Impoundment management final report

[From the U.S. Government Printing Office, www.gpo.gov]

CM-122
FINAL REPORT

"IMPOUNDMENT MANAGEMENT"
AND
"SEAGRASS MAPPING"

TC
330
-G37
F6
1986

CM 122

FINAL REPORT

11IMPOUNDMENT MANAGEMENT"

and

11SEAGRASS MAPPING"

PRINCIPAL INVESTIGATORS:

Douglas Carlson, IRMCD
Mosquito Production

G. Alan Curtis, IRMCD
Stastical Analysis

R. Grant Gilmore, HBOI
Fish and Macrocrustacean Studies

Jorge Rey, FMEL
Vegetation and Water Quality

Robert Virnstein, SEA
Seagrass Mapping

September 1986

Partial funding for this project provided by the Department
of Environmental Regulation, Office of Coastal Management,
with funds provided by the National Oceanic and Atmospheric
Administration under the Coastal Zone Management Act of 1972,
as amended.

CM 122
FINAL REPORT

"IMPOUNDMENT MANAGEMENT"
and
11SEAGRASS MAPPING"

CONTENTS

I. INTRODUCTION

A. ROTATIONAL IMPOUNDMENT MANAGEMENT AFFECTS ON
FISH, MACROCRUSTACEAN, AND.AVAIAN POPULATION DYNAMICS
AND BASIC HYDROLOGICAL PARAMETERS

R. Grant Gilmore and Dennis J. Peters
Harbor Branch Oceanographic Institution

B. VEGETATION, ZOOPLANKTON, AND WATER QUALITY

Jorge Rey, Roy Crossman and Tim Kain
Florida Medical Entomology Laboratory
Peter O'Bryan and Scott Taylor
Indian River Mosquito Control District

C. IMPOUNDMENT MANAGEMENT: MOSQUITO SAMPLING RESULTS

Douglas Carlson, Peter O'Bryan, and Robert Vigliano
Indian River Mosquito Control District

D. SPECTRAL ANALYSIS OF FISH, ZOOPLANKTON, AND MOSQUITO DATA
COLLECTED IN IMPOUNDMENT 12

Alan Curtis
Indian River Mosquito Control District

E. SEAGRASS MAPS OF THE INDIAN RIVER LAGOON

Robert W. Virnstein and Kalani D. Cairns
Seagrass Ecosystems Analysts

CM 122

"IMPOUNDMENT MANAGEMENT"
and
11SEGRASS MAPPING"

FINAL REPORT

INTRODUCTION

Enclosed are the final reports of the five principal
investigators for CM 122. Carlson, Curtis, Gilmore, and Rey
report on their work in impoundment 12, Indian River County.
Virnstein provides his report and maps of seagrasses in the
Indian River Lagoon, which was conducted as a separate project.

Dr. Virnstein's work is largely self-explanatory, and we provide
no additional summary here, except to emphasize that this is the
most complete record os seagrasses in the Indian River, and will
serve as a standard referent for any future work on seagrass
trends in this area.

The work in impoundment 12 culminates a series of NOAA/CZM funded
studies which beginnning in 1982, designed to elucidate the
impacts of various impoundment management strategies the flora
and fauna found in the high marsh. This work has been performed
in a 50 acre impounded high marsh in the Indian River lagoon
under varying passive and active management regimes. At various
periods during the last 4 years, the PIs have studied
zooplankton, fish and macrocrustaceans, vegetation, water
quality, and mosquito production. Depending on the parameter and
particular study goal, sampling hs been performed twice-weekly
(mosquitoes); twice-monthly (fish, macrocrustacean, water
quality, zooplankton); and quarterly (vegetation cover and
growth). All data has been filed on the data processing equipment
of the respective institutions, and summaries of data sets are
maintained on the local sponsoring agency's DP equipment
(provided under CM 122 funding).

The PIs each make individual recommendations concerning future
study topics, revised study methodologies, and impoundment
management. Immediately following in this section are consensus
summaries of these recommendations, and other observations by the
PIs, which we (as the local sponsoring agency) wish to emphasize.
Individual PIs may not agree in all respects with these
summaries, but all have approved their inclusion. Note that most
recommendations are tentative since they are based on data being
collected up to the present, which obviously cannot be fully
analyzed for some months to come.

IMPOUNDMENT MANAGEMENT RECOMMENDATIONS

Impoundment management recommendations are concerned primarily
with opening and closing dates of culverts, and with culvert
design, as these influence the use of the marsh by fish.

Culvert Closing Dates.

Impoundment closing dates (the date when culvert control
structures are sealed and impoundment pumping begins) are
usually set for April or'May. Impoundment managers should be
urged to refrain from pumping until forced to do so by tides
or rainfall.

Further analysis of fish and tide data for the late May-
early June period is required, but preliminary analysis
indicates that a mid-June opening of the impoundment for
several weeks may be of use -_ especially if it is
demonstrated that marsh transients who have over-wintered in
the marsh or entered in early spring will exit the marsh
during this period.

Opening Dates.

Analysis of fish populations in the marsh indicates that there
is no serious conflict between the normal September or early
October opening dates and use of the marsh by most marsh
transients and residents. Readers interested in this topic
should review Mr. Gilmore's section.

Work in this and prior years indicates some use of the open,
impounded high marsh by fish during and immediately following
the September high tide, and our preliminary recommendations
have been that impoundments should be opened for this tide.
However, analysis of seasonal tidal patterns, and observations
of several fish kills in this and other areas, lead us to
revise this recommendation.

Tidal histories for this area indicate that we can expect
estuarine water levels to fall sharply after the first
September high tide, resulting in large outflows from the open
impoundment into the estuary. This may cause water quality
problems in the adjacent estuary, and may also strand or
concentrate fish in the perimeter ditch or in pools, providing
high population densities, small water volume, very low DOs,
and high temperatures. (We may note, though, that these same
conditions are extremely advantageous for wading birds which
prey on these fish.)

We urge that September opening dates NOT be locked into
management proposals for Florida East Coast impoundments.
October opening dates may offer a better trade-off of fish
influx into the marsh versus fish kills in the marsh or
estuary from water quality, population stress, or stranding.

Impoundment managers should be free to experiment on a site by
site and season by season basis with opening schedules.
Management plans should, however, require site inspections for
fish kills after impoundment opening, with a special report
submitted to the primary permitting/regulatory agency for each
site.

Tides, culvert capacity, and the control of mosquitoes.

Agencies cannot make any generalizations about the use of
tides to control mosquitoes or augment pumping. This can only
be accomplished by studying the tidal amplitudes and
frequencies on a site by site basis. The study area, for
example, showed no monthly tidal cycle, and insufficient water
levels to flood- the marsh during the summer. Thus, use of
additional culverts to provide tidal energy to flood the marsh
would have been of no advantage in this marsh, at best
eliminating perhaps 8 to 10 hours of pumping during the study
period. Marshes several miles to the south, however, may well
have benefited from tides as an augmentation to pumping during
this same period. The only advantage additioan culvert
capacity might offer in this site would be increased transport
of fish to and from the marsh. This will be studied in the
nest year.

It remains the local sponsoring agency's feeling that the
rationales for high-volumes of culvert capacity need to be
defined more clearly before large culvert capacity becomes a
sine qua non in management plan approval. The PIs are
continuing their studies of this question.

Control of mosquitoes in impoundments by fish.

Agencies should place no confidence in the ability of marsh
resident or transient fish to control populations of
mosquitoes. For the most part, Gambusia affinis, (the
"mosquito fish") is found in low marsh areas while mosquitoes
are found in the high marsh. Fish gut analysis showed
relatively little predation on mosquito larvae. And even in
those instances where Gambusia and Aedes larvae were seen in
the same areas, the mosquito larvae persisted in high numbers
to the pupal stage. Fish production as a component of mosquito
control strategy in the impounded high marsh seems irrelevant.

Culvert construction.

Experience at this study site, in St. Lucie County, and
elsewhere has shown that inadequate engineering has often been
applied to the design and installation of the culvert and
control structure. The primary problem is the degree of
Is buoyancy inherent in flapgate/flapgate riser designs.
Permitting agencies should ask that detailed consideration be
given to ensure th@t culverts are placed securely enough to

prevent the control-structure end from shifting upwards
because of the continual floatation pressure from the water..,
Further, agencies should expect to have to re-set some
culverts a year or two after initial installation, and should
build these costs into their budgets.

Second, while it is possible to install a culvert at a
uselessly high invert elevations (leaving it out of the water
for long periods), it seems that a culvert cannot be placed
too low for fish to enter and exit the marsh. Even normally
surface dwelling fish such as Gambusia made use of a culvert
with -2 ft invert, which was entirely submerged during most
sampling periods. Thus, the limiting factors for culvert depth
are the practical problems of river-bed and impoundment-bed
elevations, not the swimming characteristics of marsh
transient or resident fish.

Finally, there is strong evidence that snook and perhaps other
fish are using the culverts as refuges prior to culvert
opening. This attractant aspect of the culvert may be
significant, and should be retained for now. Thus, for the
present, we recommend that water control structures be placed
on the inside of the impoundment, and the exterior culvert
orif--'Lce be kept open so that it can be found and entereed by
fish throughout the year.

Flooding elevations and vegetation.

One season of pumping and flooding to an elevation of 1 ft
NGVD produced vegetation stress, but no massive vegetation
kill. We will have no indication of the impact of flooding on
vegetation until next winter and spring, when we see if
seasonal species re-establish themselves. Work in this and
other impoundments has also shown that it is impossible at
this time to predict the set of vegetation which will re-
populate a marsh (either in the near term or as the climax)
when it is restored to the estuarine system.

Dr. Ray did comment, however, that provision should be made to
restore impoundments to the target flooding level after
unusual storm rains and tides. This design criteria should
take at least 10 year storms into account. Impoundment
managers should have contingency plans to achieve such water
level reductions within a maximum of fourteen days, if severe
stress on vegetation is to be avoided.

Flooding elevations and mosquitoes.

The flooding elevation of 1 ft. NGVD was the minimum which
would cover mosquito breeding sites observed in previous
studies. As expected, this elevation provided complete control
of Aedes taeniorhyncus and Aedes sollicitans. The standing
water, however, provided an opportunity for Anopholine
mosquitoes to breed. These were apparently not controlled by

fish, but never grew to sufficient @,numbprs to represent an
annoyance.

This aspect of the study simply re-confirms work performed a
generation ago: impounding works. And it is the most effective
mosquito control method which minimizes chemical applications.
only one chemical treatment was required this year during the
management period, for the brood produced by the initial
pumping. In previous years as many as 42 mosquito broods where
observed over this impoundment during the May to October
period.

AREAS OF SPECIAL INTEREST FOR FUTURE WORK

The PIs remain convinced that the most significantly changed
aspect of the impounded marsh is the perimeter ditch. It provides
refuge for predators which may cause abnormally high mortality on
transients as they leave or enter the marsh; and it may encourage
certain transients to remain in the marsh when they would
otherwise leave it, and then cause significant mortality through
low DOs or stranding.

The PIs agree that the following are significant areas of study
in the internal perimeter ditch, characteristic of most
impoundments:

Are there ways of getting fish out of the perimeter ditch
in the spring, prior to closing?

Are there practical ways of raising DOs in portions of the
perimeter ditch? And will this significantly help during the
summer months?

What are the real impacts on the total populations of those
species which may be lured into the perimeter ditch?

Additional work also needs to be done on the long term effects of
summer pumping, fall tidal flooding, and winter dry-down on
vegetation in the impoundment. These effects can only be
determined over a number of years.

PROCEDURAL RECOMMENDATIONS

We became more aware this year of the difficulties of local
government sponsorship of diverse, multi-agency research
projects, and of the impact of low funding on obtaining useful
management data from such projects.

We encountered some management difficulties with the seagrass
project caused by the fact that it was a single project, funded
to two separate agencies (IRMCD and Brevard County), with no

central project manager. Thus there was no mechanism for
resolving procedural differences. In fact, the project almost
failed because management difficulties delayed obtaining aerial
photographs.

As for the impoundment project, put at its simplest, we have
overwhelmed ourselves with data. The addition of a statistician
to the project for part of the year moved toward a unified view
of the impoundment project, but could not (in the time available)
provide anything approaching a coherent and complete synopsis. We
are convinced that a multi-disciplinary approach is the only
possible mode of achieving useful data for a complex ecological
system -- but we also feel that the only way to utilize such data
is to dedicate funding which allows the PIs to concentrate on
data analysis' and which provides for at least one scientifically
knowledgeable co-ordinator devoted to developing an over-view of
the data.

We estimate we have at least one year of primary analysis
remaining for certain data sets, in addition to data to be
gathered from the same site in the coming year. Further analysis
of the combined data sets may lag for several more years.

In retrospect, we feel we should have sought additional funding
for combined project design and data analysis earlier. When only
one team is working on one problem, this is not a concern. But if
speedy procurement of management-related data is a primary
objective, DER and the local sponsoring agencies should insist
that multi-agency projects specifically provide for a unified
approach. This is more expensive -- it may add another level of
management between the local sponsoring agency and the various
PIs -- but when possible, would yield better integrated, more
complete recommendations in a far shorter time.

In part, we did not seek such additional funding because of the
one-year nature of the funding. When we began this project, we
had no idea how long we could continue work at this site. Thus,.
each year's work was, to a degree, a series of ad hoc studies,
and there seemed no justification for a long-term data
coordinator. Three or five year funding schedules would have
provided more consistent data sets, and better correlations among
data sets.

Longer-term funding is not merely desirable from a data-analysis
perspective -- it is essential for ecological studies. One or two
years of data on seasonal phenomena yields only one or two data
points per parameter studied, and we can draw firm conclusions
neither about populations nor about their trends from such
numbers. Dr. Virnstein, for example, in this year's work on
seagrass found unexpected abundance. Is this a sign of a robust
estuarine system? or is it an aberrant efflorescence, masking a
trend of decline? A single year's work cannot pretend to address
these questions -- and so can have no management implications. If
CZM funding wishes to addKess long term estuarine management
strategies, it should fund longer-term estuarine research projects.

ROTATIONAL MANAGEMENT IMPOUNDMENT AFFECTS ON FISH, MACROCRUSTACEAN AND AVIAN

POPULATION DYNAMICS AND BASIC HYDROLOGICAL PARAMETERS

CM - 122

FINAL REPORT

R. Grant Gilmore and Dennis J. Peters

Harbor Branch Oceanographic Institution

5600 Old Dixie Highway

Fort Pierce, FL 33450

TABLE OF CONTENTS

I. Introduction .................................................... 3

II. Methods ......................................................... 6

III. Results ......................................................... 7

A. Physical/hydrological parameters ............................ 7

B. Community composition ....................................... 10

C. Temporal/Spatial dynamics ................................... 12

IV. Discussion ...................................................... 16

V. Literature Cited ................................................ 20

VI. List of Tables ................................................. 21

VII. List of Figures ................................................ 22

INTRODUCTION

Historical research accounts of Impoundment 12, Indian River County,

Florida, date from 1955 to the present. Throughout the 31 years of

investigation, numerous researchers have implemented an array of sampling

techniques in order to survey the flora, fishes, and invertebrates occurring

in this marsh area. Nearly all of these studies were conducted as

interdisciplinary research programs typically treating the interaction of

mosquito populations or mosquito control strategies with the indigenous

fauna and flora. A concise review of the past research efforts in

Impoundment 12 will better elucidate the scientific developments of the

present study.

The pioneer investigators, Harrington and colleagues, made their first

ichthyofaunal collections over a three year period, 1955-1957, in an

unimpounded marsh (site of Impoundment No. 12) inundated only by heavy

rainfall or annual autumn tides (Harrington and Harrington, 1961). In 1966,

a dike was constructed and the marsh site was impounded and managed by the

Indian River Mosquito Control District as Impoundment No. 12. This salt

marsh was again investigated by the Harringtons in 1968 in an effort to

compare the food habits of fishes from Impoundment No. 12 with previous

studies of the unimpounded salt marsh (Harrington and Harrington, 1961,

1982). During 1979, the impoundment was no longer managed and was kept

closed, having no tidal connection with the adjacent estuary. It was at

this time (1979-1980) that Gilmore made his first collections from the sites

where Harrington and Harrington (1961) had made their collections (Gilmore

et al., 1982). Comparisons of this data with data from a neighboring opened

impoundment revealed the importance of tidal access to the immigration of

marsh transient species.

From 1982 through 1984 (CZM-47 and CZM-73), Gilmore initiated one of

the most intensive qualitative and quantitative studies of Impoundment 12 to

be conducted along the east coast of Florida. Spatial, temporal, and diel

distribution and migratory habits of fishes and macrocrustaceans throughout

the marsh and adjacent estuary were studied (Gilmore, 1984). In addition, a

comprehensive study of the trophic habits of the most abundant fishes was

initiated. Due to the significance of this data, state, local, and private

interests recognized the need for the continuation of the above research in

order to accurately propose management strategies for similar impoundments

of the area. Much of the information derived from this research was

disseminated through professional presentations on local, state, regional

and national levels.

From 1982 to present, the combined efforts of the Harbor Branch

Oceanographic Institute (HBOI), the Florida Medical Entomology Laboratory

(FMEL), and the Indian River Mosquito Control District (IRMCD) have engaged

in a collaborative research investigation of Impoundment 12 (CZM-47, 1982-

83; CZM-73, 1983-84; CZM-93, 1984-85; and CZM-122, 1985-86). During the

1984-85 CZM-93 research program, the marsh impoundment remained unmanaged,

open to tidal inundation through two 45.7 cm diameter culverts separated by

approximately one mile of dike road. As in the 1982-1984 (CZM-47 and CZM-

73) study, fish and macrocrustacean distribution patterns relative to

spatial and temporal parameters, fish population trophic analyses, and

documentation of fish dispersal associated with marsh inundation were

studied producing an extensive data base from which to compare future

management scenerios. An important supplement to this study was a biweekly

survey of the bird populations feeding in the impoundment.

During the past year, 1985-1986 (CZM 122), water levels in Impoundment

12 were managed by IRMCD to passively control mosquitos for the first time

since 1978. Flapgates were installed on each of the culverts in late May

and the impoundment was manually pumped. Inside impoundment water levels

were maintained through September 16 at which time the flapgates were

removed and the impoundment was again opened to tidal inundation. This

management strategy is known as Rotational Impoundment Management (RIM) and

.now practiced by all counties with impounded marshes along the east coast of

Florida (though with varying degrees of acreage under RIM management

strategies). This year's sampling regime was an approximate duplication of

the previous year's work. Special attention was focused on comparing the

'opened' period, May 21 - September 16, 1985 with the 'closed' period, May

21 - September 16, 1986. The primary objectives were to determine the

effect of RIM on (1) major hydrological parameters and (2) fish,

macrocrustacean, and avian spatial and temporal population dynamics. In

addition to the CZM-93 data, an extensive historical data file from CZM-47

and CZM-73 can be used to further evaluate the effects of RIM on the

indigenous wetland biota.

METHODOLOGY

Sampling equipment and intensities remained the same at all site

locations as outlined in the CZM-93 final report. Culvert traps were

simultaneously set for three hours in each culvert (stations 72 and 61) on

each tide (4/24 hrs.). Replicate cast net throws were done at the 'mole

hole' site (station 30). Additionally, day and night pull nets were taken

adjacent to each culvert site (stations 71 and 60). Upper marsh sampling

continued with ten throw net casts at each the 'low breed' (station 54) and

'high breed' (station 55) sites and a single heart trap set (24 hrs.) at the

'rain guagel site (station 56). Biweekly avifaunal observations resumed

with supplimental counts taken from the dike road in addition to the counts

taken from the bird blind.

Physical data was monitored at all sites each time a particular trap

was set or sample was taken including; temperature, dissolved oxygen,

salinity, and pH. Continuous water level recordings were taken from inside

and outside the impoundment at station 61 near the south culvert. Several

months of upper marsh (station P3) water level data was taken just prior to

the flapgate installation.

RESULTS

A. Physical parameters.

Figures I - 6 present the mean values for various physical parameters

at specific locations and in broad regions of the marsh on the day of faunal

collections. The annual mean of the respective parameter is also given.

The mean marsh elevation, approximately 16.9 cm, is nearly equal to

the mean annual water level, 16.5 cm, recorded from September 1984 to August

1986. Therefore, the plot of mean water level in the figures approximates

the mean surface elevation of the marsh and the inundation of the marsh can

be predicted from the water level plots over time (Fig.s 1 -6). In the

unmanaged marsh, 1984-85, the marsh surface was infrequently inundated from

December through September. This means that the majority of the marsh

surface was not available to the aquatic fauna for nine months of the year

when the marsh was open to tidal influence. Under RIM, spring/summer 1986,

the marsh surface was artificially flooded on 21 May, over three months

before the natural seasonal late summer/fall inundation observed during the

preceding two years, 1984, 1985. Flapgates were left in place from 21 May

to 16 September. As impoundment water levels during this period exceeded

the estuarine high tide amplitudes the flapgates did not open to allow

estuarine water to enter the impoundment until early September.

Therefore,the impoundment was effectively isolated from the open estuary

during the RIM spring/summer control period, which in this case was 119

days.

The pH of the impoundment waters showed both spatial and temporal

variations. The upper marsh flats had the highest pH values while the

perimeter ditch (lower marsh) had the lowest (Fig. 1, 2). The culvert sites

had an intermediate mean pH level between the perimeter ditch and upper

marsh (Fig. 3, 4). Seasonal patterns are complicated by the malfunction of

several pH meters used throughout the study period, therefore producing

missing values during, October 1984, 1985, December 1985, January and March

1986. The most consistent seasonal phenomena is the general increase in pH

during the summer months, particularly during RIM. The minimum pH values

were recorded during the fall, October and December and during the spring,

April - May.

Dissolved oxygen values showed significant spatial and temporal

differences. Nearly all D.O. values from the upper marsh were above the

mean for the impoundment (Fig. 1). The lower marsh D.O. values were lowest

within the perimeter ditch (Fig. 5, 6), somewhat higher at the culvert sites

(Fig. 3,4). Dissolved oxygen reaches its seasonal low during the summer at

all locations and was lower during RIM than in the preceding year,

particularly within the perimeter ditch at station 71 (Fig. 5).

Salinities were highest on the upper marsh (Fig. 1), most moderate at

the culvert sites (Fig. 3,4). The most hypersaline conditions were reached

on the upper marsh during the RIM closure period, i.e. 53 ppt.

Water temperatures also reached their highest levels during RIM on the

upper marsh, i.e. mean temperature of 29.9 C. Temperatures and salinities

were most moderate at the culvert sites (Fig. 3,4). The lowest seasonal

temperatures were observed within the perimeter ditch, 2.0 C (Fig. 5,6).

B. Community Composition

A total of 33 fish and 6 decapod crustacean species were collected

between October 1984 and September 1986 in Impoundment 12 (CZM - 93 + CZM

122; Table 1). During the summer with tidal access 19 fish and 4 crustacean

species were collected, while during RIM the summer collections were limited

to 12 fish and 2 crustacean species. The species composition of the

impoundment fauna was reduced 40% by RIM, the total number of individuals by

31%. The resident species collections most significantly affected by

impoundment closure was Palaemonetes spp. and the Gulf killifish, Fundulus

grandis, both of which dropped to 0 individuals captured from the preceding
0 ummer's collection of 1,693 and 63, respectively. Other collections were
:lso affected negatively, i.e. marsh killifish, Fundulus confluentus,

sheepshead minnow, Cyprinodon variegatus and the sailfin molly, Poecilia

latipinna, with population drops of from 535 to 19, 10,310 to 1,976 and

9,396 to 3,675, respectively. These reduced captures may not be due to

actual reduced numbers of individuals due to increased available habitat and

dispersal from the standard collection sites as will be discussed below.

The only resident species increasing in number collected were the

mosquitofish, Gambusia affinis, with from 1,339 to 8,985, and the inland

silverside, Menidia beryllina, increasing from 166 to 811.

Transient species were most greatly affected with only 8 of the 28

recorded during 1984-86 collected during the impoundment closure period, 8

of 14 recorded from the preceding summer under tidal influence (Table 1).

Yet except for the ladyfish, Elops saurus, which dove from 247 to 0

individuals collected, the numerical reductions were not significant as the

summer period is not considered a major transient recruitment period. The

striped mullet, Mugil cephalus, actually increased in numbers captured, from

55 to 398. It is anticipated that many of the transient species collected

during RIM were in the impoundment before its closure and were subsequently

trapped and captured later during the summer.

Twenty-eight avian species, 6,267 individuals, were counted within the

impoundment from October 1984 to September 1986 (Table 2). A total of 23

during CM-93, 24 during CM - 122. Except for significant increases in wood

storks (21 to 203), belted kingfishers (13 to 31) and green-backed herons

(12 to 54) counts, and decreases in tricolored herons (167 to 95), little

blue herons (151 to 88), ospreys (136 to 97) and double-crested cormorants

(47 to 6) the species rankings based on number of individuals changed little

between the years. No major species changes took place during the RIM

period, except for the increase in great egret counts from 0 to 44 and red-

bellied woodpecker, an anhinga and a black skimmer were observed for the

first time during the summer months.

C. Temporal/ Spatial Dynamics

The major spatial regions of the marsh to be considered in this

analysis are the culvert sites, stations 61 and 72, perimeter ditch sites,

stations 60 and 71, and the upper marsh sites, stations 54, 55 and 56.

Faunal density estimates are based only on stations 54, 55, 60.and 71.

Stations 54, 55 and 56 are considered upper marsh, while stations 60, 61, 71

and 72 are considered lower marsh.

Figures 7, 8, 9 and 10 present the biweekly distribution of all

collections and transient/resident species between the upper and lower

marsh. Fish densities were comparable between the habitats, though

generally higher in the perimeter ditch (Fig. 7). The densities were

highest on the upper marsh when they were lowest in the perimeter ditch,

principally during the fall. Fish presence on the upper marsh corresponds

with the seasonal inundation of that habitat both under natural conditions,

during the fall, or artificial conditions during the summer of 1986 in RIM

(Fig.s 7 and 8). Transient species populations were lowest during the

summer in both the tidal year, 1985 and the RIM year, 1986 (Fig. 9).

Transient use of the upper marsh was minimal even during seasonal fall

inundations (Fig. 10). Resident use of the upper marsh was heaviest during

the fall inundation, but also obvious during RIM. Resident populations

reached their peak during the winter sea level recession, particularly
0

during year one, 1984-1985. Transient species also reached their population

peak during this period.

The three most abundant transient species were the striped and white

mullets and the ladyfish. All show similar patterns in population

dynamicswith no captures on the upper marsh flats and minimum population

levels in occurring in the marsh during the summer (Figures 11,12, 13).

Striped mullet populations reached peak seasonal abundance in the marsh

during the winter and spring, white mullet during the fall and late spring.

Ladyfish were most abundant both years during the fall and spring with

limited summer recruitment which was nearly completely stopped during the

RIM months.

Spatial and temporal population dynamics of the three most abundant

resident species, the sheepshead minnow, sailfin molly and the mosquitofish

are presented in figures 14, 15 and 16. Densities of sheepshead minnows
frequently reached level at or above 10 fish per m2 between December and

April/May (Fig. 14). After a summer population low in the perimeter ditch

the densities began to increase August and September until the upper marsh

was inundated during the fall at which time the dispersal of sheepshead

minnow across the upper marsh reduced perimeter ditch densities. During RIM

the sheepshead minnow dispersed across the upper marsh. This species is the

only species to be found on the upper marsh during the summer of 1985, prior

to RIM. The sailfin molly revealed a quite similar pattern (Fig. 15).

However, the mosquitofish did not show the same level of density declines in

the perimeter ditch during either fall inundations or during RIM upper marsh

inundations. Mosquitofish populations actually increased in the perimeter

ditch during RIM and their use of the upper marsh flats was significantly

less than the sheepshead minnow and sailfin molly.

Palaemonid shrimp were totally absent from collections made during RIM.

Their capture was principally most successful in the culvert traps. Culvert

trap collections were significantly reduced during RIM due to the absence of

water flow through the culvert. This pattern was also observed for penaeid

shrimp and portunid crabs. All of these crustaceans are most abundant in

fall to spring collections whether under tidal influence or not, however,

palaemonid shrimp are most significantly impacted by RIM.

Bird populations showed great variation in individual species counts

and total annual count patterns. This is typical of mobile widely ranging

bird species, such as the wading birds which numerically dominate the

avifauna. A comparison between the CZM-93, 1984-85, total counts and and

the CZM-122, 1985-86, total counts for the biweekly periods reveals this

variation (Fig. 17). The primary analogous seasonal patterns observed

between the two years are the January/February peak associated with the

winter sea level recession and concentration of fishes in the lower marsh.

A similar peak was observed in April/May also associated with a sea level

recession. Wading bird use of the marsh was somewhat elevated during RIM

but not significantly so.

Individual species revealed great variation in seasonal use of

impoundment 12. This variation can be seen in white ibis, snowy egret and

great egret population pattern revealed in figure 18. There is a general

fall and winter peak in all three species, a spring peak in snowy and great

egrets. White ibis populations reached a seasonal low during the spring and

early summer. The only species of the three to show an obvious positive

reaction to RIM was the great egret.

Two endangered wading bird species that utilized the marsh in

significant numbers were the roseate spoonbill and wood stork. Roseate

spoonbills were present in the marsh both during RIM and during the tidal

access period with principally a consistent summer use. However, spoonbills

were observed during the winter months the second year but not during the

first year of study, indicating that their may be considerable population

variations between years. The wood stork use of impoundment 12 is

principally limited to the fall and winter months with increased use during

the second year.

Avian spacial distribution patterns within the marsh observed during

RIM revealed an increased use of the south and north quadrats during RIM

over the previous years observations.

DISCUSSION

The overall objective of the CZM - 122 research program was to

determine the affects of a Rotational Impoundment Management program on the

impounded marsh ecosystem. These affects were to be determined based on

several years of historical study of this particular marsh and impoundment

(Harrington and Harrington 1961, 1982; Gilmore et al. 1982; CZM - 47, CZM

73 and CZM - 93). Our knowledge of this ecosystem has progressed to the

point that some conclusions and recommendations can be made relative to

indigenous fish, decapod and avian popu lations.

A. Water Level Management Affects

The positive aspects of water level management are complex enough at

our present level of analyses to allow only conjectural statements to made.

There is little question that resident fish and crustacean populations are

permitted to utilize the upper marsh surface for a longer period of time.

These organisms disperse over nearly the entire marsh surface. The most

abundant fish species, the sheepshead minnow, forms territories for mating

and feeding across the marsh. Their densities on the upper marsh typically
varying from 0.1 to 5.0 per m2 with alternating high and low densities on a

four week cycle. The densities during RIM were similar to those observed

during the fall inundation. This observation further demonstrates that this

species is territorial with some optimum density across the upper marsh

flats. The sailfin-molly and mosquito are also commonly captured on the

upper marsh but are not territorial. These latter species show significant

variation between the fall inundation habitation of the upper marsh and the

RIM period. These observations suggest that their is a carrying capacity or

limitation on the number of fish that can occupy the upper marsh flats,

particularly concerning the most abundant and territorial sheepshead

minnows. Therefore, the time at which this carrying capacity is reached is

critical when comparing RIM with natural fall inundations. Figure 4

indicates that it is reached within the first few weeks of marsh inundation.

The longevity of sheepshead minnows without predation is such that the same

fish could survive throughout the RIM and fall inundation period, i.e. 6 to

7 months. Therefore, if no predation occurs the turnover of individuals

would be low, larval and juvenile mortality high (to territorial adults).

This would in turn mean that longer flood periods would not export

additional protein from the upper marsh to the lower marsh or estuary than

would a typical fall inundation period of 2 - 3 months. The only way it

would is if piscivorous birds and fish could consume more sheepshead minnows

during the RIM period. With increased consumption sheepshead minnows there

would be increased turnover per unit time and with a longer inundation

period more export of biomass (in the form of consumed sheepshead minnows)

from the upper marsh. However, there is no indication that predator

consumption of sheepshead minnows increases during the RIM period,

particularly as the estuarine access to the marsh is closed and wading use

of the impounded marsh is not great during the summer RIM period. It is

quite possible that the reduction in piscivorous predators during RIM may

have had a positive affect on mosquitofish populations within the perimeter

ditch allowing the observed rapid increase in their numbers.

A negative aspect of water management is the decreased populations of

certain marsh residents which depend on periodic marsh exposure for

effective reproduction. Major examples are the marsh killifish, Fundulus

confluentus and the Gulf killifish, F. grandis. Both declined significantly

during RIM and were also found to be eradicated by impoundment closure in

previous studies (Gilmore et al. 1982).

B. Impoundment Closure Affects

Impoundment closure has been found to effectively stop transient fish

and crustacean migration between the estuary and the impounded marsh. Those

species which due make tidal migrations into the impoundment during the

summer months typically had populations reduced to 0 during RIM (i.e.

ladyfish and palaemonid shrimp). An exception was the increased catch of

white mullet during RIM over the preceding annual catch under tidal

conditions. This closure would also limit resident fish migration to the

adjacent estuary.

Closure also does not permit the emigration of fishes from the

impoundment during periods of physiological stress. The summer months are

the period of greatest hypoxic stress and we have recorded fish mortalities

during this period in closed systems (Gilmore et al. 1985).
0

A positive aspect of closure may be the reduced predation on

mosquitofish as mentioned above. As mosquitofish are not territorial they

may be able to reproduce in significant enough numbers free of major

predatory mortalities and therefore allow a larger standing crop to develop

prior to the opening of impoundment. The fate of this standing crop after

the impoundment is opened to tidal influence is open to speculation. The

mosquitofish may migrate into the estuary and subsequently be consumed in

another habitat or they may be consumed within the impoundment as

piscivorous fishes and birds reach their peak populations for fall through

spring. In either case the production of fish and subsequent export of this

production from the salt marsh to other ecosystem will have increased.

Management Improvements - Water Level Control

The maintenance of a water level at a desired height is both necessary

for adequate mosquito control and preservation of the indigenous vegetation.

It also allows fish access to the upper marsh flats. The significance of

this latter consideration is not yet determined so it is difficult to

maintain that water levels promote fish, crustacean and avian use of the

impounded upper marsh plain and increase resident populations and biological

export over beyond the levels typical of the fall sea level inundation.

There has been some difficulty in setting control structures for water

level control. Control of culvert elevation has been most difficult.

Therefore, there should be some allowance for this variation within the

design of the flapgate and weir structure fitted to the end of the culvert

that actually controls the managed water height. The north culvert at site

71 was not functional in water height management due to the floatation of

the inside end of the culvert. As the weir structure only covers the upper

end of the water control assembly it was not effective in allowing water

release. This could be effectively modified in two ways: (1) Attach a

flapgate weir in which the entire structure is a weir which can be easily

modified by pulling riser boards, or (2) set the culvert elevation at lower

levels and brace during the initial set, which will insure a lower elevation

for the weir structure which is at the top of the water control assembly.

There is some hypothetical argument for the placement of the water

control assembly on the impoundment side of the culvert. This would allow

the fish queing on water release, which is now thought to occur in snook, to

aggregate in a protected environment within the culvert rather than along

the estuarine edge outside of the culvert where predation pressure by larger

fish can be quite intense.

Management Improvements Impoundment - Estuarine Access

The initial closure of the impoundment is critical in that several

transient species can be captured in the impoundment without the ability to

egress until the fall inundation. This in effect means that they will have

to survive the most anoxic period of the year, the summer, in the perimeter

ditch habitat. Periodic fish kills have been observed at other sites during

this period and nearly all of the fishes suffering hypoxia were transient

species. For this reason it would be quite beneficial to develop and test a

mechanism for controlling impoundment water levels and yet allow or induce

transient fish emigration from the impoundment. I suggest that serious

consideration be given to this problem or RIM impoundment management

strategies could in effect be blamed for trapping and killing fishery

species during a critical period in their life histories.

A mid-summer dry down of the impoundment is another possible access

improvement that may be examined under experimental conditions. This would

allow transient fish egress during a period in which typically leave the

confines of the marsh. The possible affects on mosquito oviposition must be

considered, but the July low in sea level and low dissolved oxygen

conditions within the impoundment would allow an opening at this time to

flush the perimeter ditch without flooding the principal breeding areas on

the upper marsh. Rainfall would. There is also the possibility that a

release of anoxic water from the perimeter ditch could also produce a fish

kill in the adjacent estuary which has now been documented at other sites.

Conclusion

The present RIM program is basically compatible with the migratory

habits of the principal organisms utilizing the marsh. The impoundment

closure comes during a period in which most transient species do not utilize

the regional salt marsh habitat. The artificial inundation during the

summer is deleterious to some marsh residents but also allows other
0 residents to take an advantage of the increased availability of the marsh

flats for both spawning and feeding. Avian use has only been seen to

increase significantly in one species, the great egret, but no overall

deleterious effects have been seen in avian populations due to RIM. This is

only the initial evaluation of the Rotational Impoundment Management program

and we strongly suggest that there be further study of both standard

management practices and modified management under experimental conditions.

These results are preliminary.

LITERATURE CITED

Harrington, R.W., Jr. and E.S. Harrington. 1961. Food selection among

fishes invading a high subtropical salt marsh; from onset of flooding

through the progress of a mosquito brood. Ecology, 42:646-666.

Harrington, R.W., Jr. and E.S. Harrington. 1982. Effects on fishes and

their foraging organisms of impounding a Florida salt marsh to prevent

breeding by salt marsh mosquitos. Bull. Mar. Sci., 32:523-531.

Gilmore, R.G., D.W. Cooke and C.J. Donohoe. 1982. A comparison of the fish

populations and habitat in open and closed salt marsh impoundments in

central-east Florida. Northeast Gulf Science, 5:25-37.

Gilmore, R.G. 1984. Fish and macrocrustacean population dynamics in a

tidally influenced impounded subtropical salt marsh. Final Report:

Florida Department of Environmental Regulation-CZM-47: 35 pp.

Gilmore, R.G., D.J. Peters, J.L. Fyfe and P.D. O'Bryan. 1986. Fish,

macrocrustacean, avian population dynamics in a tidally influenced impounded

subtropical salt marsh. Final Report FDER-CZM 73, -93: 25 pp.

ACKNOWLEDGEMENTS

Mr. Peter O'Bryan of IRMCD aided with field collections and laboratory

treatment of specimens. Mr. Peter Hood, Ben McLaughlin, Derek Tremain, Doug

Scheidt and Ron Brockmeyer of HBOI aided with various aspects of field and

laboratory activities.

LIST OF TABLES

Table 1. Species list of fishes and macrocrustaceans collected in

Impoundment No. 12 from September 1984 through August 1986. Tabulations

represent a comparison of the total number of individuals collected from all

sites during each year of the study periods, the opened management periods,

and the closed management periods.

Table 2. A species list of the avifauna observed in Impoundment No. 12 from

September 1984 through August 1986. Tabulations represent a comparison of

the total number of individuals observed over the entire impoundment during

each year of the study periods, the opened management periods, and the

closed management periods.

Table 1. Species list of fishes and macrocrustaceans collected in
Impoundment No. 12 from September 1984 through August 1986. Tabulations
represent a comparison of the total number of individuals collected from all
sites during each year of the study periods, the opened management periods,
and the closed management periods.

Total Study Period Open management Period Closed Management Period
October 1984 - September 1986 October - May June September
Fish and Macrocnist ans CH93 CM122 CM93+CM122 6Q493 0qM22 CH93 C2qH122
R,anked by Abundance 84-85 85-86 84-86 84-85 85-86 84-85 85-86

Cyprrinodon variegatus 83579 28020 111599 58452 26044 10310 1976
Poecilia. latip 39640 20884 60524 13710 17209 9396 3675
Gambusia, affinis 20707 20834 41541 13846 11849 1339 8985
Palaemonetes spp 10644 5321 15965 8761 5187 1693 0
Fundulus confluentus 3374 803 4177 2775 720 535 19
Mugil cephalus 2732 6727 9459 2663 6708 55 398
Mugil cerema 1297 485 1782 1189 87 103 241
Menid beryllina 1184 2703 3887 992 2462 166 811
Elops saurus 951 2075 679 1264 247 0
Centropomus undecimalis 664 814 1478 625 814 12 0
Gerres cinereus 199 46 245 193 46 1
Fundulus grandis 132 913 1045 28 755 63 0
Penaeus 0M 111 71 182 99 70 8 1
Lucania parva 71 34 105 15 28 7 6
Callinectes spp 57 15 72 36 15 14 0
Bucinosoma harengulus 48 0 48 48 0 0 0
Gobiosoma robustun 41 1 42 40 1 1 0
Diapte auratus 15 0 15 15 0 0 0
Brevoortia 14 14 28 14 14 0 0
Syngnathus scovelli 7 11 18 3 11 4 0
Lutjanus griseus 7 3 10 1 2 3 1
Rxinostamus argenteus 6 1 7 6 1 0 0
Anguill rostrata 5 1 6 5 1 0 0
Gobiosama bosci 5 0 5 5 0 0 0
Alpheus amillatus 4 1 5 3 1 0 0
Ara spp 3 1 4 0 1 3 0
rbamboides 3 4 7 2 4 1 0
Pogonias cromis 3 52 55 2 43 1 9
Uca soo 3 13 16 3 1 0 12
Harengul. 3 0 3 3 0 0 0
Microgobius gulosus 3 0 3 3 0 0 6
Dormitator maculatus 2 6 8 2 0 0 0
Eucinostcuis gula 2 0 2 2 2 0 0
Leiostcmis xanthirus 2 2 4 2 2 0 0

Trachimtus falcatus 2 0 2 2 0 0 0
Achirus lineatus 1 0 1 0 0 1 0
Fundulus similis 1 3 4 0 2 1 1
zosterae 1 1 2 1 1 0 0
barractida 1 0 1 1 1 0 0

Totals 165524 89875 255399 104226 73356 23964 16519

Table 2. A species list of the avifauna observed in Impoundment No. 12 from
September 1984 through August 1986. Tabulations represent a comparison of
the total number of individuals observed over the entire impoundment during
each year of the study periods, the opened management periods, and the
closed management periods.

Total Study Period Open Management Period Closed Management Period
October 1984 - September 1986 October May June September
Avian Species CM93 CM122 CM93+CM122 CM93 CM122 CM93 CM122
Ranked By Abundance 84-85 85-86 84-86 84-85 85-86 84-85 85-86

White This 1370 1618 2988 1145 1464 245 154
Snowy Egret 311 411 722 220 305 91 106
Least Sandpiper 326 243 569 326 243 0 0
Great Egret 209 227 436 209 183 0 44
Tricolored Heron 167 95 262 135 67 32 28
Little Blue Heron 151 88 239 122 64 29 24
Osprey 136 97 233 118 75 18 22
Wood Stork 21 203 224 21 203 0 0
Roseate Spoonbill 69 86 155 31 63 38 23
Great Blue Heron 35 41 76 28 25 7 16
Brown Pelican 37 32 69 37 32 0 0
Green-backed Heron 12 54 66 9 30 3 24
Fish Crow 40 26 66 38 18 4 8
Double-crested Cormorant 47 6 53 47 6 0 0
Belted Kingfisher 13 31 44 12 24 1 7
Red-bellied Woodpecker 1 18 19 1 15 0 3
Pileated Woodpecker 12 0 12 12 0 0 0
Mottled Duck 0 7 7 0 7 0 0
Ring-billed Gull 6 -0 6 6 0 0 0
Turkey Vulture 2 3 5 2 3 0 0
Black Bird 4 0 4 4 0 0 0
Anhinga 1 2 3 1 1 0 1
Black-crowned Night Heron 1 1 2 0 1 1 0
Semipalmated Plover 0 2 2 0 2 0 0
Common Ground Dove 2 0 2 2 0 0 0
Screech-Owl 0 1 1 0 1 0 0
Black Skimmer 0 1 1 0 0 0 1
Red-shouldered Hawk 0 1 1 0 1 0 0

Totals: 28 Species 2973 3294 6267 2504 2823 469 471

LIST OF FIGURES

Figure 1. Physical data comparisons of temperature, salinity, dissolved

oxygen, pH, and inside impoundment water level for the combined lower marsh

sites (stations 60, 61, 71, 72) of Impoundment No. 12 from September 1984

through August 1986.

Figure 2. Physical data comparisons of temperature, salinity, dissolved

oxygen, pH, and inside impoundment water level for the combined upper marsh

sites (stations 54, 55, 56) of Impoundment No. 12 from September 1984

through August 1986.

Figure 3. Physical data comparisons of temperature, salinity, dissolved

oxygen, pH, and inside impoundment water level for culvert station 72