appendix C for species collected). 17 VI. Equipment, Fish Collection and Processing Most of the fish in this study were collected using gill nets. Methods of collection are noted with other general information on the data sheets for each of the individual sample results. Besides gill nets, fish were captured with dip nets (from Trout Run), boop nets in Presque Isle Bay and the lagoons on Presque Isle State Park and book and line through the ice on Presque Isle Bay. Gill nets varied, but generally were 50 to 100 yards in length and were 6 feet in height. Nets were set on the bottom of the lake or bay. Net mesh sizes ranged from 1-1/4 inch square to about 2-1/2 inch square. The smallest net, 1-1/4 inch square, was a multifilament float and lead type net. All other gill nets were monofilament lead and float type nets. Some of the gill nets bad the lead weights internally inserted in the lead line on the bottom of the nets. Other nets bad the lead weights attached to the exterior of the bottom lines. The gill nets after use were usually stored in galvanized metal wash tubs. Nets were generally set one day and pulled early the next day. Nets were carefully pulled by band into the boat and loaded directly into the metal tubs. When very large fish were encountered, they were generally taken from the net and then placed in another wash tub. The procedure we generally used was to line an empty tub with clean.aluminum foil, dull side up (the shiny side is treated and may not be suitable for our testing purposes), and place the large fish on the 18 clean foil to avoid contamination. We were careful never to buy gasoline when we bad fish on our boat (to avoid the possibility of getting any gasoline on the fish). When we did purchase gasoline, we were very careful to avoid spilling gasoline or otherwise contaminating the work area. Smaller fish were brought aboard the boat and fish and nets were placed in the metal tubs. Later, after all nets were pulled in, we picked the fish from the gill nets and placed them in a prepared tub or a clean, stainless steel bucket until they could be processed. The same general procedures were used when we used hoop nets or fish caught on book and line. In selecting fish for analysis, where there were more than five fish of the sate species, we tried to pick the five biggest fish. We assumed that the largest fish would likely be the most contaminated and also are the ones most likely to be kept by fishermen for consumption (see Appendix H). Where we caught less than five fish of a particular target species, we used all the fish that were available, unless there was a very significant size difference. For example, if.we captured three large fish of a given species and one very small f.ish, our policy was to keep the three similar size fish for analysis and discard the very small fish. Therefore, our composite would include three, rather than four, fish. We believed that radically mixing sizes of fish would provide less meaningful information. 19 Once the fisb were collected, they were prepared as follows: lengtbs and weigbts were taken and recorded, along witb date of capture, location of capture, metbod of capture and any otber relevant information. Before tb1s CZM project was initiated, our procedures were discussed witb the Bureau of Water Quality Management of the Pennsylvania DER. We agreed to use their sampling and processing protocol (see example of DER's field data collection sbeet that we used, Appendix D). It was agreed that we would analyze standard skin on fillets. This was in keeping witb work being done in otber Great Lakes areas. The idea was to try to provide sample results that could be compared to similar samples in otber areas. It sbould be noted that FDA uses skinless fillets for catfiab. However, we left the skin on catfisb and bullbead fillets as we believed this would be the more conservative approacb. Any future studies are expected to bave the skin removed from catfisb samples wbicb is in keeping witb the FDA protocol and more in keeping witb bow most fisbermen clean their catfisb. Wbere fisb species were encountered witb pelvic fins on the 'standard fillet," we elected to remove these fins (see Appendix D). Otber fisb were scaled and cleaned witb commercially available metal fisb scalers. The scalers were cleaned eacb time prior to use. Our protocol for the cleaning procedure involved rinsing the scaler. After being cleaned in this mariner, the scaler was rinsed witb pesticide grade bexane to attempt to remove any possible organic contaminants. 20 After the fisb were scaled, they were filleted using a stainless steel knife. The knife was cleaned eacb time in a manner similar as described for the fisb scaler. It sbould be pointed'out that the scaler and the knife were cleaned between eacb composite sample. This was done to eliminate cross-contamination of fisb samples. Filleting took place on clean aluminum foil, dull side towards the fisb. Aluminum foil was replaced witb new, clean foil between composite fisb samples. The foil was not replaced nor was the knife cleaned for individual fisb witbin a composite. Wben we started collecting fisb in the fall of 1987, once the fisb were filleted the fillets were scraped reasonably clean, wrapped and frozen. Later we learned that the fillets were not later wasbed by the laboratory, ratber the fillets were ground up in a frozen or semi-frozen state and processed. This meant any slime or scales, etc., still on the fisb fillets ended up in the edible fillets. To remedy tbis, on at least one occasion (5/4/88), some fiab were rinsed in "clean" nearsbore waters. The wisdom of this was later questioned and subsequently the practice was discontinued. Later, more care was given to more carefully scrape as mucb mucus and loose scales off the fillets as reasonably possible witb the knife blade. However, there is little doubt some mucus and scales remained on at least some of the fillets. 21 We filleted both sides of each fi sb. We securely wrapped all left side fillets for each composite sample in clean aluminum foil, dull side towards the fish. The bright side of aluminum foil has a special factory coating that might interfere with test results. The foil was taped closed and labeled with species name, location, date and an identifying number (see Appendix D). The right fillets were processed in a similar manner. The packages were then placed in food grade plastic bags to further prevent the fish from contamination, prevent tearing of the aluminum foil and to retard dehydration. It was our general policy to send the left fillets to the laboratory and keep the right fillets for duplicate sample "backups." The backups proved very useful when additional testing on the composite was done. As fish were collected and frozen, the laboratory was contacted and at appropriate times fish were shipped to the laboratory in coolers with dry ice. The laboratory was always alerted as to when they should expect to receive the fish so that they could put them in their freezers as soon as they were received. 22 FIELD PROCEDURES To eliminate variables due to procedures, it is suggested that in future studies, where possible, whole fish be shipped.to the contract laboratory. This would help eliminate possible contamination in the field during fish cleaning operations. An associated problem with field preparation we encountered was rough water, making it more difficult to use measuring scales to come up with precise live fish weights. Questionable sources of "clean" water were another concern. We assumed that as the fish themselves came from the water, it "wouldn't hurt" to wash our scalers, rinse our knives and bands in the water. Our policy was not to rinse instruments on the downwind side of the boat and use the water only on the "clean" side of the boat. We followed up wasbings of the scalers and knives with a hexane rinse. Although we were always very careful to avoid contamination, there still is always the question of bow clean was the water alongside the boat. Rinsing knives with bexane itself presented a potential danger on the boat. Hexane is extremely flammable and needed to be carried in a clean glass bottle. Care bad to be taken not to spill the bexane or break the bottle on the boat. The danger from fire is increased when such a volatile substance is carried in a glass bottle. Other problems included waves, which could increase the probability of the worker getting cut on a knife or producing a fillet that was not as uniform as hoped for. 23 One of the protocols we used included taking*dry ice on each and every trip. Dry ice would be needed in certain cases where there would be quite some period of time between catching tbe fish and when they are put in a freezer. Often we processed the fish at the end of the day. The fish were then placed on dry ice, but perhaps an hour or two after being put on the ice the fish were removed from the coolers and placed in our freezer. Fish fillets were cold when removed from coolers but were usually not frozen, as they bad not been in contact with the dry ice long enough. It is believed that using dry ice is a good idea, but in some cases should be left to the discretion of the collector. For example, during the winter, according to protocol, we took coolers with dry ice out on the bay ice; however, we do not believe this was practical or necessary. Having tried this, it is recommended that fish not be processed at all in the field in the winter. If it is necessary to process the fish before sending them to the contract laboratory, they should be processed in an appropriate inside location. Difficulties we encountered were trying to bold wet fish and a sharp knife with very cold bands. In fact, it is dangerous to do so, as fingers lose control over the knife when it is very cold. Accidents are very likely. Also, if gloves are worn, even intermittently to warm bands, their cleanliness both inside and out comes to question. Also, ice may start forming on the knife. Therefore, It is obvious fish cleaning preparations during the winter should not be conducted in the field. 24 As noted above, not only was clean water a concern for rinsing our knives, but the question arose as to bow to properly clean the slime and scales off the fisb once they were filleted. This could be important because mucus can in certain instances apparently affect lead results. 9 Scales and slime on test fisb may not necessarily represent wbat a fisberman migbt be eating. The problem is bowever finding suitable rinse water or a way to clean the fisb. Wben we commenced the project in the fall of 1987, we did not rinse fisb. Some scales and slime were likely unavoidably included in the sample. After learning that the laboratory did not rinse the fisb, we did rinse some fisb in nearsbore waters on May 4, 1988. However, the practice was discontinued. During our fisb collection trips, we did not take water witb us. We cbose not to use plastic containers because of the possibility of organic contaminants from the plastic. The general way we bandled the fillets, at least towards the end of the study, was to carefully "wipe" the fillets as reasonably free of scales and slime as possible using a knife blade. However, some scales and slime likely were still left on the fillets. Again, this is anotber reason not to process fisb in the field. 25 RECOMMENDATIONS ON FIELD PROCEDURES Ideally, it seems fish would be shipped whole to the testing laboratory and they would process fish. However, if field processing is to continue, some mechanisms for washing the fillets prior to freezing should be considered. Fish should not be processed in the field during very cold conditions. It is recommended that edible fillets continue to be collected from both sides of the fish. The right half and left half can be frozen in separate packages. One side can be held in reserve should the original be lost or destroyed or if confirmation of a particular contaminant is needed. These samples also may have future value if the same or another laboratory wishes to use the fish to check their results or perhaps even check for additional parameters if problems are found or suspected or if detection techniques improve. It is suggested that investigators consider keeping the "other half" of the samples for a minimum of six months after any final reports are issued. This would better assure that outside concerned parties would have time to read the report. If comments were generated regarding the results, it would be possible to re-evaluate the issues by making the frozen samples available for more testing. With respect to the particular study area, it is suggested, if funds become available, some sampling be conducted near the Ohio line and towards the international border to the north of Erie. These areas were not sampled 26 during this current study. Also, as laboratory capabilities improve And additional environmental contaminant concerns develop, some fish should be checked periodically to monitor present contaminant levels, as well as providing background information should contaminant concentration changes be noted. When future studies are undertaken, it might be well to consider the advisability of checking composites of only three to four species, for example, and obtaining a rather fast "turn around time." Larger studies may tend to prolong the time between when samples are collected and when they are finally presented to the public. Consideration should be given in comparing quantity of samples versus the timely reporting of a very few samples. For example, had we tested three or four samples and found questionable chlordane and/or lead results at the start of the program in 1987, we would have been more likely to have changed field and/or laboratory methods. Likewise, it was noted that the large lake trout (#0692264) bad elevated levels of cblordane and PCBs. Had we received the results in the Fall of 1987, we could'bave focused more attention towards collecting more lake trout In the Spring of 1988. As it was, all the fish collections were completed before we received the final laboratory results. Had this study been staged into three or four smaller reports, it is likely some of our sampling priorities would have been changed. 27 LABORATORY PROCEDURES Discussion of Cblordane Results Cblordane is a mixture of cbemicals ratber than a pure compound. According to the Handbook of Environmental Data on Organic Cbemicals, tecbnical cblordane consists of a mixture of many compounds. Tecbnical cblordane consists of approximately 19% alpba cblordane (cis cblordane), 24% gamma cblordane (trans cblordane) and 10% beptacblor epoxide, as well as otber compounds. 10 Because cblordane is not a single-peak compound, identifying it is not a simple matter. Some laboratories or agencies pick some of the isomers of cblordane and add up the quantities. The Federal Food and Drug Administration's metbodology addresses two metbods for determining cblordane (see analytical notes, Appendix I). One sums individual components of cblordane 0.02 ppm or bigber. The otber, if the pattern matcbes tecbnical cblordane, indicates the results sbould be quantitated against a tecbnical cblordane standard. In reviewing the quality assurance information, it was noted that in some cases DER's PCB results were lower at times than otber laboratories. However, cblordane was generally a factor of up to 4 (or more) times bigber than otber agencies were finding In the same sample, raising the question wbetber the actual levels for cblordane were being overestimated. 28 Upon reviewing the testing procedures, the laboratory indicated it reported technical cblordane rather than summing isomers. This led to the question, would it be expected tb.at "technical cblordane" as such would be found in fish. We speculated that the various components of technical cblordane would have different decay rates in the environment, different water solubilities, different attractions to silt and clay particles in the water, as well as biomagnification rates. It would not be expected that "technical cblordane" as such would be found in fish. Because of these questions, we contacted several agencies, including the Michigan State Department of Health. They agreed to help us look at the situation. They agreed to look at the other sides (right fillets) of the lake trout sample (CZM 6A), largemoutb bass sample (CZM 18) and the channel catfish sample (CZM 32) (see Appendix J). They found.less cblordane using different testing proc edures. Acting on this and other information, it was agreed to do further research into the matter. Both laboratories cooperated and later split samples with the U.S. Environmental Protection Agency (EPA). It appeared, following rather extensive testing, that the DER was overestimating the cblordane results (see excerpts from attachments, Bruzel memo, Appendix-K). However, it should be pointed out because of matters far beyond the scope of this report, that at this time we do not have a definite answer as to which metbod(s) will ultimately be acceptable. 29 Alan Bruzel points out in his memo dated July 7, 1989 that there will be a cblordane conference held in Missouri. At that time, hopefully, the issue of the correct, or acceptable metbodology.-will be resolved. Until then, the reader is cautioned that, although we can clearly see that the cblordane results of the DER are higher than others, it is difficult, if not impossible, to compare the answers. It seems likely that the quantification of isomers might be the preferred choice. If this is so, in the future we might see lower cblordane results than the DER is now reporting. 30 Discussion on Lead Results The FDA bas no action level for lead in fisb. However, the International Joint Commission (IJC) 1985 Annual Report, Revision of.'October 1986, suggests that long-term consumption for an adult sbould not exceed 2 mg/kg.11 This lead level included botb organic and inorganic lead. The IJC also suggested jurisdictions could adopt more stringent standards to protect their respective populations that migbt be exposed to otber sources of lead. They also suggested protecting sensitive subgroups, sucb as cbildren and women capable of bearing cbildren. It was empbasized that this proposed limit sbould be considered tentative. Wben the Erie County Department of Healtb received the lead results in the fisb, we noted two issues. One was that the duplicate samples in some cases were considerably different from the original sample.. The otber was that the results of some of the lead test results were considered bigb. Had the ratber bigb lead level results in our study been found in a particular area, we would bave suspected a localized source of lead. However, some of the results sbowed ratber bigb levels, even in the open lake. After the biologist from the Erie County Department of Healtb made several pbone calls to otber agencies, it became clear that the apparent bigb lead results migbt be related to a problem witb eitber field preparations and/or laboratory procedures. 31 A document prepared by Schmitt and Finger showed that there was generally a significant difference in lead results between fish prepared in the field as compared to fish prepared under ultra-clean laboratory conditions. As a result of the questions, a meeting was held in Harrisburg to discuss the issue of high lead results and other sample procedures. Regarding lead sampling, it was agreed at that meeting that the Harrisburg laboratory would attempt to secure a certified "fish tissue" lead standard and recheck their QA/QC procedures and recheck mathematics, etc. It was also agreed that the Erie County Department of Health would look into field preparations as a possible cause for the elevated lead levels. It was agreed additional testing would be done ,by the DER for lead. Four samples that were held frozen (the right fillets from previously tested fish) with high lead or relatively high lead'were chosen to be retested. They were as follows: CZM #1, yellow perch, 0692260: ori inal sample, left fillet retest, right fillets 0.798 ppm after rinsing by ECDH 0.317 ppm CZM #1, right fillets, allquat stored at laboratory from original sample, no additional cleanup by ECDH: orl inal sample 1989 retest 0.798 ppm 0.067 ppm duplicate 0.073 ppm 32 CZM #19, bluegills, 0692289: original sample 1989 retest 2.35 PPM 0.211 ppm CZM #22, sunfish, 0692278: original sample 1989 retest 1.18 ppm 0.293 ppm duplicate 0.249 ppm The frozen rigbt,fillets from the above fish bad been taken to a private residence for processing before shipping them to Harrisburg. The fish were thawed and then vigorously washed in the City of Erie drinking water that bad previously been tested and found to be of a currently acceptable lead level (<0.05 ppm). All scales and slime and any pieces of true rib bones inadvertently missed in the filleting process were washed and/or picked off the fillets. The fillets were rewasbed, drained and then carefully wrapped in clean aluminum foil, put in a food-grade plastic bag and refrozen. They were then shipped to DER's Harrisburg laboratory and retested, using slightly different methodology because of additional knowledge for testing for lead in fish flesh. Additional testing was also done on CZM #29, 0692287, which consisted of 5 walleyes (see Appendix G). As these walleye fillets were from large fish, it was decided to try to isolate possible sources of lead in the fish. The fillets were subdivided into 3 packages. All 3 packages (composites B, C and D) contained portions of all 5 fish. Hence, each package remained a composite of part of each of the 5 original fish. 33 Composite A was the result of the original left fillets. In composite B the skin and scales were removed from the 5 rigbf-balf fillets. The bones imbedded in the fish were removed by cutting a wedge of flesh down both sides of the bones to remove a piece of flesh with imbedded bones. Composite B was flesh only. All detectable skin and bones were removed. The fillets were then vigorously rinsed in the City of Erie drinking water under a kitchen spigot. Composite C consisted of the wedge of bones along with a small amount of flesh on either side of the bones. This composite was rinsed under tap water. Composite D consisted of the residue skin, "slime" and any loose scales and tips of bones removed during the skin removal process. Skin removal consisted of using a long, sharp knife blade and holding it down against the skin. The fish fillet was placed skin down on clean aluminum foil (dull side towards the fish). A sl iding downward cutting motion with the knife was then used to separate the flesh from the skin. The remaining skin and other parts were not rinsed, as it was suspected. Perhaps this fraction of the fish could,be much higher in lead. It can be seen (in Appendix G) that there Is a considerable difference between the original lead result and the followup samples. It did not appear that the lead, based on the walleye sampling, was concentrated in the skin or bones in this particular case; rather, the highest lead 34 concentrations were found in the flesb. It will be noted, bowever, that overall, the concentration of lead in the 1989 testing was approximately one-tentb the original amount reported in-1988. Based on the above information, the DER laboratory reviewed their findings and reported their findings in a memo dated August 24, 1989. The memo reads in part: . . .Your review of your fisb tissue data indicated elevated levels of lead in some samples. Because they were bigber than permitted levels you were very concerned, as we were. We reviewed all our work and felt it was correct. We agreed to run more samples that (sic) you requested. These samples were submitted to us, and along witb some of the original samples, we set up an analytical program in wbicb we altered some of the original preparation procedures. In the first assay run we dipsted one gram of wet fisb tissue. We felt this migbt Dot be a true representative sample so for the second run we used two grams of material. "The wet digestion was performed as normal. During the first analytical procedure we diluted the digested material one to five and performed the normal AA Furnace metbod for lead. In the study of the data from botb runs we discovered the dilution factor caused a bigber result that did Dot sbow up in the undiluted sample. The diluted sample reading was very close to our MDL, and this way (sic) be the reason for the elevated results. "We feel the results reported from our second run more nearly represent the lead levels in the fisb tissue samples. We do not bave enougb data to make a statement on otber metal parameters reported. We do feel that most of the reported data looked good and the QA results indicated that it was accurate. If any of the reported data indicated excessive levels or levels above the limits, of any of these parameters, we would be glad to analyze similar fiab tissue samples. "In conclusion, it is very difficult to get a truly representative sample of fisb tissue. We try very bard, but we are dealing witb sucb a small amount it makes it difficult. One gram of wet tissue willonly yield 0.3 grams of dried 35 material. We do have a new technique available. We can now freeze dry larger amounts of tissue, grind it, mix it and thus get a much more representative sample to analyze. We will be using this technique for our 89 samples . . . 1112 The grant that funded this study was to be terminated at the end of August 1989. Because of the lack of time, we were not able to have additional work done on lead or on the other metals. It appears that those "original" lead levels listed in this report, likely, are inflated values. As stated in the above memo, we do not know if the other metals are overestimates of the "true" values or not. The process of fish flesh analysis is an evolving process and almost certainly will be revised as more refinements are made In analytical techniques. 36 VII. General Discussion As can be seen in the species list, there is a good diversity of fish in the bay and nearby Lake Erie waters. We collected several game fish (terminal predators) from the bay. Only one gamefisb from the bay, a largemoutb bass, showed any contaminant (cblordane) above the FDA action level. This analysis result is in question because the 'Michigan Department of Public Health did not detect any in a sample from the same fish. It is interesting to note that there were no organics identified in the 35-incb muskellunge. The five-fisb composite of northern pike revealed only 0.02 ppm pp' DDE (average size of these pike was 30.1 inches). No PCB's or cblordane were found in these fish. There was no noticeable change in organic contaminates in yellow perch from the bay over a period covering October, February and June. All perch sampled were reported as none detected (ND) regarding the fourteen organics. At some future time it might beinteresting to sample perch roe (eggs) to see if organics accumulate in the roe because some fisherman do eat fried perch roe. The roe might be more contaufto-ted than the flesh. 04.tbe five species of fish that showed levels of organics over the FDA action level, four came from Presque Isle Bay. However, this may not be directly related to the bay water quality, but more to the type species sampled (and analytical methods used). Carp and catfish are known to be 37 high in at least one organic contaminant, as evidenced by the existing warning advising against eating these fish. There were no channel catfish or carp from the lake to compare tbe--results to. Gizzard shad are a particularly oily fish; it was suspected they might show some higher results than other species. Sampling confirmed this to be true. The largemoutb bass cblordane result was unexpected, but, as noted above, at this time there is serious doubt that the cblordane was over 0.3 ppm, especially when there were no PCB19 detected. (Generally, fish have more PCB's than cblordane in them.) The PCB/cblordane ratio alone casts doubt on the cblordane issue. One large (by Pennsylvania standards) lake trout (32 inches) revealed elevated levels of cblordane. This was confirmed by the Michigan Department of Public Health. However, even though they did find cblordane, Michigan found about four times less cblordane than did Pennsylvania. Michigan did find PCB's in this fish above FDA standards, while Pennsylvania found a lower amount. It should be noted this fish was unusual in that it bad a very large head for its body weight. It may not have been representative of another fish that same length. In fact, a fish of similar length with a more normal (i.e., heavier) weight quite likely could have different levels of contaminants. One smaller lake trout did also have cblordane slightly above the FDA action level. This was determined by the Pennsylvania laboratory and was not confirmed. 38 As noted earlier, as the tecbniques are refined, better data will be generated. It sbould be noted that this study is not unique in coming up witb different results compared to otber agencies. In the 1987 Report on Great Lakes Water Quality, it was pointed out, "Differences in absolute values between agency programs may result from differing analytical metbods It Is obvious that more uniform protocols be developed so that fiob samples can be more easily compared and verified. In looking at the reported PCB contaminant levels, it was found that our study (witb the exception of fresbwater drum "sbeepsbead") mirrors information put out by the Ontario Ministry of the Environment.14 In their 1987 Guide to Eating Ontario Sport Fisb they sbow an inverted pyramid. At the top of the pyramid are Lake Trout (most PCB's) and at the bottom are nortbern pike and fresbwater drum (least PCB1s). In ascending order are: level one - nortbern pike and fresbwater drum; level two cobo salmon, rainbow trout and cbinook salmon; level tbree brown bullbeads and wbite bass; level four - cbannel catfisb; level five - brown trout; level six - carp; and the top, level seven - lake trout. In looking at PCB levels (except the drum) in our study u&tng the larger lake trout and larger of the two rainbow (trout composites), we find that the progression is identical witb Ontario's relative levels (see Appendix F, Table II). If this "condition" reflects actual lake-wide 39 conditions in Pennsylvania, because there are current advisories on cbannel catfisb and carp, it seems more attention needs, to be focused on brown trout-and lake trout. (Altboogb outside the scope of this report, it bas been agreed to focus attention in 1989 on collecting and analyzing additional Lake trout samples. Large brown trout are not particularly common; consequently, getting enougb to sample is more difficult. However, if possible, it migbt be well to also look at tbem.) 40 VIII. Recommendations It is recommended that, in future studies, not only should internal quality control/quality assurance be practiced but, as was done in this study, an outside laboratory or laboratories should also analyze the fish as early in the study as is practical. This would better insure that any di screpancies in test results could be resolved prior to testing all the fish. The issue of lead analysis should be followed up as well as looking at and verifying other metals besides lead. TIbe DER laboratory is committed to further . investigating the cblordane issue. This should help resolve this matter. There remains some question as to the PCB results. While not as an apparent difference, as was noted in the cblordane issue, there is evidence to indicate that the Pennsylvania DER PCB results are not always consistent with outside references. It is also evident that there is a question as to whether the PCB's identified are 1260 or 1254. It is recommended that procedures be reviewed regarding PCB interpretation, as is currently being done with cblordane. It is also recommended that all Greatlake states and appropriate federal agencies come to a consensus of opinion as to proper sampling techniques, field preparation and laboratory protocol, in order to better assure reliable, valid results. 41 It is recommended-tbat lake trout, particularly larger size fish, be evaluated, and if a reasonable amount of larger brown trout can be collected that they also be checked for-organics. It is recommended that improved field protocol be developed in cleaning the scales and slime from fish prior to their being homogenized (ground up) in the laboratory- Possibly de-ionized water in a suitable, contaminant-free container could be employed to rinse the fish fillets. This would better duplicate what fishermen actually do. Presently, if scales are not rinsed from the sample before freezing, they are homogenized into the sample, as it is easier to grind up frozen fish than soft, tbawed-out fish. It is recommended, if at all possible, fish not be processed in the field in the winter. This is due to problems associated with frozen knife blades, cutting fingers, etc. When future studies are conducted, it would be better to run fewer samples during phases of a study so that the results could be made public in a more timely fashion or even change direction in sampling priorities as data comes in. In other words, if the lab and field protocol was better established and in a pre-printed format, it would be relatively easy to prepare a report by just adding species, locations and test results. Perhaps the time from sampling to usable results could be reduced from months or years to perhaps weeks or months. 42 Ix. Summary This study provided meaningful insigbt into the complexities of fisb contaminant studies, and as a result there will likely be cbanges in procedures botb in the field and in the laboratory. Bay fisb that were sampled, including yellow percb, sunfisb, bluegills and black crappies, did not bave detectable levels of the organic contaminants. This provides some degree of reference for potential consumers, if they cboose to apply the FDA action level. One fisb, the large lake trout, bad levels of cblordane above the FDA action level, and this was confirmed by more than one laboratory. Four otber species of fisb (carp, cbannel catfisb, bass and gizzard sbad) bad cblordane levels exceeding the FDA action limit. However, wben Micbigan cbecked the same catfisb and largemoutb bass they found a cblordane level below the FDA action level on the catfisb and did not detect any at all in the bass. Tberefore, if one were to make allowances for metbodology, that is, looking for a tecbnical cblordane pattern versus adding up components of tecbnical cblordane, the FDA limits may or may not bave been exceeded (otber than the lake trout). Some otber fisb, as determined by DER, bad bigber than expected cblordane results approacbing the FDA action level. Again, the cblordane results are in question. None of the DER PCB results exceeded the FDA action level. However, Micbigan did find elevated levels of PCB in the large lake trout. Because of the differences in quality 43 assurance in test results among laboratories, the DER's PCB results sbould not necessarily be taken as absolute values. They may be bigher or lower than reported values. Lead results likely are mucb lower than this document's "first-run" reported values. Otber metal results sbould be looked at in the future by the laboratory to see if only the lead results were bigb or if the tecbniques used also inflated (or underestimated) otber reported metal results. It sbould be noted that in all fisb tested, all mercury results were not only below the FDA action level but were also all below 0.5 pp m, wbicb is a standard used by some otber agencies. Because of the variables in testing procedures and consequently in the analysis results, it is recommended that Pennsylvania, along witb all Great Lake states and otber appropriate governmental agencies, adopt strict uniform testing procedures. It is difficult to evaluate risk assessments or determine the validity of issuing a consumption advisory (or not issuing a consumption advisory) based on only a few fisb, wben, in fact, laboratory results among agencies are not always consistent. Wbile the results of this study sbow the presence of some contamination in fisb, it is somewbat comforting to note that at least some fisb species sbowed all test results below the laboratories detection levels for the fourteen organics. This study also reaffirms the fact that our area does not appear to be as contaminated as some otber waters 44 in the Great Lakes area. Hopefully, with increased emphasis on cleaning up the environment, someday all fish will be found to be below even lower detection levels, and for all practical purposes will then all be termed "safe." At least for now, however, prudence and caution should be exercised until more is known about the subject. 45 I I I I I I I I A P P E N D I X A I FDA Flab "Action Levels" I I I I I I I - I I - I United States Department of Healtb and Huinan Services Public Healtb Service Food and Drug Administration FDA "Action Levels" for Fisb Aldrin and Dieldrin (edible portion) 0.3 ppm DDT, TDE and DDE (edible portion) 5.0 ppm Endrin (edible portion) 0.3 ppm Heptacblor and Heptacblor Epoxide (edible portion) 0.3 ppm Mercury (metbyl mercury in edible portion) 1.0 ppm Mirex (edible portion) 0.1 ppm PCB 2.0 ppm Toxapbene (edible portion) 5.0 ppm I I I I I I I I A P P E N D I X B I Sampling Locations I I I I I I I I I I Locations of Study Areas Study Area A Presque Isle Bay (soutb-central area of bay) Nortb Latitude: 42007'36" and Nortb Latitude: 42007137" West Longitude: 8000,7'09" West Longitude: 80006'48" Study Area B Presque Isle State Park (lagoons - backwater off Presque Isle) Nor.tb Latitude: 42009'33" West Longitude: 8000.6'00" Study Area C Presque Isle Bay (western end of bay) No rtb Latitude: 42007'03" Nortb Latitude: 4200651" and West Longitude: 80008'28" West Longitude: 80009'00" Study Area D Trout Run is a small tributar to Lake Erie, about 10 miles west from Erie, near PA Rt. U Study Area E off Sbades Beacb (35 - 40 ft. water) Nortb Latitude: 42011'58 Nortb Latitude: 42014'19" and West Longitude: 79057'53" West Longitude: 79057'39" Note: Nets for yellow percb sample #CZM291 were set off Sbades Beacb in about 64 ft. water. I I I I I I I I A P P E N D I X C @ I Species Sampled Common and Scientific Names for Fisb I I I I I I I I -1 I Species Sampled Common Name Scientific Name Bass (largemoutb) Micropterus salmoides Bass (smallmoutb) Micropterus dolomieui Bluegill Lepomis macrocbirus Bullbead (brown) Ictalurus nebulosus Carp Cyprinus carpio Catfisb (cbannel) Ictalurus punctatus Crappie (black) Pomoxis nigromaculatus Gizzard Sbad Dorosoma cepedianum Muskellunge Esox masquinongy Percb (wbite) Morone americana Percb (yellow) Perca flavescens Pike (nortbern) Esox lucius Sbeepsbead (fresbwater drum) Aplodinotus grunniens Sucker (wbite) Catostomus commersoni Sunfisb Lepomis gibbosus Trout (lake) Salvelinus namaycusb Trout (rainbow) Oncorbyncbus mykiss (formerly Salmo gairdneri) Walleye Stizostedion vitreum vitreum I I I I I I I . . I I A P P E N D I X D I Field Collection and Preparation Protocols I I I I I I I I I I Field Protocol Fish Tissue Sampling 1. Collect fish (Electrofishing, Seine, Gill Net, Rotenone, Angling, other) taking care not to contaminate specimens with gasoline, motor oil, sediment, or soil. Record method on Field Data Sheet. 2. Measure total length in MM of each specimeit in sample. 'Record on Field Data Sheet. Weigh each specimen in sample to nearest gram. Record on Field Data Sheet. 3. Note general condition, tumors, lesions. Record on Field Data Sheet as needed. 4. Prepare sample: A. Whole Fish - Wrap composite sample (or individual Ti-sb IT-necessary, for specific study) in clean, commercial (restaurant) grade aluminum foil allowing only the dull foil surface to contact fish tissue. Indicate sample type on Field Data Sheet. B. Fillets - Rinse clean fillet knife with purif ied hexane labeled as suitable for pesticide residue analysis. 1. Inland Waters - Remove entire edible portion (fillet) from both sides of each specimen and remove skin. Wrap composite sample. (or individual f ish samples if necessary) in clean aluminum foil (dull side in contact with fish). Indicate sample type on Field Data Sheet. 2. Lake Erie - Follow above procedure, but do not remove skin. Scale.each specimen prior to filleting and leave skin on fillet. This complies with our agreement with the EPA Great Lakes National Program Office and the other Great Lakes states to provide uniform methodologies for Great Lakes tissue samples. 5. Clearly label each sample with the station number or water body name and location, date, time, and collector. number (if necessary). 6. Place foil wrapped sample in a food grade protective plastic bag and freeze sample immediately on dry ice, if possible. 7. Be sure Field Data Sheet has been completed. FIEL DATA SEEET Tisell Sampling Commonvealth of Pennsylvania Station Water Body: Date: Location:- County: Municipality: Collector: Agency: Co1l.# Method: Electrofishing Seine Gill Net Rotenone Angling Other Reason: SPECIES TL_MM WT_G *CONDITION 2. 3. 4. S. 6. 7. 9. 10. *Note tumors, lesions, & general condition (if needed). Tissue Type:- Whole Fish )'Skinless Fillet Skin-on Fillet Scaled (Y or N) Blood Organ Other Co=nents (water/weather conditions,man-hours expended,problems etc.) FILLETS USED BY ERIE COUNTY DEPARTMENT OF HEALTH SKIN ON FILLET FOR FISH-WITH ANTERIOR PELVIC FINS (SCALES REMOVED) EG YELLOW PERCH BASS NOTE SOME SMALL IM13EDDED BONES REMAIN IN FILLET BUT TRUE RIBS ARE REMOVED *SKIN ON FILLET FOR FISH WITH P TERIOR PELVIC FINS (SCALES REMOVED) % EG NORTHERN PIKE RAINBOW TROUT NOTE SOME SMALL LAKE TROUT BONES REMAIN IN FILLET TRUE RIBS ARE REMOVED NOTE PELVIC FINS REMOVED SKIN ON.FILLET FOR MEMBERS OF THE CATFISH FAMILY oat, EG CATFISH BULLHEAD NOTE PELVIC FINS REMOVED I I. I I I I I I A P P E N D I X E I Study Areas, Lengtbs and Weigbts of Fisb, I Collection Netbods and Dates I I I I I I I I I Species, collection location, length and weight of Presque Isle Bay fish submitted for chemical analysis S T U D Y A R E A A Date Method of Length Weight Sample Collected Species Collection in. mm lb./oz. .--grams 0692266 10/20/87 Walleye Gill Net 21 533 3 lb. 8 oz. 1,589 27 686 6 lb. 10 oz. 3,008 22 559 3 lb. 8 oz. 1,589 20 508 2 lb. 15 oz. 1,334 0692267 10/20/87 Northern Gill Net 34 864 7 lb. 3 oz. 3,263 P,ike 32 813 7 lb. 3 oz. 3,263 25.5 648 3 lb. 1,362 28 711 4 lb. 1 oz. 1,844 31 787 6 lb. 2,724 0692268 10/20/87 Yellow Perch Gill Net 8.5 216 4 oz. 113 11.5 292 10 oz. 283 8.5 216 4 oz. 113 7.5 191 3 oz. 85 9 229 4.5 oz. 127 0692269 10/20/87 Black Gill Net 9 229 6 oz. 170 Crappie 8 203 5 oz. 142 7.75@ 197 4 oz. 113 8 203 5 oz. 142 0692276 5/4/88 White Sucker Hoop Net 18.5 470 2 lb. 1 oz. 936 16.45 418 1 lb. 8 oz. 681 14.2 361 15 oz. 425 16.4 417 1 lb. 12 oz. 795 15.27 388 1 lb. 3 oz. 539 Species, collection location, length and weight of Presque Isle Bay'fisb submitted for chemical analysis S T U D Y A R E A A Date Method of Length Weight Sample Collected Species Collection in. mm lb./oz. grams 0692277 5/4/88 Brown Hoop Net 13.46 342 1 lb. 4 oz. 562 Bullbead 12.8 325 15 oz. 430 12.7 323 15.6 oz. 442 12.2 310 13.7 oz. 390 12.2 310 14 oz. 400 0692279 5/12/88 Yellow Perch Gill Net 8.66 220 2.8 oz. 80 9.25 235 3.6 oz. 103 9.84 250 4.9 oz. 140 9 230 3.5 oz. 100 9 230 2.8 oz. 80 0692280 5/12/88 Smallmoutb Gill Net 17.1 435 2 lb. 10 oz. 1,192 Bass 17.5 445 2 lb. 7 oz. 1,107 18.1 460 3 lb. 2 oz. 1,419 16.5 420 2 lb. 5 oz. 1,050 0692281 5/12/88 Muskellunge Gill Net 35.2 895 10 lb. 12 oz. 4,881 0692282 5/12/88 White Perch Gill Net 9.8 250 9 oz. 255 11.4 290 15 oz. 425 9.4 240 8 oz. 227 8.66 220 7 oz. 198 8.8 225 6 oz. 170 0692283 5/12/88 Gizzard Shad Gill Net 15.74 400 1 lb. 4 oz. 568 15.74 400 1 lb. 4 oz. 568 17.7 450 1 lb. 13 oz. 823 Species, collection location, lengtb and weigbt of Presque Isle Bay fisb submitted for cbemical analysis S T U D Y A R E A A Date Metbod of Lengtb Weigbt Sample Collected Species Collection in. mm lb./oz. grams" 0692284 6/22188 Yellow Percb Gill Net 9.2 233 5.2 oz. 148 6/30/89 9.5 242 6.2 oz. 175 9.9 251 5.6 oz. 160 0692285 6/30/88 Cbannel Gill Net 18.8 478 2 lb. 8 oz. 1,135 Catfisb 22.2 564 2 lb. 7 oz. 1,107 21.45 545 3 lb. 8 oz. 1,589 0692286 6/22/88 Carp Gill Net 17.7 450 2 lb. 8 oz. 1,135 20.08 510 3 lb. 4.5 oz. 1,490 0692287 5/12/88 Walleye Gill Net 25.2 640 5 lb. 13 oz. 2,639 24.8 630 5 lb. 15 oz. 2,696 20.5 520 3 lb. 9 oz. 1,617 23.8 605 4 lb. 15 oz. 2,242 21.25 540 3 lb. 13 oz. 1,731 0692288 5/4/88 Largemoutb Hoop Net 12.6 320 14 oz. 398 Bass 0692289 5/4/88 Bluegill Hoop Net 7.7 195 5 oz. 138 7.4 188 4 oz. 122 8.3 210 7 oz. 200 7.2 184 4 oz. 110 Species, collection location, lengtb and weigbt of Presque Isle Bay fisb submitted for cbemical analysis S T U D Y A R E A B Date Metbod of Lengtb Weigbt. Sample Collected Species Collection in. mm lb./oz. grams 0692278 5/4/88 Sunfisb Hoop Net 7.24 184 4.58 oz. 130 6.89 175 4.51 oz. 128 6.89 175 4.44 oz. 126 7.28 185 5.2 oz. 148 7.04 179 4.23 oz. 120 Species, collection location, lengtb and weigbt of Presque Isle Bay fisb submitted.for cbemical analysis S T U D T AREA* C Date Metbod of Lengtb Weigbt Sample Collected Species Collection In. Imm lb./oz. grams 0692272 3/2/88 Bluegill Angling .7.2 183 4.2 oz, 118 7.4 188 5.2 OZ. 148 .6.9 175 3.8 Oz. 108 7.5 191 4.6 oz. 130 7.8 198 5.6 oz. 160 0692273 2/17/88 Yellow Percb Angling 9.25 235 5.4 OZ. 154 7.1 180 2.5 oz. 72 7.5 191 2.7 oz. 78 7.25 184 2.25 oz. 64 7.25 184 2.3 oz. 65 Species, collection location, length and weight of Trout Run (Lake Erie) fish submitted for chemical analysis S T U D Y A R E A D Date Method of LeDgtb Weight Sample Collected Species Collection in. mm lb-./oz. grams 0692270 11/19/87 Rainbow Netted 24.3 617 6 lb. 3 oz. 2,810 Trout 21.93 557 4 lb. 1,820 23.03 585 4 lb. 3 oz. 1,900 22.64 575 4 lb. 12 oz. 2,160 23.07 586 5 lb. 2,280 0692271 11/19/87 Rainbow Netted 17.5 445 1 lb. 15 oz. 900 Trout 15.4 392 1 lb. 7 oz. 660 15.7 400 1 lb. 11 oz. 780 16.3 415 1 1b. 9 oz. 720 18.2 463 2 lb. 3 oz. 1,000 Species, collection location, lengtb and weigbt of Lake Erie fisb submitted for cbemical analysis S T U D Y A R E A E (35' depth) Date Method of Length Weight Sample Collected Species Collection in. Mm lb./o'z. grams 0692260 10/20/87 Yellow Porcb Gill get 11.5 292 10 oz. 283 11.25 286 9 oz. 255 10.0 254 oz. 170 11.0 279 .'9 oz. 255 11. 0 279 9 oz. 255 0692261 10/20/87 Walleye Gill Net 21.5 699 7 lb. 9 OZ. 3,433 26.0 660 6 lb. 10 oz. 3,008 0692262 10/20/87 Smallmoutb Gill Net 20.0 508 4 lb. 12 oz. 2,157 Bass 21-75 :5 5 2 5 lb. 2 oz. 2,327 18.0 3 lb. 3 oz. 1,447 19.0 483 3 lb-: 3 oz. 1,447 19.0 483 3 lb. 1,362 0692263. 10/20/87 Sbeepsbead Gill: Net 18.0 @.457 2 lb. 13 oz. 1,277 0692264 10/20/87 Lake T.Fout Gill Net 32.75 .832 10 lb. 9 oz. 4,795 0692265 10/20/87 Lake Trout Gill Net 23.5 597 4 lb. 6 oz. 1,986 Species, collection location, length and weight of Lake Erie fish submitted for chemical analysis S T U D Y A. R E A E (64' deptb) Date Method of LeDgtb Weight Sample Collected Species Collection in. mm lb. /O'z. grams 0692291 7/8/88 Yellow Perch Gill Net 10.1 257 6.7 oz. 190 9.8 249 7.04 oz. 200 9.7 246 5.46 oz. 155 9.2 234 5.46 oz. 155 9.2 234 5.64 oz. 160 I .1 I I I I I I A P P E N D I X F I Sample Results I Organic and Inorganic, I Percent Moisture and Percent Lipids. I I I I I I I I Concentrations of elemental and organic contaminants in edible fillet samples collected from Presque Isle Bay area (results in ppm wet weight*) S T U D Y A R E A A Sample Number 0692 and Species** 266 267 2-68 -269 276 Walleye Northern Yellow Black White Pike Perch Crappie Suckers %Lipid 1.36 .81 .21 .58 ' 32 7.Moisture 78.5 78.4 75.8 77.6 75.6 Elements Lead*** 0.623 0.599 0.530 0.478 0.162 Cadmium <0.013 0.015 0.014 <0.013 0.017 Chromium 0.187 0.187 0.202 0.214 0.212 Arsenic <0.5 <0.5 <0.5 <0.5 <0.5 Copper <0.5 <0.5 <0.5 <0.5 <0.5 Selenium 0.277 0.294 0.437 0.327 0.427 Antimony <1.25 <1.25 <1.25 <1.25 <1.25 Silver <0.05 <0.05 <0.05 <0.05 <0.05 Barium <0.25 <0.25 <0.25 <0.25 0.299 Beryllium <0.025 <0.025 <0.025 <0.025 <0.025 Mercury 0.253 <0.1 <0.1 <0.1 <0.1 Organics PCB's .13 ND ND ND ND Cblordane .061 ND ND ND ND Toxapbene ND ND ND ND ND pp'DDE .021 .020 ND ND ND pp'-DDD ND ND ND ND ND A '-DDT ND ND ND ND ND Ydrin ND ND ND ND ND Dieldrin ND ND ND ND ND Endrin ND ND ND ND ND Lindane ND ND ND ND ND Heptacblor ND ND ND ND ND Heptacblor epoxide ND ND ND ND ND Metboxycblor ND ND ND ND ND Mirex ND ND ND ND ND ND not detected (see Appendix J for information on detection limits) ppm parts per million - reported on lab sheets as migrograms/gram ( = milligrams/kilogram) *wet weight = as received basis, i.e., what a consumer would buy in a market **All scaled species of fish consisted of skin left on but scales removed fillets (skin was also left on the catfish family samples) ***See special section on lead analysis difficulties, page 39 and Appendix G Concentrations of elemental and organic contaminants in edible fillet samples collected from Presque Isle Bay area (results in ppm wet weight) S T U D Y A R E A A Sample Number 0692 and Species 277 280 281 282 Brown Small Muskel- White Bullhead Mouth lunge Perch (Note: skin-on fillets) %Lipid 2.19 2.09 1.09 7.37 '/.Moisture 78.4 77.4 75.5 71.9 Elements Lead 0.525 0.552 0.176 1.25 Cadmium 0.029 0.034 0.014 0.038 Chromium 0.175 <0.125 0.264 0.663 Arsenic <0.5 <0.5 <0.5 <0.5 Copper <0.5 <0.5 <0.5 1.7 Selenium 0.492 0.610 0.334 0.630 Antimony <1.25 <1.25 <1.25- <1.25 Silver <0.05 <0.05 <0.05 <0.05 Barium 0.35 0.274 <0.25 0.375 Beryllium <0.025 <0.025 <0.025 <0.025 Mercury <0.1 0.125 0.135 0.1 Organics PCB's .690 .270 ND Chlordane .230 ND ND .170 Toxapbene ND ND ND ND pp'-DDE .070 .071 ND o52 pp'-DDD .060 ND ND ND pp'-DDT ND ND ND ND Aldrin ND ND ND ND Dieldrin ND ND ND ** Endrin ND ND ND ND Lindane ND ND ND ND Heptacb1or ND ND ND ND Heptachlor epoxide ND ND ND ND Metboxychlor ND ND ND ND Mirex ND ND ND ND ND = not detected *Possible trace, estimated at .170 **possible trace, estimated at .013 Concentrations of elemental and organic contaminants in edible fillet samples collected from Presque Isle Bay area (results in ppm wet weigbt) S T U D Y A R E A A Sample Number 0692 and Species 283 284 285* 286 287 Gizzard Yellow Cbannel Carp Walleye Sbad Percb Catfisb (Note: skin on fillets) %Lipid 12.67 0.24 9.73 8.33 5.86 %Moisture 71.8 76.7 69.4 75.9 75.2 Elements Lead 0.716 0.797 1.69 0.853 1.2 Cadmium <0.013 0.034 0.034 0.035 0.019 Cbromium 0.272 0.511 0.717 0.702 0.451 Arsenic <0.5 <0.5 <0.5 <0.5 <0.5 Copper 1.11 0.847 <0.5 <0.5 0.801 Selenium 0.568 0.272 0.269 0.439 0.636 Antimony <1.25 <1.25 <1.@25 <1.25 <1.25 Silver <0.05 <0.05 <0.05 <0.05 <0.05 Barium <0.25 0.349 0.553 0.251 <0.25 Beryllium <0.025 <0.025 <0.025 <0.025 <0.025 Mercury <0.1 0.108 0.18 0.205 <0.1 Organics PCB's 1.2 ND .920 1.000 .190 Cblordane 0.53 ND .720 .560 .220 Toxapbene ND ND ND ND ND pp'DDE 0.21 ND .240 .200 .070 pp'-DDD 0.069 ND .130 .040 ND pp'-DDT ND ND ND ND ND Aldrin ND ND ND ND ND Dieldrin ND ND ND ND ND Endrin ND ND ND ND ND Lindane ND ND ND ND ND Heptacblor ND ND ND ND ND Heptacblor epoxide ND ND ND ND ND Metboxycblor ND ND ND ND ND Mirex ND ND ND ND ND ND = not detected *See results - Micbigan CZM 32, duplicate analysis, rigbt fillets of same fisb Concentrations of elemental and organic contaminants in edible fillet samples collected from Presque Isle Bay area (results in ppm wet weigbt) S T U D Y A R E A A Sample Number 0692 and Species 288* 289 Large Bluegill Moutb Bass %Lipid 1.01 '/.Moisture 78.8 79.5 Elements Lead 1. 04 2.35 Cadmium 0.024 <0.013 Cbromium 0.401 0.574 Arsenic <0.5 <0.5 Copper 0.652 1.2 Selenium 0.391 0.432 Antimony <1.25 <1.25 Silver <0.05 <0.05 Barium 0.652 1.20 Beryllium <0.025 <0.025 Mercury <0.1 0.141 Organics PCB's ND Cblordane .320 Toxapbene ND pp'DDE .030 pp -DDD ND pp'-DDT ND Aldrin ND Dieldrin ND Endrin ND Lindane ND Heptacblor ND Heptacblor epoxide ND Metboxycblor ND Mirex ND ND = not detected *See results Micbigan CZM 18, duplicate analysis, rigbt fillet of same fisb Concentrations of elemental and organic contaminants in edible fillet samples collected from Presque Isle Bay area (results in ppm wet weigbt) S T U D Y A R E A B Sample Number 0692 and Species 278 Sunfisb %Lipid .32 '/.Moisture 76.0 Elements Lead 1.18 Cadmium 0.031 Cbromium 0.125 Arsenic <0.5 Copper <0.5 Selenium 0.387 Antimony <1.25 Silver <0.05 Barium <0.25 Beryllium <0.025 Mercury 0.170 Organics PCB's ND Cblordane ND Toxapbene ND pp'DDE ND pp'-DDD ND pp'-DDT ND Aldrin ND Dieldrin ND Endrin ND Lindane ND Heptacblor ND Heptacblor epoxide ND Metboxycblor ND Mirex ND ND not detected Concentrations of elemental and organic contaminants in edible fillet samples collected from Presque Isle Bay area (results in ppm wet weigbt) S T U D Y A R E A C Sample Number 0692 and Species 272 273 Bluegill Yellow Percb %Lipid .2 .16 %Moisture 79.8 75.9 Elements Lead 0.545 0.627 Cadmium 0.026 0.028 Cbromium 0.396 0.589 Arsenic <0.5 <0.5 Copper <0.5 3.66 Selenium 0.277 0.366 Antimony <1.25 <1.25 Silver <0.05 <0.05 Barium <0.25 <0.25 Beryllium <0.025 <0.025 Mercury 0.108 <0.1 Organics PCB's ND ND Cblordane ND ND Toxapbene ND ND pp'DDE ND ND pp'-DDD ND ND pp'-DDT ND ND Aldrin ND ND Dieldrin ND ND Endrin ND ND Lindane ND ND Heptacblor ND ND Heptacblor epoxide ND ND Metboxycblor ND ND Mirex ND ND ND = not detected Concentrations of elemental and organic contaminants in edible fillet samples collected from Trout Run, a tributary of Lake Erie (results in ppm. wet weight) S T U D. Y A R E A D Sample Wumber 0602 and Species 270 271 Ra inbow Rainbow Trout* Trout %Llpid 7.56 6.7 */.Moisture 68.6 80.6 Elements Lead 0.663 0.843 Cadmium 0.013 0.0201 Chromium 0.251 0. _,@97 Arsenic <0. 5 <0. 5 Copper <0. 5 <0. 5 Se 1 en ium 0.36(,, 0.1-89 Antimony <1.25 <1.25 Silver <0.05 <0.05 Bar j um <0 . 2@ 5 <0.25 Beryllium not -run <0.025 Mercury <0.1 0.178 Organics PCBIs 0.44 0.24 Chlordane 0.28 0.18 Toxaphene ND 14D pp'DDE 0.095 0.084 pp'-DDD ND 0.018 pp'-DDT' ND ND Aldrin ND ND Dieldrin ND Endrin ND ND Lindane ND ND Heptachlor N@D ND Heptachlor epoxide ND ND Metboxycblor ND -'ND Mirex ND ND ND not detected *possible trace, estimated at .012 Concentrations of elemental and organic contamiribnts in edible fillet samples collected from Lake Erie off Shades Beach (results in ppm wet weight) S T U D Y A R E A E 'Sample Number 0692 znrd Species 260 261 262 @263 Yellow Walleye ISMall Sbeeps- Perch Mouth head Bass %Lipid 0.15 5.39 4.16 11.88 /.Moisture 78.7 74.8 75.5 68.5 Elements 'ead 0.758 0.410 <0.125 0.785 Cadmium 0.015 0.019 <0.013 0.0137 Chromium 0.386 0.348 0.226 0.224 Arsenic <0.5 <0.5 <0.5 <0.5 Copper <0.5 <0.5 <0.5 <0.5 Selenium <0.25 0.463 0.819 0.377 Art imoijy <1.25 <1.25 <1.25 <1.25 Silver <0.05 <0.05 <0.05 <0.05 Barium <0.25 <0.25 <0.25 <0.25 Beryllium <0.025 <0.025 <0.025 <0.025 Mercury 0.197 0.171 0.306 0.121 Organics PCB's ND 0.21 0.35 0.46 Cb1ordane ND 0.22 0.16 0.15 Toxaphene ND ND ND ND pp'DDE ND 0.072 0.067 0.058 pp'-DDD ND 0.019 0.016 pp'-DDT ND ND ND ND Aldrin ND ND ND ND Dieldrin ND trace ND Endrin ND ND ND ND Lindane ND ND ND ND Heptacb1or ND ND ND ND Heptacblor epoxide ND ND ND INTD Metboxycblor ND ND ND- ND Mirex ND ND Xl@ ND ND = not detected *possible trace, estimated at .015 **possible trace, estimated at .007 Concentrations of elemental and organic contaminants in edible fillet samples collected from Lake Erie off Shades Beach (results in ppm wet weight) S T U D Y A R E A E Sample Number 0692 and Species 264* 264 265 Lake Lake Lake Trout Trout Trout (First (Second Analysis)__Analysis) %Lipid 10.91 9.63 11.55 %Moisture 68.0 Elements Lead 0.690 Cadmium <0.0125 Chromium 0.188 Arsenic <0.5 Copper <0.5 Selenium 0.296 Antimony <1.25 Silver <0.05 Barium <0.25 Beryllium <0.025 Mercury <0.1 Organics PCB's 1.2 1.3 0.24 Chlordane 1.1 0.0 0.32 Toxaphene ND ND ND pp'DDE 0.39 0.4 0.09 pp'-DDD 0.11 0.12 ND pp'-DDT ND ND ND Aldrin ND ND ND Dieldrin ND ND Endrin ND ND ND Lindane ND ND ND Heptacblor ND ND ND Heptacblor epoxide ND ND ND Metboxycblor ND ND ND Mirex ND ND ND BHC ND = not detected *See results - Michigan CZM 6A, duplicate analysis, right fillet of same fish **estimated at 0.010 ***estimated at 0.012 Concentrations of elemental and organic contaminants in edible fillet samples collected from Lake Erie off Sbades Beacb (results in ppm wet weigbt) S T U D Y A R E A E Sample Number 0602 and Species 291 Yellow Percb %Lipid 0.14 7.Moisture 77.5 Elements Lead 0.423 Cadmium 0.024 Cbromium 0.361 Arsenic <0.5 Copper 0.373 Selenium 0.632 Antimony <1.25 Silver <0.05 Barium 0.373 Beryllium <0.025 Mercury 0.112 Organics PCBIS ND Cblordane ND Toxapbene ND pp'DDE ND pp'-DDD ND pp'-DDT ND Aldrin ND Dieldrin ND Endrin ND Lindane ND Heptacblor ND Heptachlor epoxide ND Metboxycblor Mirex ND ND not detected Table I Fisb fillets in descending order of % lipids ("fat") actual percentage of individual fisb or study averages, wbere more than one fisb of the same species was encountered from combined lake and/or bay samples Gizzard Sbad 12.67* Sbeepsbead (Fresbwater Drum) 11.88 Lake Trout 11.23* Cbannel Catfisb 9.73* Carp 8.33* Wbite Percb 7.37 Rainbow Trout 7.13 Walleye 4.20 Smallmoutb Bass 4.16 Muskellunge 1.09 Largemoutb Bass 1.01* Nortbern Pike 0.81 Black Crappie (Calico Bass) 0.58 Sunfisb (Pumpkinseed) 0.32 Wbite Sucker 0.32 Bluegill 0.2 Yellow Percb 0.18 *Fisb fillets exceeding FDA "Tecbnical Cblordane" action level as determined by the Pennsylvania DER (please read special discussion on laboratory procedures on quality assurance and interlaboratory differences in results) Table II Concentrations of PCB's found by the Pennsylvania Department of Environmental Resources. This study mirrors the Ontario Ministry of the Environment "Relative Levels of PCB's in Lake Ontario Fish Species." PCB's Lake Trout (large) 1.3 ppm Carp 1.0 ppm Brown Trout no information on Pa. fish Channel Catfish 0.92 ppm Brown Bullhead 0.69 ppm Rainbow Trout 0.44 ppm (large composite) Northern Pike none detected Sheepshead 0.46* (Freshwater Drum) *Freshwater drum is the exception of the relative Ontario PCB concentrations. It may be due to the fact that only one fish was sampled in Pennsylvania and that fish may or may not reflect the average PCB concentrations. I -1 I I I I I I A P P E N D I X G I Lead Results - Walleye Sections I I - I I . I I I I I I LEAD RESULTS WALLEYE CZK329 FIVE FISH COMPOSITE A ORIGINAL LEFT SIDE FILLETS SCALED SKIN ON - NOT RINSED B RIGHT SIDE FILLETS SKIN REMOVED - WEDGE CUT OUT REMOVING SOME FLESH & ALL IMBEDDED BONES - RIBSED IN TAP WATER C WEDGE REMOVED FROM ABOVE FLESH & BONES - NO SKIN-RINSED IN TAP WATER RESIDUAL SKIN FROM THE 5 RI GHT FILLETS D SOME "TIPS" FAOM BONES LOOSE SCALES & "SLIME" NOT RINSED ORIGINAL 1988 RESULTS RETEST IN 1989 A B (FLESH) C (WEDGE) D (SKIN) 1.20 PPM 0.106 PPM 0.091 PPM 0.037 PPM DUPLICATE 1.75 PPM I I I I I I I I A P P E N D I X H 1, Preparing PCB Contaminated Fish I for Human Consumption I I I I I I I I I COMMONWEALTH OF PENNSYLVANIA DEPARTMENT OF ENVIRONMENTAL RESOURCES PREPARING PCB CONTAMINATED FISH FOR HUMAN CONSUMPTION Polychlorinated biphenyls (PCB's) are a group of very stable industrial chemicals used since 1929. PCB's were discovered in fish in the United States in 1967, and domestic production was banned in 1977. PCB's are, however, ,still used in closed electrical equipment. Because of their stability, PCB's do not break down in the environment and, in fact, tend to accumulate through the food chain. The U.S. Food and Drug Administration establishes limits for deleterious substances in human food. The current FDA "Action Level" for PCB's in fish flesh is 2 parts per million (ppm). This standard became effective August 20, 1984. The level bad previously been 5 ppm. While the hazards of PCB consumption are largely undetermined, they are a suspected carcinogen. Therefore, anglers and others may wish to limit their consumption of fish containing PCB's. Pregnant women and those breast feeding their children may wish to avoid all food contaminated with PCB's. This information sheet does not recommend consumption of PCB contaminated fish. The consumption of contaminated fish is a matter of personal choice. The following guidelines are intended to help the angler and his family reduce their dietary intake of PCB's. Ways to Limit Intake: 1. Learn to differentiate between stocked and stream-bred fish - Recently-stocked fish will contain less PCB than carryover or native fish. Stocked trout can be differentiated from native trout by observing the fish's color and fin size. Stocked trout will be a dull color and bav6 much smaller, often misshapen or tattered fins. 2. Eat smaller, younger fish - PCB's are stored in fat tissue; therefore, the older, larger, and fattier a fish becomes, the more likely it will contain PCB's. A reduction in dietary intake,of PCB's can be achieved by keeping smaller fish for the table. 3. Prepare and cook the fish properly - While the method of cooking does not appear to greatly affect PCB content, proper preparation and care of the fish before and during cooking will reduce the amount of PCB's consumed. a. Remove the skin before cooking. b. Remove fat - The dorsal layer of fat (under the dorsal fin along the fish's back), the belly fat, and a layer of fat along the lateral line (along the mid-line of the side of the fish) should be trimmed off. The Wisconsin Department of Natural Resources found that such trimming reduced PCB content of the fish by 30%. c. The juices and fat s that cook out of the fish should not be eaten or reused for cooking other foods. The cooked fish should not be left in the pan with its juices and fats. Information concerning PCB contamination of particular fish species in specific streams in Pennsylvania is disseminated through press releases. It is also available in "PCB's in Pennsylvania Waters," Publication #51 of the Bureau of Water Quality Management, Department of Environmental Resources. The address is P.O. Box 2063, Harrisburg, PA, 17120. Free copies are available. This information sheet was prepared by Donald F. Knorr, Water Pollution Biologist, in cooperation with the Division of Water Quality. Publication #12-3600(78) of the Wisconsin Department of Natural Resources, was the chief data source. I I I I I I I I A P P E N D I X I I Food and Drug Administration I Guides .- PCB's and Cblordane I I I I I I I .I I FOOD AND DRUG ADMINISTRATION COMPLIANCE POLICY GUIDES GUIDE 7108.19 CHAPTER 8 - FISH AND SEAFOOD INDUSTRY SUBJECT: Polychlorinated Biphenyls (PCB's) in Certain Freshwater Fish BACKGROUND: The Commissioner of Foods and Drugs promulgated a regulation (21 CFR 109.30a in the Federal Register of July 6, 1973 (38 F.R. 18096) that established limits (temporary tolerances) on the amounts of PCB's that may be lawfully present in food as a result of unavoidable contamination. PCB's are a class of toxic industrial chemicals that are highly stable, heat resistant and nonflammable. One consequence of the contamination of the environment with PCB's is the indirect contamination of certain foods, principally those of animal origin. Surveillance data gathered by FDA and USDA subsequent to the effective date of the temporary tolerances have shown that, with the exception of certain freshwater fish, the presence of PCB's in those individual foods subject to the tolerance continues to be sporadic, and that there has been an overall and substantial decline in frequency and levels. Fish collected from commercial establishments contained PCB's substantially below levels which would pose a risk to health. However, considerably higher levels were detected in various species of freshwater fish-- particularly sports fish, e.g., coho and chinook salmon, lake trout--from several areas in the United States. The commissioner is concerned about the public health implications for sportfishermen and others who may regularly consume fish that are caught in PCB contaminated waters. On October 20, 1975, the New York Department of Environmental Conservation was informed that FDA will initiate seizures of interstate shipments of fish found to exceed the temporary tolerance for PCB's and would support the state in similar action against intrastate shipments. Should additional regulatory actions become necessary in the future, FDA will promptly take all appropriate steps consistent with its statutory authority. POLICY: Even though the temporary tolerances for PCB's in selected commodities (21 CFR 109.30(a) are currently under review, they remain in effect for unavoidable contamination of food products. Date: 10/01/80 PAGE 1 ISSUING OFFICE: EDRO, Division of Field Regulatory Guidance AUTHORITY: Associate Commissioner for Regulatory Affairs NOTE: The detailed "Statement of Policy", concerning PCB's in certain freshwater fish, appears in the Federal Register of February 26, 1976 (39 F.R. 8409/10). GUIDE 7141.01 ATTACHMENT B.3 CHLORDANE The following action levels are for residues of chlordane Including heptachlor and its epoxide, cis and trans chlordane, cis and trans nonachlor, oxychlordane (octachlor epoxide) alpha, beta and gamma chlordene and. chlordene. Levels of individual components must be quantitated at 0.02 ppm o above and confirmed in order to be added into the "chlordane" total value. Also see "analytical notes at the bottom. Commodity Action Level (ppm) Animal fat# rendered 0.3 Animal feed, processed 0.1 Asparagus 0.1 Bananas 0.1 Beans 0.1 Beets, (with or without tops) 0.1 Beet greens 0.1 Brassica (cole) leafy vegetables (except broccoli raab, Chinese mustard cabbage, and rape greens) 0.1 Carrots 0.1 Celery 0.1 Citrus fruits 0.1 Corn 0.1 Cucumbers 0.1 Eggplant 0.1 Fish 0.3 (edible portion) Lettuce melons 6.1 Okra 0.1 onions 0.1 Papayas 0.1 Parsnips 0.1 Peanuts 0.1 Peas 0.1 Poppers 0.1 Pineapple 0.1 Pome fruits (except crabapples and loquats) 0.1 Potatoes 0.1 (with or without tops) 0.1 Radish tops 0.1 Rutabagas (with or without tops) 0.1 Rutabaga tops 0.1 Small fruits and berries (except cranberries# currants# elderberries, gooseberries, and o1allie berries) 0.1 PAGE DATE April 1, 1987 FORM FDA 2678c (6/86) GUIDE 7141.01 ATTACHMENT 8.3 Spinach 0.1 Squash 0.1 Stone fruits (except Chickasaw, Damson, and Japanese plums) 0.1 Sweet potatoes 0.1 Swiss chard' 0.1 Tomatoes 0.1 Turnips (with or without tops) 0.1 Turnip greens 0.1 aAction levels for crop groups cover all commodities specified in 40 CFR 180.34(f), except wherean.exception is noted. Analytical Notes: (1) When the GLC pattern of the residue matches that of technical chlordane, quantitate against a reference standard of technical chlordane. When the residue consists of identifiable individual components quantitate against their respective reference standards and sum the components. (2) If heptachlor and/or its epoxide are proportionally higher t han the amount of chlordane present, follow the guidance in PAM Volume 1, section 300.64c and apply the action levels established specifically for heptachlor and heptachlor epoixde. QT NO. PAGE 2 Date: April 1, 1987 FORM FDA 2307 c(2/80) ACTUAL SIZE (BY, X 11) GUIDE - ATTACHMENT A P P E N D I x i Quality Assurance Data Comparisons to Other Agencies - Fish and Wildlife Service, U.S. Food and Drug Administration, U.S. Environmental Protection Agency and Nichigan Department of Public Health Quality Assurance/Quality Control Project Plan Organics Analysis Pesticide/PCB analysis in fish flesh - work-up of samples: 1. Fish flesh samples will be prepared following procedure as described in the Method Section (see Methods Manual "Draft," Appendix D). 2. Samples will be analyzed by GC/ECD according to the following protocol: When available, sets of 6 samples (including QA samples) will be run at a time. One set will consist of 4 samples, a spike and a duplicate; another set will consist of 4 (different) samples, a spike and a blank, as follows: 4 fish flesh samples I fish flesh sample to be run as a duplicate I fish flesh sample to be run as a spike alternating with 4 fish flesh samples I fish flesh sample to be run as a spike I fish flesh sample to be run as a blank Spiking will be performed as follows: Parameter Spiking Level (mg/kg) Aroclor 1260 0.5 Cblordane 0.1 Dieldrin 0.05 Toxapbene 1.0 These parameters will be separately spiked into selected samples. 3. The Department of Environmental Resources will analyze for the following: Approximate Reportin Parameter Limit (mg/kg wet wt.@ PCB's 0.2 Cblordane 0.05 Toxapbene 0.2 pp'DDE 0.02 pp'-DDD 0.02 pp'-DDT 0.05 Aldrin 0.1 Dieldrin 0.02 Endrin 0.02 Lindane 0.01 Heptacblor 0.02 Heptacblor epoxide 0.02 Metboxycblor 0.1 Mirex 0.02 Additional pesticides will be identified and/or quantified as matrix conditions permit. Percent lipids will also be determined and reported. Precision for the above parameters in fish flesh can be determined from the duplicate samples; accuracy can be estimated from analysis of samples with known concentrations of parameters. 4. Data from fish flesh duplicates and blanks are to be written on one of the sheets of the sample sets. The lab will supply its own blanks that contain no demonstrable pesticides/PCB's. The lab will also do duplicate analysis of selected fish. Metals Analysis 1. Sample preparation for metal analysis of fisb tissue A. Samples will be analytically weigbed prior to digestion procedure. B. Samples will be eitber lyopbilized or dried by microwave tecbnique. C. Dried samples will be digested by microwave procedures. D. Exception - Mercury separate sample is used for this digestion. A modified Metbod 245.1 procedure is used. 2. Analytical procedures for metal analysis of fisb tissue A. The following metals will be analyzed using the designated EPA metbods: Mercury Cold vapor, manual Metbod 245.1 Antimony AA, furnace Metbod 204.2 Arsenic AA, furnace Metbod 206.2 Barium ICP Metbod 200.7 Beryllium ICP Metbod 200.7 Cadmium AA, furnace Metbod 213.2 Cbromium AA, furnace Metbod 218.2 Lead AA, furnace Metbod 239.2 Selenium AA, furnace Metbod 270.2 Silver AA, furnace Metbod 272.2 3. Quality assurance procedure for analysis of metals in fisb tissue A. Samples will be analyzed in groups of six and will be as follows: 4 fisb tissue samples 1 duplicate of one of the 4 above 1 spike of the duplicate sample B. There will be a reagent blank run after every 3 groups. C. Standard operating procedure for AA furnace metbods is to run a spike for eacb metal determination and sbow a recovery factor of + 20% before the data is considered valid. 4. Reporting Limits A. The Division has not established MDL's on fish tissue. B. Reporting limits will be as low as the QA program allows. 5. Moisture determination procedure A. We propose to perform the moisture determination of the fish tissue using the CEM microwave moisture balance. B. The quality assurance for the moisture determination will be as follows: 5 fish tissue samples 1 duplicate of one of the 5 above C. Report the determination as calculated from the wet and dry weights. QUALITY CONTROL Units are ug/gram wet weight Amount Spiked Duplicates and % Recovered Sample Element Sample Duplicate Spike % Recovery 0692266 Lead 0.623 0.550 3.18 86.5 Walleye Cadmium <0.013 <0.013 1.17 108 Chromium 0.187 0.200 2.88 99.5 Arsenic <0.5 <0.5 2.45 98 Copper <0.5 <0.5 11.7 94 Selenium 0.277 0.290 1.66 110 Antimony <1.25 <1.25 6.27 100 Silver <0.05 <0.05 1.04 104 Barium <0.25 <0.25 24.5 97.4 Beryllium <0.025 <0.025 2.47 98 Mercury 0.253 0.219 0.605 77.3 Amount Spiked DuRlicates and % Recovered Sample Element Sample Duplicate Spike % Recovery 0692276 Lead 0.162 0.225 2.19 80 White Cadmium 0.017 0.013 1.1 110 Sucker Chromium 0.212 0.175 2.2 81.5 Arsenic <0.5 <0.5 2.47 101 Copper <0.5 <0.5 11.3 91 Selenium 0.427 0.544 1.56 67 Antimony <1.25 <1.25 6.14 100 Silver <0.05 <0.05 1.01 82.6 Barium 0.299 0.399 26.2 104 Beryllium .<0.025 <0.025 2.27 91 Mercury <0.1 <0.1 0.489 90 Amount Spiked Duplicates and % Recovered Sample Element Sample Duplicate Spike % Recovery 0692282 Lead 1.25 1.36 3.27 80 White Cadmium 0.038 0.031 1.17 112 Perch Chromium 0.663 0.623 2.92 89.5 Arsenic <0.5 <0.5 2.39 95 Copper 1.7 1.7 13.2 100 Selenium 0.630 0.698 1.90 101 Antimony <1.25 <1.25 6.51 104 Silver <0.05 <0.05 *931 92.5 Barium 0.375 <0.25 26.1 100 Beryllium <0.025 <0.025 2.59 100 Mercury 0.1 <0.1 .439 72 QUALITY CONTROL Units are ug/gram wet weigbt Amount Spiked Duplicates and % Recovered Sample Element Sample Duplicate S2ike % Recovery 0692283 Lead 0.716 0.583 2.33 65 Gizzard Cadmium <0.013 0.016 1.09 110 Sbad Cbromium 0.272 0.261 2.99 110 Arsenic <0.5 <0.5 2.2 89 Copper 1.11 0.744 12.69 101 Selenium 0.568 0.551 1.84 106 Antimony <1.25 <1.25 6.10 08.8 Silver <0.05 <0.05 0.969 58 Barium <0.25 <0.25 23.9 95.2 Beryllium <0.025 <0.025 2.36 95 Mercury <0.1 <0.1 0.439 76 Amount Spiked Duplicates and % Recovered Sample Element Sample Duplicate Spike % Recovery 0692287 Lead 1.2 1.75 3.18 78 Walleye Cadmium 0.19 0.014 0.855 83 Cbromium 0.451 0.52 2.36 76 Arsenic <0.5 <0.5 2.11 84 Copper 0.801 1.44 14.1 107 Selenium 0.636 0.582 1.72 86.4 Antimony <1.25 <1.25 5.68 90.4 Silver <0.05 <0.05 0.935 93 Barium <0.25 <0.25 27.1 109 Beryllium <0.025 <0.025- 2.61 105 Mercury <0.1 <0.1 0.416 78 Amount Spiked Duplicates and % Recovered Sample Element Sample Duplicate Spike % Recovery 0692270 Lead 0.663 1.19 2.82 80.5 Rainbow Cadmium 0.013 <0.013 0.963 98.6 Trout Cbromium 0.253 0.531 2.60 97.5 Arsenic <0.5 <0.5 2.27 94.0 Copper <0.5 <0.5 12.1 100.2 Selenium 0.366 0.425 1.58 93.4 Antimony <1.25 <1.25 6.27 104 Silver <0.05 <0.05 0.934 96.7 Barium <0.25 <0.25 23.5 97.9 Beryllium not run not run not run not run Mercury <0.1 <0.1 0.562 80.0 QUALITY CONTROL Units are ug/gram wet weigbt Amount Spiked Duplicates and % Recovered Sample Element Sample Duplicate Spike % Recovery 0692271 Lead 0.843 not run not run not run Rainbow Cadmium 0.020 not run not run not run Trout Cbromium .397 not run not run not run Arsenic <0.5 <0.5 2.29 92 Copper <0.5 <0.5 11.71 94.4 Selenium 0.489 0.449 1.75 101 Antimony <1.25 <1.25 5.55 89.2 Silver <0.05 <0.05 1.00 100 Barium <0.25 <0.25 24.8 99.8 Beryllium <0.025 <0.025 2.53 102 Mercury 0.178 0.179 0.679 90 results in mg/kg Fish I Wildlife 85 C Food and DrugAdmin US EPA Organochlorine pest 1098 EPA Kentucky MFGL DER kniown DER DER known DER 7/6/88 5/12188 5/19/88 6/16/88 Araclor 1242 0.352 Araclor 1254 1.2 2.5 2.53 1.7 1.5 Araclor 1260 0.266 PC8S 1.31 1.3828 0101&in 0.184 0.145 0.14" 0.22 .0.361 0.20 0.16 Tech chlordans O.S28 1.8 0.67 0.47 G-chlordam 0.035 0.049i A-chlordane 0.191 t-chlordarm 0.069 c-chl"ne 0. IrA 0.158 oxy,chlordans 0.051 0.064 0.0301 t-nonschlor 0.109 0.146 C-nmachlor O.OS6 0.0512 heptachlor epox 0.049 0.0393 toxapherw 1.58 hexachloro,- 0.011 benzerm itachloro- 0.011 0.0251 0.041 0.056 yclobexarm ppf@m 0.563 0.476 0.3S95 0.69 2.34 2.1 pp 6 41100 0.075 0.010 0.06R I.S9 0.00 @pp#.= 0.047 0.067 CON 0.87 0.68 en&In 0.013 lipidi 15.7 17.21 14.3129 1.0 0.1 0.11 78.2 walstum @ y pp DER spike results DER blank results samp I e spiked with 'k recovery sample conMent ORG-6064 0.50 ppm Aroclor 1260 104 DER blank 10.0 ORG-5610 0.50 ppm Aroclor 1260 98 solvent/reagent blank N.0 ORG-6606 0.50 ppm Aroclor 1260 118 ORG-5602 0.50 ppe Aroclor 1260 86 ORG-5598 0-50 PPM Aroclor 1260 96 ORG-3693 0.50 ppm Aroclor 1260 96 ORG-4607 0.50 ppm Aroclor 1260 118 ORG-3685 0.50 ppm Aroclor 1260 74 ORG-5310 0.10 ppm chlordane 73. ORG-3303 1.00 Ppm Aroclor 1260 107 ORG-3773 1.00 PPO Aroclor 1260 99 ORG-3681 0.20 pm chlordane 74 ORG-3677 0.05 Ppm dlieldrin 74 ORG-3773 0.05 ppm dieldrin 78 ORG-3303 0.05 ppm dleldrin 70 ORG-4603 0.1 Ppm chlordane 71 ORG-4607 O-S PPm Aroclor 1260 118 DER blank 0.5 PPm Aroclor 1242 132 DER blank 0.5 Ppm Aroclor 1248 134 DER blank 2.0 ppm toxaphene 100 MR blank 0.2 PPm chlordane 110 DER blank 0.05 ppm dieldrin 106 00 duplicate results --------------------------------------------- sample parameters found ppm ORG-3673 WE 0.034 0.033 chlordans 0.14 0.15 Araclor 1260 0.19 0.18 dieldrin trace trace 4 lipids 0.57 0.53 ORG-3680 ODE 0.10 0.094 000 -0.009 -0.008 chlordme 0.17 0.16 Araclor 1260 0.48 0.40 dieldrin trace tram 4 lipids 6.69 7.03 ORG-3683 WE -0.017 -0.014 chlordane O-OSS 0.053 Araclor 1260 -0.06S Indications a-WC -0.019 -0.017 dieldrfn trace trace % lipids 0.74 0.66 ORG-3689 WE -0.011 -0-012 Wordane -0.024 -0.02S Araclor 1260 -0.12 -0.12 % lipids 0.51 0.46 ORG-4603 DOE 0.013 0.019 Araclor 1260 0.560 0.620 chlordane 0.045 0.069 ORG-3693 WE -0.011 -0.019 Aroclor 1254 -0.09 -0.17 chlordane -0.021 -0.04S % lipids 0.88 1.59 ORG-3694 Aroclor 1260 indications indications chlordane Indications-0.041 % lipids 3.19 3.12 ORG-5600 N.D. N.D. % lipids. 0.15 0.18 ORG-SM N.D. N.D. * lipids 0.14 0.09 ORG-5607 Aroclor 1260 1.4 1.4 chlordane 0.08S 0.11 dieldrin -0.011 -0.010 % lipids 1.15 1.06 ORG-5612 WE 0.040 0.043 Aroclor 1260 0.19 0.22 chlordane 0.050 0.050 dieldrin trace trace Aroclor 1242/1248 indications indications % lipids 2.35 2.45 ORG-6063 WE 0.072 0.072 000 0.019 0.019 chlordane 0.22 0.21 Aroclor 1260 0.21 0.23 dieldrim -0.015 -0.014 % lipids 5.39 5.20 ORG-5308 WE Q,39 0.37 Ow 6.11 0.10 Aroclor 1260 1.2 1-2 a-SHC -0.012 "0.010 dieldrin 10.010 0.021 % lipids 10 .91 10.74 MT-CHIGAN DE"ARTMENT OF PUBLIC KEALTH Lake trout 0692264 CELIS 179 (2) CEHS--DIVISION OF LABORATORY SERVICES-FISH MONITORING PROGRAM Sarmple No. 88-202 Site: Pernsylvania Set: Ld ID: 6A Species: %Fat: 13 - 5 (M-P 13 Len.-Ith (cm) Weight (gm): Sarple Type: 03-TCUND RESULT pm,"An RDL M/Kg Duplicate mg/Kg Hexachlorobenzene 0.012 1 0.012 0.001 Mirex <0.005 <0.005 0.005 OmLmc (Lindane) <0.005 Q-005 0.005 Heptachlor <0.005 <0.005 0.005 Aldrin <0.005 <0.005 0.005 Octachlorostyrens 0.011 0.011 0.001 Pentachlorostyrene 0.010 est. 0.010 est. 0.001 Iiexachlorostyrene 0.003 est - 0.003 est. 0.001 Heptachlorostyrene 0.003 est. 0.003 est. 0.001 oxychlordane 0.014 0.011 0.003 Heptachlor Epoxide 0.016 0.014 0.003 g !-Mordane 0.030 0.029 0.003 trans-Nwachlor 0.153 0.149 0.003 !i@_Chlordane 0.086 0.084 0.003 4,41-DDE 0.560 0.522 0.003 Dieldrin 0.074 0.073 0.005 cis-mnw.h1or 0.071 0.074 0.003 4,4#-MD 0.243 0.236 0.005 4,41-Mr 0.063 0.09 0.005 FF-1 (PBB) <0.005 <0.005 0.005 Aroclor 1242 (PCB) 0.025 Aroclor 1254 (PCB) 2.30 2.24 0.025 Aroclor 1260 (F M) 0.025 Total Pm 2.30 2.24 0.02S Toxaphene (0.050 <0.050 0.050 Tc,_- al' Chlordane 0.340 0.336 0.020 MICHIG&N DEPARTMENT OF PUBLIC HEALTH large moutia bass 0692288 CEHS 179 (2) CEHS-DIVISION, OF LABORATORY SERVICES FISH MONITORING PROGRAM Sam .1e No. 88-201 Site: PW=Ylvania Set: Field ID: CZm#18 Species: %Fat:0-65(0UP.0.5 rangth (CM) Weight (9m): Sw9le Type: CCHPOW RESULT RMAM RDL mg/Kg Duplicate rrq/yg Hexachlorobenzene (0-001 (0.001 0.001 mirex (0-005 <0.005 0.005 SMA-MC (Lindane) <0.005 (0-005 0.005 fleptachlor (0.005 (0-005 0.005 -4 Aldrin (O.OOS (0-005 0.005 octachlorostyrene (0-001 (0.001 0.001 Pentachlorostyrene (0.001 (0.001 0-.001 Hexachlorostyrens (0-001 (0.001 0.001 Heptachlorostyrene (0.001 (0.001 0.001 Cxychlordans (0-003 <0-003 0.003 Heptachlor Epoxide (O.CO3 (0-003 0.003 k-<hlordane <0.003 (0-003 0.003 trans-NonachlOr (0-003 (0.003 0.003 m k-Mordans (0-003 (0.003 0..003 4,4'-ME 0.010 0.009 0.003 Dieldrin <0.005 (O.OOS 0.005 c is-Ncnac hlor <0-003 <0-003 0.003 0.00S <0.005 0.005 414, -mr <0-005 (0.005 O.OOS FF-1 (PBB) (O-OOS (0-005 0.005 Aroclor 1242,(PM) o.oZ5 ArOclOr 1254 (PM) 0.075 0.079 0-02S Aroclor 1260 (PM) 0.02S Tvtal Pcs 0.075 0.079 0.025 Toxaphm-m (0.050 (0-050 0.050 Chloevne Not letected Not detected 0.020 7 MICHIGAN DEPARTMENT OF PUBLIC HEALTH channel catfish 0692285 CEHS 179 (2) CEHS-DIVISION OF LABORATORY SERVICES F TSH MONITORING PROGRAM Sar-p-le No. 88-200 Site: Pennsylvania Set: F I d I D :=v32 Species: %Fat:15.3(DUP 15. Lengt:-j (crn) : Weight (gn): SarTle Type: C34:K= RESULT REMARK RM mg/Kg Duplicate wq/xg Hexachlorobenzene 0.008 0.007 0.001 Mirex (0.005 (0-005 0.005 )a-MC (Lindane) (O.OOS 40-005 0.005 Heptachlor (0.005 (0.005 1 0.005 Aldrin (0.005 (0-005 0.005 Octachlorostyrene 0.007 0.007 0.001 Pentachlorostyrene 0.005 est. 0.005 est. 0.001 Hexachlorostyrens 0.002 est. 0.002 est. 0.001 Heptachlorostyrene 0.002 est. 0.002 est. 0.001 oxychlordans 0.008 0.006 0.003 Heptachlor Epoxide 0.014 0.012 0.003 3--=ordarie 0.028 0.028 0.003 trans-Nonachlor 0.070 0.069 0.003 i-Chlordane 0.066 0.065 0.003 4,41 -DDE 0.236 0.230 0.003 Dieldrin 0.058 0.059 0.005 cis-Nmachlor 0.032 0.035 0.003 4, 41 -DM 0.120 0.118 0.005 4,41 -D[)r 0.079 0.072 0.005 FF-1 (FM) (0.005 (0-005 0.005 Aroclor 1242 (PCB) 0.025 Aroclor 1254 (PCB) 1.40 1.43 0.025 Aroclor 1260 (FCB) 0.025 Total Fm 1.40 1.43 0.02S Toxaphene (0.050 0 .050 Total Chlordane 0.196 0.197 0.020 :;@ (3) L @ca_::GkN DEr-ARMIE-W OF 7,:37JIC FEAL7H MG-DIVISION C)F LABCRA"l-MY SERVICES ETSH MONITORDG PROGRAM EPA FAT CCNTROL #137 88 ADV. SET CONME OL NO. I OOMFCLW %RE=,orMY MDPH VA= TRUE VALVE mg/,(g rng/Kg Hexachlorobenzene 102 0.050 0.049 beta-BIC 89 0.268 0.300 Oxychlordans 101 0.113 0.112 Heptachlor Epoxide 103 0.077 0.075 trans-Nonachlor 100 0.119 0.119 4,4 0 -ME 96 2.120 2.200 Dieldrin 103 0.041 0.040 4,4 0 -Mr 103 0.180 0.175 Aroclor 1254 100 1.'00 1.00 m Hirmc 99 0.128 0.129 MICHMAN DEPARTMDM CF PUBLIC HEALTH CEM-DIVISICN CF LABCRATCRY SERVICES risg mcmaumPFCGRAM FM LIPTD CCt?rRM #4 aWIROL NO. 2 CCMPCLM VOOOVERY HDPH VAUM mglKg TME VAUX mg Dieldrin 100 0.6% 0.658 Hexach.lorobenzens 105 0.043 0.041 cis-Nonachl or 94 0.955 1.02 Oxychlordans 99 0.820 0.830 Heptachlor Epoxide 98 0.691 0.705 trans-,4machlor 97 1.01 1.04 Fp,p,nm 104 7.34 7.09 p,p'DM 89 1.18 1.32 P,Plwr 96 1.47 1.53 Aroclor 1254 99 14.74 14.89 Al I gpha-Chlordans 97 0.714 0.736 gM:.a-Ctaord&ie 97 0.732 0.7S5 I I I I I I I I I A P P E N D I X K I Memo - Discussion on Testing Procedures I I I I I I I I I I The following are excerpts from a memo to Richard Sbertzer, Chief Quality Assessment Unit, Bureau of Water Quality Management, Pennsylvania DER, from Alan Bruzel, Organic Chemistry Section, Bureau of Laboratories, Pennsylvania DER, dated July 7, 1989 The discussion below refers to conversations/correspondence with the following concerning pesticide residue analysis in fish tissue: John Aus-tin, US EPA, Annapolis, MD Dan Donnelly, US EPA, Annapolis, MD Charles Finsterwalder, US FDA, Cincinnati, OH Jim Longbottom, US EPA, Cincinnati, OH Bob Welch, Michigan Dept. of Public Health, Lansing, MI Mr. Longbottom is coordinating a study for an upcoming Cblordane Conference in Missouri that may ultimately yield a uniform fish extraction/analysis procedure. The results of this study should lead to a situation where there is one fish analysis procedure whose results are comparable and laboratory independent. (Currently, there are a variety of techniques which can differ in extraction, chromatography, and interpretation for fish flesh analysis). Participating labs in this study will use capillary column chromatography (probably DB-5) with a temperature gradient; Bureau of Labs currently uses packed column isothermal chromatography for fish tissue analysis. The extraction procedure will probably involve Soxblet extraction followed by a cleanup step. Mr. Longbottom expects the results to be published in a few months time. In terms of enforcement action, Mr. Finsterwalder notes that the results of both original and check analyses must exceed the tolerance l'imit for the analyte in question before it is necessary to take action. He indicated to me that the results of original and check analyses may either be from one laboratory or from different laboratories. The action guidelines require that both the original and check analyses are to be in "reasonably close agreement" which has been interpreted by Mr. Finsterwalder to be a judgement decision made on a case-by-case basis. Precision between analyses has also been addressed by Mr. Austin who stated that the maximum difference between duplicate cblordane in water analyses is to be no more than eighteen percent per US EPA Method 608. We have calculated that historically the results from the Bureau of Labs for cblordane analysis (in EPA Water Pollution and Water Supply studies) have only varied between 1% more than actual EPA values and 12% less than actual EPA values - a difference within the 18% limit. Mr. Welch sent the Bureau of Labs his original preparation of fish tissue extracts. After check pesticide7PCB analysis by the Bureau of Labs, aliquots of Mr. Welcb's extracts were sent for further check analysis to Mr. Donnelly's EPA Annapolis, MD laboratory. The following is a brief description of the results obtained: Fish extracts supplied by the State of Michigan were analyzed by Michigan, PA DER Bureau of Labs (BurLabs) and EPA, Annapolis laboratories for alpha- and gamma-cblordane (a- and g-cblordane). Each lab used the same extracts for analysis thereby eliminating differences in quantitation of analytes due to extraction technique. There were differences, however, in the gas chromatographic conditions that were used by the different labs. For ease of presentation, the data from packed columns is presented below. Analytical results of two fish extracts prepared by the State of Hl-cbigan and analyzed by Michigan, BurLabs, Tn-dUSEPA: below are in ppm Sample 88-0202 Sample 88-0202 Dup Bur us Bur us analyte Michigan Labs EPA Michigan Labs EPA a-cblordane .086 .27 .089 .084 .20 .085 g-cblordane .03 .07 .061 .029 .07 .064 The following table details the gas chromatographic analytical columns used in analysis of the fish extracts. These are all packed columns. It should be stressed that each participating lab used different columns in the analysis of cblordane. We have at band a memorandum from Mr. Austin through Mr. Donnelly that shows that responses to both a- and g-chlordane differ depending upon what chromatographic columns are used to analyze the compounds. DB-5 and DB-608 capillary columns used by EPA showed a higher response to a-chlordane; g-cblordane's responses did not vary as much. Packed gas chromatographic columns used by labs in chlordane anaiysis of fish tissue BurLabs Michigan US EPA 1.5% SP-2250/1.95% SP-2401 3% SE-30 OV-1 4% SE-30/6% SP-2401 1.5% DV-17/1.95% DV-210 4% SE-30/6% DV-210 From the above it can be concluded that in comparison to US EPA's results: 1.) BurLabs gives bigber quantitation for botb a- and g-cblordane: 2.7 times bigber for a-cblordane 1.1 times bigber for g-cblordane 2.) Micbigan gives lower quantitation for g-cblordane: 1.0 times lower for a-cblordane, i.e. results not different from US EPA 2.1 times lower for g-cblordane It would tbus appear from the data tbat, compared to US EPA's results, BurLabs overestimates a-cblordane (and, to sser extent, g-cblordeneTand Micbigan under estimates g-cblordane. Because of the apparent discrepancies of a- and g-cblordane results between all labs, the Bureau of Labs compared its a- and g-cblordane standards used in the above quantitations witb otber available a- and g-cblordane standards. This was done to eliminate the possibility that the Bureau of Labs incorrectly prepared its a- or g-cblordane standards. It was found that the gas cbromatograpbic responses of a- and g-cblordane standards used in the analysis of the Micbigan extracts were comparable to responses from otber available a- and g-cblordane standards (US EPA and Accu-Standard, New Haven, CT). Incorrect preparation of cblordane standards by the Bureau of Labs car) be ruled out as the origin of the discrepancies. The differences in the results of this sbared fisb extract may be resolvable once the interlaboratory comparisons performed as part of the upcoming Missouri Cblordane Conference report are publisbed. That report may answer the question regarding acceptable precision between fisb tissue analyses. Different cbromatograpby conditions may be . responsible for the bigber a-cblordane levels reported by the Bureau of Laboratories and the lower g-cblordane levels reported by the State of Micbigan (in comparison witb the results from US EPA). It is conceivable that a co-eluting metabolite or cblordane constituent is responsible for the enbanced a-cblordane value reported by the Bureau of Labs. The Bureau of Labs will refrain from quantitating a- and g-cblordane using packed columns and will quantitate tecbnical cblordane only using packed columns. We will keep current witb the pro ress of Mr. Longbottom's study so that we can be prepared lor wbatever metbodology be proposes for pesticide residue analysis in fisb tissue. He anticipates the analytical metbod will involve Soxblet extraction, a cleanup step and capillary cbromatograpby witb quantitation of eacb cblordane constituent reported. We will try to-get details from this as yet unwritten Missouri Cblordane Conference metbod as bis study progresses. References 1. Comprebensive Water Quality Management Plan for the Pennsylvania Portion of the Lake Erie Drainage Basin and the Remaining Portion of Erie Count repared by Enginee'ring-Science, Inc., September l9h, PCbapter I, Page 4. 2. Great Lakes Water Quality Board, Re ort to the International Joint Commission, 1989 Report on Great Lakes Water Quality. Presented at Toledo, Obio, November 1987, pages 143-144. 3. Food and Drug Administration Compliance Policy Guides, Guide 7108.19, Cbapter 8, Fisb and Seafood Industry, Polycblorinated Bipbenyls (PCB's) in Certain Fresbwater Fisb. October 1, 1980, page 1. 4. Safe Seafood an Analysis of FDA Strategy, Report of the Seafood Task Force. U.S. Food and Drug Administration, April 1989, page 16. 5. Public Healtb Advisory. Micbigan Department of Public Healtb Sciences, December 1988 update, page 1. 6. Lake Mic.bigan Sport Fisb - Sbould You Eat Your Catcb. P@e ared by the National Wildlife Federation, Lake Mic9igan Sport Fisb Consumption Advisory Project, Barbara S. Glenn, Jeffery A. Foran, Mark Van Patten, 1400 Sixteentb Street, NW, Wasbington, DC, page 0. 7. Erie County Department of Healtb - Cascade Creek and related files, see solid waste spills, industrial discbarges, combined sewer ove@flows. Erie County Department of Healtb, 606 West Second Street, Erie, Pennsylvania, 16507. 8. 1987 Report on Great Lakes Water Quality, Apgendix B, Great Lakes Surveillance, Volume I. Comp ie and edited by David E. Ratbke, Obio State University, Columbus, Obio, and Gil McRae, Researcb Associate, International Joint Commission, Great Lakes Regional Office, Windsor, Ontario, Marcb 1989, pages 2.4-11. 9. Arcbives, Environ. Contam. Toxicol. 16, 185-207 (1987), page 202. The Effects of Sample Preparation on Measured Concentrations of Eigbt Elements in Edible Tissues of Fisb from Streams Contaminated by Lead Mining. Cbristopber J. Scbmitt and Susan E. Finger, U.S. Department of the Interior, Fisb and Wildlife Service, Colombia, Missouri, 65201. 10. Handbook of Environmental Data on Organic Chemicals, Second Edition. Karel Verscbueren, VanNostrand Reinhold Company, New York, Cincinnati, Toronto, London, Melbourne, page 351. 11. Great Lakes Water Quality Board/Great Lakes Science Advisory Board, Report to the International Joint Commission, 1985 Annual Report, Committee on the Assessment of Human Health Effects of Great Lakes Water Quality, Revision of October 1986, page 55. 12. Memo - Vincent C. White, Chief of the Division of Inorganic Chemistry and Biological Services, Pennsylvania Department of Environmental Resources, Bureau of Laboratories, to Robert J. Wellington, Aquatic Biologist of the Erie County Department of Health, dated August 24, 1989. 13. Great Lakes Water Quality Board, Report to the International Joint Commission, 1987 Report on Great Lakes Water Quality, Appendix B, Great Lakes Surveillance, Volume I, page 2.5-14 14. Guide to Eating Ontario Sport Fish, 1987. Ministry of the Environment, Ministry of Natural Resources, Ontario Ministry of the Environment, 135 St. Clair Avenue, West Toronto, Ontario, Canada, M4V 1P5, page 35. [email protected] a .IIIIIIIIIIIIND 3 6668 14103 4977 1 I i I i I I i I I i I 11 i I I I i -@