1 State University of New York At Syracuse, NY

October, 2000

        Our higher level trophic research was comprised of three components: (1) changes in species assemblage relative to environmental and habitat changes; (2) autecology of spotted seatrout; and (3) growth, recruitment and survival of snappers, grunts and barracuda. The assemblage of juvenile and small resident fishes, monitored since 1984 to present, differed markedly between 1984-1985 and 1994-1996, and present data indicate an assemblage that resembles that observed in 1984-1985. Densities of bay anchovy (Anchoa mitchelli), a pelagic zooplanktivore, had dramatically increased in 1994-1996 collections (following major environmental changes in the bay), but markedly deceased in recent collections. On the other hand, the standing crop of seagrasses observed in 1999-2000 mirror those observed in 1994-1996, which were considerably lower, particularly in the Western Subdivision, than observed in 1984-1985. Our ichthyoplankton monitoring program described the distribution and abundance of ichthyoplankton throughout the bay, and the spatial and temporal spawning habits of the spotted seatrout in Florida Bay. Trout spawn mainly in the Gulf, Western and Central Subdivisions of the bay. Results from spotted seatrout otolith microstructure analysis indicated: (1) growth, based on size at age of capture, did not differ among areas (Gulf, Western and Central Subdivisions); (2) spawning is protracted with 50% and 100% completed by mid-July and October, respectively; (3) using time series data and a covariance model, otolith increment growth during the early ages is influenced by salinity. In addition to our otolith work, a individual based bioenergetic model is being developed for larval and juvenile spotted seatrout. Large interannual differences in juvenile gray and lane snapper abundances have been observed. The distribution, abundance, growth and spawning dates of four species of grunts in Florida Bay was determined. Sailors choice (Haemulon parra), white grunt (H. plumieri), bluestriped grunt (H. sciurus) and pigfish (Orthopristis chrysoptera) settle in seagrass habitats at comparable sizes ranging from 9 to 17 mm SL. Sailors choice and white grunt are both about 40 days old at settlement. Pigfish are only slightly older, but bluestriped grunts were found to be about a month older. Pigfish grow the fastest at about 0.75 mm per day. Sailors choice and bluestriped grunts both grow about 0.50 mm per day. White grunt grew the slowest at about 0.25 mm per day. Spawning is protracted for all grunts studied. Barracuda have a protracted spawning period. Back-calculated spawning dates, corrected for natural mortality, of 375 juveniles collected from 1984 to 2000, ranged throughout the year. Peak spawning occurred during the summer with 50% of the survivors being a product of the June-August spawn. Nearly 15% of the juveniles were spawned during late fall through winter and are possibly of Carribean origin. Barracuda enter Florida Bay at a size of about 20 mm and an age of 27 days. The distribution of the youngest recruits indicates that barracuda enter Florida Bay through the Atlantic passes and the open gulf. The growth rates of barracuda varied between annual cohorts, ranging from about 1.6 to 2.1 mm d^-1. Average growth rate across all years for fish less than 250 mm SL was 1.9 mm per day.

        Proposed Everglades National Park restoration plans, which include the restoring of historic freshwater flows to the Everglades, could have a significant influence on the species assemblage, species diversity, growth, recruitment and survivorship of fishes in Florida Bay. To determine the effects of restoration plans on Florida Bay’s fisheries resources, it is essential to understand the life histories of fishes and the role bay habitats play in supporting upper trophic level fishes and their prey. This information could then be incorporated into predictive ecological models to guide managers in restoration decision making.
        The goal of our work was to obtain biological information to collaborate with ecological modelers to develop ecological models useful to restoration managers. Process-oriented research and monitoring activities were established to quantify recruitment, growth, and survivorship of fishes relative to varying habitat and environmental conditions, particularly resulting from proposed increases in freshwater flow. Research focused on species exemplifying two life history types: species that spend their entire life history in Florida Bay (spotted seatrout) and species that spawn outside of the bay (snappers, grunts and barracuda), but as juveniles use habitats in Florida Bay.
        Specific objectives were: (1) Monitor changes in the species assemblage structure of juvenile and small resident fishes in relation to habitat (including environmental factors). Results are expected to be used to verify a statistical model that evaluates the effects of water management policies on the ecosystem In addition, a major product will be a synthesis document describing the early life history of fishes in Florida Bay; (2) Examine growth, recruitment and survival of snappers, grunts and barracuda. to determine if differences in snapper recruitment and survivorship are related to habitat and environmental changes in the bay, or planktonic processes outside the bay using an ‘analysis of survivors’ approach. This work will provide a framework for managers to assess the role of anthropogenic and natural processes in affecting fish growth, recruitment and survivorship in Florida Bay. (3) Conduct autecological studies of spotted seatrout to determine the effects of environmental factors on larval and juvenile growth. This work will contribute to the development of an individual based bioenergetic model of spotted seatrout, which, in collaboration with ecological modelers, can be coupled with ecosystem models allowing predictions of the effect of restoration plans on spotted seatrout growth, recruitment and survivorship.

Identification of problem
         Florida Bay is a valuable nursery area for important commercial and recreational fishes, forage fishes, and decapod crustaceans. Recently Florida Bay has received a great deal of attention due to environmental and habitat changes that have occurred over approximately the last 13 years. Rapid changes during the period 1987-1991 included drought-induced hypersaline conditions, seagrass die-off, increased turbidity, and algae blooms (Fourqurean and Robblee, 1999). Accompanying these changes in habitat were changes in fish and decapod crustacean populations (Sheridan et al., 1997; Matheson et al., 1999; Thayer et al., 1999). A plan to restore historic flows of freshwater to the Everglades was initiated in the 1990's and prompted questions as to how these flows could influence the recruitment, growth, and survivorship of higher trophic level animals in Florida Bay.
        Variation in abundance and production of adult fish populations is largely determined by processes that act during the egg, larval and juvenile stages (Rothschild, 1986). Relatively small changes in mortality during these early life history stages can have dramatic effects on the numbers of fish surviving through the juvenile stage (Houde, 1987). Survival during these early stages is affected by a combination of predation (Bailey and Houde, 1989), interactions between growth and mortality (Meekan and Fortier, 1996; Hare and Cowen, 1997; Sogard, 1997), successful transport from spawning to juvenile nursery habitats (Boehlert and Mundy, 1987; Hare and Cowen, 1996) and changes in abiotic and biotic parameters of juvenile nursery habitats (Sogard, 1992, 1994). Proposed restoration plans in Florida Bay could have a significant influence on species assemblage, species diversity, growth, recruitment and survivorship of fishes in Florida Bay through changes in the abiotic and biotic habitats.
        Two distinct life history types of fishes inhabit Florida Bay and changes in environmental or habitat characteristics will have differential affects on these life history types. One life history type, exemplified by the spotted seatrout, Cynoscion nebulosus, spends its entire life history in Florida Bay. The other, exemplified by the gray snapper, Lutjanus griseus and lane snapper, L. synagris, spawn outside Florida Bay in Gulf of Mexico and Atlantic Ocean waters, and enter habitats in Florida Bay as juveniles. Potential change in Florida Bay habitats caused by water management policies could affect these target species differently owing to their different uses of Florida Bay.
        The spotted seatrout is a valuable sport fish and recruitment, growth and survivorship are dependent on factors that can be studied in the bay. Restoration of freshwater flows into Florida Bay could influence life history characteristics of this valuable resource. Lowered salinities can result in loss of spawning habitat as pelagic eggs sink at approximately <20-25 psu and are not viable (Alsuth and Gilmore, 1994; Johnson and Seaman, 1986), and a change in the prey assemblage that could influence growth of the early life history stages. In addition, because of poorly developed osmoregulatory systems in larvae, reduced larval growth and feeding rates could result (Holt et al., 1989). A spatially- explicit ecological model that could be applied to environmental scenarios to assist management decision making should be the ultimate goal of research or monitoring programs (Figure 1). To develop such a model requires a strong understanding of the autecology of the target species (spotted seatrout) and determination of important environmental factors that directly control recruitment, growth (either directly or by affecting the abundance and distribution of important prey) and survivorship.
        The juvenile stages of the lane and gray snappers encounter similar abiotic and biotic environments as juvenile spotted seatrout in Florida Bay, but because they spawn in offshore waters, larvae are planktonic and depend upon transport processes to reach Florida Bay. Because prevailing currents on the inner southwestern Florida shelf are to the south and southeast (Lee et al., 1998; 1999; Smith, 2000), larvae spawned there are transported to nursery habitats in Florida Bay and may represent an important source for juvenile recruits to reef habitats in the lower Florida Keys. To interpret observed patterns of variability in the recruitment and species composition of these life history types in the bay, it is necessary to understand planktonic processes that affect larval growth, transport, survivorship and in-bay processes that affect juvenile growth, recruitment and survivorship. This is critical in understanding the affects of anthropogenic changes in habitat on fishes within Florida Bay.
        Our work examined the relationship between environmental and habitat change and the recruitment, growth and survivorship of fishes in Florida Bay (Figure 1). This information is also required to distinguish anthropogenic induced change due to restoration efforts versus natural variability in fish abundance and growth. The results of our work can then be incorporated into ecological models accompanied by water management scenarios that should guide managers in restoration decision making.

        The goals of our work were to obtain information on growth, recruitment and survivorship of fishes that utilize Florida Bay, and synthesize a long-term database on juvenile and small resident fishes to support the development of statistical and spatially explicit ecological models, that will be useful to restoration managers and future research endeavors. (Figure 1, bottom boxes). Process-oriented research, coupled with monitoring results, provide information on recruitment, growth, and survivorship for valuable fishery resources, relative to habitat use under varying environmental conditions. The objectives were tightly coupled to address Central Question #5 of the South Florida Ecosystem Restoration Prediction and Modeling (SFERPM), Program Management Committee (PMC), " What is the relationship between environmental and habitat change and the recruitment, growth, and survivorship of animals in Florida Bay"? (Figure 1). Objectives were:
        (1) To examine changes in species assemblages and the habitats species occupy by monitoring the distribution and abundance of juvenile and small resident forage fishes, habitat and environmental characteristics. Results will also be used to verify a statistical model now under development (staff of the National Marine Fisheries Service, Southeast Fisheries Science Center, Miami, FL), and to evaluate the effects of the implementation of water management policies.
        (2) To synthesize life history information for abundant juvenile and small resident fishes (~30 species) based on a long-term database from 1984 to present. The emphasis of this synthesis will be to produce a source document for future modeling and research studies of the Florida Bay ecosystem. This synthesis will be prepared after 22000-2001 sampling is completed (January 2001).
        (3) To examine the distribution, abundance, growth and survival of larval snapper cohorts on the southwest Florida shelf, and examine the distribution, areal abundance, birthdate distributions, size-at-recruitment and growth of the surviving juveniles from the same cohort to determine if differences are related to habitat or environmental changes in the bay or processes outside the bay.
        (4) Undertake autecological studies of spotted seatrout by examining abundance, distribution, growth, hatchdate distributions, spatial and temporal spawning habits, a time series of environmental factors on growth, and develop an individual based bioenergetic model.

        Ichthyoplankton monitoring: Ichthyoplankton was monitored in 1984-1985 at 14 stations throughout Florida Bay (Figure 2) using standard ichthyoplankton sampling techniques to compare the distribution and abundance of ichthyoplankton between 1984-1985 and 1994-1995, following major environmental changes in the bay, and to provide a spatially comprehensive description of the abundance and distribution of ichthyoplankton that was not provided by our 1984-1985 spatially limited sampling design (Powell et al., 1989; Thayer et al., 1999). Due to financial constraints, from April to September, 1996, ichthyoplanton monitoring was concentrated at eight stations (Figure 2–stations 5, 6, 9, 10, 11, 12, 13, and 15) where spotted seatrout were historically collected. From July 1997 through November 1999, we targeted spotted seatrout at four stations - - two stations where spawning has only recently been observed (entrance to Little Madeira Bay and Whipray Basin) and two stations where spawning had historically been observed (Palm and Bradley Keys). A bottom sled was employed in the latter sampling scheme to collect recently settled fishes, especially spotted seatrout. These recently settled spotted seatrout were used in the spotted seatrout autecology aspect of our work (see below).
        

        Juvenile and small resident fishes and seagrass habitat: Five series of bi-monthly sampling trips for juvenile and small resident fishes have been conducted since 1984-1985. From May 1984 through January 1999, sampling was conducted with a two-boat otter trawl( Thayer et al., 1999). Sampling was conducted at 18 stations divided among three strata. In March 1999, we modified our sampling design to conform with the South Florida Ecosystem Restoration Prediction and Modeling (SFERPM), Program Management Committee’s (PMC) subdivision designations. Florida Bay was stratified into six subdivisions that were divided into one nautical mile squares. Because the number of trawlable squares in each subdivision were not equal, the number of stations to be sampled were weighted by trawlable area. Thirty-six stations were randomly chosen (Gulf Transition: 4; Western: 3; Central: 10; Atlantic Transition: 7; Northern Transition: 3; Eastern: 9). Sampling was conducted with a 5.5 m boat and the same 3- m otter trawl previously used. The habitat monitoring of this work was directed at water column and seagrass characterization. Temperature, salinity, water clarity (secchi disk) and water depth were recorded. Three 0.008 m² cores were taken along a transect adjacent to the trawled area for seagrass short-shoot densities, composition, blade length, and standing crop.

        Autecology of spotted seatrout (Cynoscion nebulosus) in Florida Bay Juvenile spotted seatrout used for otolith microstructure analysis were obtained from collections made in 1995 by Florida Marine Research Institute staff to examine: (1) size at age data to determine if growth differed by location; (2) hatchdate distributions to infer temporal spawning; and (3) the influence of temperature and salinity on otolith growth, a surrogate for somatic growth. Otolith methodology consisted of using polished, transverse sagittal sections from juvenile spotted seatrout. Increment counts, examined under immersion oil at 1000x magnification, were used to estimate size at age, and hatchdates. Standard regression analysis was used to compare growth (size at age) among subdivisions. Hatchdate distributions were standardized to account for mortality. Sagittal otoliths from 347 fish were aged and daily otolith increment measurements were obtained using image analysis. The counting path changed at 21 increments. Therefore the analysis was divided into two sets associated with each of the counting paths. Age estimates differed significantly between the standard counting method and the image analysis system for some fish. To eliminate inaccurate growth estimates due to ageing errors caused by reading otoliths by the two techniques, any otolith that differed by greater than 7 days or 10% was excluded from the analysis. One-hundred and seventeen otoliths were removed using these criteria. To compare growth across distinct intervals and reduce noise the observations for otolith increment width, salinity and temperature were averaged across calendar weeks for 1995. A random coefficient model was used to determine a relationship between temperature or salinity and the natural log of otolith increment width. This model was implemented using SAS Proc Mixed procedures (SAS 1997). An age interaction was included to account for differences in growth across ages. Autocorrelation between observations from the same fish was modeled using a Markov covariance structure. Observations between fish were assumed to be independent.
        Daily salinity and temperature values in 1995 were obtained from the Everglades National Park Service for Johnson Key basin, Little Madeira Bay, and Little Blackwater Sound. Values for other stations were estimated for 1995 using a relationship derived between stations from 1997-1998 data and applying to 1995. The values for Murray/Bradley Key were then predicted from the 1995 Johnson Key Basin values using the regression obtained. When there was not a monitoring station close to the sampling station, temperature and salinity were predicted using a relationship between values obtained when fish samples were taken and the closest monitoring station where 1995 data were available. In some cases a good relationship could not be determined and these fish were excluded from the study.

        Spotted seatrout bioenergetic model: Larval spotted seatrout (Cynoscion nebulosus) occur over a wide range of temperatures and salinities in Florida Bay. The significance of temperature and salinity on the suitability of estuarine nursery areas for this species was investigated in laboratory experiments by measuring the routine metabolism, ad libitum feeding rate, and growth rate. Routine oxygen consumption rates of larval spotted seatrout were measured over a range of temperatures (24, 28, 30 and 32°C) and salinities (5, 10, 20, 35 and 45psu). Spotted seatrout were collected from Florida Bay in 1999 and 2000 to determine energy density and daily ration. After wet weight and dry weights were determined, the energy density of individual fish was determined through bomb calorimetry. Stomach contents were analyzed to determine the weight dependence of maximum consumption. The mean weight of food in the stomach and laboratory-derived estimates of gut evacuation are then used to estimate the daily ration for fish in Florida Bay. These ration estimates, along with length at age data from otolith analysis will provide a field validation of the bioenergetics model. The bioenergetics model will then be used to spatially map spotted seatrout growth potential throughout Florida Bay under different water quality scenarios.

        Recruitment, growth and survival of offshore spawning upper trophic level fishes: We are seeking to determine if differences in snapper (Lutjanidae) and grunt (Haemulidae) recruitment and survivorship are related to habitat and environmental changes in Florida Bay, or planktonic processes outside the bay using an analysis of survivors approach. This work will provide a framework for managers to assess the role of anthropogenic and natural processes in affecting fish growth, recruitment and survivorship in Florida Bay.
        Over the last 15 years, we have observed large interannual differences in juvenile snapper abundance. We have also observed seasonal differences in juvenile snapper abundance, which is partially attributable to patterns in spawning and larval survival; these two combine to create patterns in back-calculated hatchdates. Also observed are seasonal differences in juvenile growth rate, which may also contribute to seasonal patterns in survivorship of snapper species (see below). We are examining the distribution, abundance, growth and survival of larvae on the southwest Florida shelf to understand how oceanic processes affect patterns of juvenile growth, recruitment and survivorship in Florida Bay. We emphasize unique life history attributes (i.e., hatchdates, size and age at settlement, growth rates, areal abundance) of surviving juveniles in the bay. Our work also examines the importance of juvenile growth and distribution on the survivorship of snappers in Florida Bay. In October 1998 we initiated a separately funded study of ichthyoplankton on the inner continental shelf west of Florida Bay during the peak spawning seasons of several species of grunts and snappers. Depth-stratified samples are collected with a Multiple Opening/Closing Net and Environmental Sensing System (MOCNESS) (Wiebe et al., 1976) and an opening/closing Tucker Trawl bottom sled along three transects extending for 50 nautical miles over depths ranging from 10 to 30 m. we are planning to continue this sampling in July and August 2000. Our ichthyoplankton stations generally lie to the south and west of stations sampled by Collins and Finucane (1984), to the east of stations sampled by Houde et. al. (1979), and to the west of stations sampled by Powell et al. (1989). As such, these collections represent an important area of the inner and middle shelf not previously sampled. Ichthyoplankton samples are initially sorted at the Polish Sorting Center. Subsequently, snapper larvae will be identified at the Beaufort Laboratory (see below). Otoliths will be removed and hatchdate distributions and growth rates will be determined. Hydrographic data will be analyzed to provide cross-shelf sections of temperature, salinity and density. Vertical distributions of larvae will be analyzed to provide preliminary biological data for input into physical circulation models. In addition, spatial and temporal patterns in larval growth rate will be analyzed to begin to define water column habitat quality relative to the transport of larvae through this habitat.
        Six weeks after each offshore cruise, juvenile snappers which successfully recruited to Florida Bay habitats are sampled with a 3-m otter trawl. In the lab, snappers are measured, weighed and their sagittae (otoliths) are removed. Otoliths are sectioned on a transverse plane and processed according to Secor et al. (1991). Otoliths are examined under immersion oil using light microscopy. All counts are made on the ventral lobe along the sulcus at 1000X magnification. Ahrenholz (in print) found that otolith increments formation is daily in snappers. Hatchdates are estimated by subtracting increment count from collection date. Growth rates are estimated from regression models. Through the use of otolith analysis, we will examine growth prior to, and after, ingress to Florida Bay nursery habitats and compare the birthdate distributions of larvae with those of the juveniles that successfully recruited. This information will provide insight on the sources of variability in species-specific patterns of growth and survival among and between cohorts that use Florida Bay.
        Accurate species-level identification of larval snappers is prerequisite for the success of this research. Through collaboration with the NOAA Charleston Laboratory, we will use genetic characters to positively identify problematic larvae. Two approaches will be used: (1) link discernable types of larvae with adults of known species based on genetic characters, and (2) use genetic characters to identify larvae where traditional characters are not useful and types can not be distinguished. Approach (1) is preferred, but both approaches might be required. Sarver et al. (1996) demonstrated species-specific variability in the 12s region of the mtDNA genome, however, they did not quantify intraspecific variability. We are currently examining sequences from multiple individuals to verify the ability to identify Lutjanus larvae to species using 12s mtDNA sequences. We will determine larval types through classic meristic and morphometric analysis. Each larvae will be digitally photographed and notes on ontogeny, spination, pigmentation will be made. We will then conduct genetic characterization of representatives of each larval type. If larval type proves to be mono-specific, all larvae previously assigned to that type can be identified to species and future specimens can be identified using traditional characters of the type (e.g., pigmentation and spination). If a larval type is not species-specific, genetic analysis will be conducted on an individual larval basis to determine species identity.
        In July 1997 we began trawl sampling every 6 weeks targeting early juvenile stages of gray snapper and lane snapper for growth studies. Young grunts and barracuda (Sphyraena barracuda) were also saved when encountered. These collections were not randomized throughout Florida Bay like our juvenile survey work discussed earlier, but, rather directed at specific habitats where we knew we could collect sufficient numbers of individuals for growth studies. We sampled edges of grass banks, mouths of channels, edges of mangroves and shallow submerged banks.

        Ichthyoplankton monitoring (1994-1996 collections): Clupeiformes (unidentified Clupeiformes, Clupeidae and Engraulidae) dominated ichthyoplankton collections. In 1994-1995 and 1996 these pelagic zooplanktivores comprised 49% and 67% of the total ichthyoplankton collected, respectively. The families Gobiidae and Callionymidae (Diplogrammus pauciradiatus) were notable components of the ichthyoplankton.
        Densities of the majority of the most abundant taxa ($1.0 larvae m^-3) differed significantly (P<0.05) spatially and temporally and densities appeared to differ notably within subdivisions. Clupeiform larvae dominated collections at nine of the 14 stations in 1994-1995. Very high densities (>100 larvae m^-3) of total clupeiform larvae occurred in the Western (station 13), Gulf Transition (station 9) and Central Subdivision stations (station 5 and 6), high densities (10.0 - 99.9 larvae m^-3) at numerous stations except those in the Atlantic Transition Subdivision, moderate (1.0 - 9.9 larvae m^-3) and low (0.1 - 0.9 larvae m^-3) densities in the Atlantic Transition Subdivision (stations 8 and 1, respectively). Collections at Gulf Transition Subdivision stations 11 and 12 in 1994-1995, that exhibited significantly lower densities relative to 1996, suggested that considerable variability in clupeiform recruitment existed at least in the Gulf of Mexico Transition Subdivision. Significant differences in total clupeiform densities were observed temporally, but because numerous clupeiforms could not be identified to family, it was not appropriate to analyze larval clupeid and engraulid densities temporally. Clupeids and/or engraulids appeared to be most abundant in summer months.
        Gobiid larvae were the second most abundant taxa overall. This taxa comprised 26% and 23% of the total ichthyoplankton collected in 1994-1995 and 1996, respectively. This demersal taxa was a dominant component at all stations, and when clupeiforms are excluded, they were the dominant taxa at 11 of 14 stations. The highest density of gobiid larvae was observed at a Gulf Transition site (station 12), and lowest at Eastern sites (stations 4 and 2). Although there was a significant difference (P< 0.05) in monthly densities, there did not appear to be a seasonal pattern, except relative low densities were observed in December.
        Clinid larvae although a dominant component of the ichthyoplankton overall, never dominated collections at any one station. This demersal taxa was not a dominant component of the ichthyoplankton at most stations in the northwest section of the bay. Highest densities ($10.0 larvae m^-3) were observed at an Atlantic Transition station (station 8), lowest densities (0.1 - 0.9 larvae m^-3) at Central and Gulf Transition stations (stations 14, 12 and 11) and were never collected at two Central and one Western station (5, 6 and 13). At all other stations, moderate densities were observed. Densities of clinid larvae were significantly different (P< 0.05) among months, and it appeared that spawning was minimal in winter months.
        Blenny larvae, a dominant component of the ichthyoplankton, was the least dominant component of the ichthyoplankton overall. This demersal taxa was never a dominant component of the ichthyoplankton at any one station. Blenniiid larvae occurred at moderate densities at two stations in the Central Subdivision (stations 5 and 6).

        Ichthyoplankton monitoring (1997-1999 collections): Bow-mounted, push- net ichthyoplankton collections were dominated (>10 fish100m^-3) by taxa that spawn within the bay and produce planktonic larvae (Table 1). Goby larvae were collected in high densities at Johnson Key Basin, but recently settled larvae were rarely collected there (Table 2 and 3). The absence of engraulids, callionymids and clupeids at Johnson Key Basin suggested that spawning of these taxa rarely occur at this site. Engraulids appeared to be more abundant in the Central and Northern Transition Subdivisions (Table 3). Callionymids, which previously were found to be dominant in the Atlantic Transition and Central subdivision (1994-1996 collections), were rare to absent at all stations except Whipray Basin, a station were clupeids also dominated. Overall, densities of larvae (taxa that spawn in Florida Bay and have planktonic larvae) appeared greatest at Whipray Basin (Table 3).

        Autecology of spotted seatrout (Cynoscion nebulosus) in Florida Bay: Spotted seatrout spawning habits in Florida Bay have been established (1994-1996 collections; Powell et al. 1989; Rutherford et al., 1989). Based on length-frequency distributions of spotted seatrout larvae collected in 1984-85 (Powell et al., 1989) and collections from 1994-1996, spotted seatrout spawn mainly in the Gulf Transition, Central, and Western Subdivisions with limited spawning in the Eastern Subdivision (Figure 2). With the exception of station 8, all stations where larval trout were collected indicated recent spawning. Spawning also occurs in the far northeastern portion of Florida Bay in Little Blackwater and Blackwater Sounds (Rutherford et al., 1989).
      Based on 1997-1999 data, where only four stations were sampled, spotted seatrout larvae were collected at relatively high densities in Whipray Basin (Table 4), and length-frequency distributions of larvae indicate spawning occurs in this area (Figure 3). Yet we rarely collected recently settled larvae, and juveniles in Whipray Basin (Table 4, Figure 4). This site has relatively sparse Thalassic testudinum standing crop (12 gms/m²) compared to Palm Key (28 gms/m² T. testudinum and 14 gms/m² Halodule wrightii). Although spotted seatrout appeared to spawn intensively throughout the spawning season in Whipray Basin, the value of this basin as a nursery area for juvenile spotted seatrout could be minimal.
        Larval spotted seatrout were collected at the entrance to Little Madeira Bay on two occasions (Table 4 ). Surprisingly, these collections were taken at salinities <15 ppt, salinities at which spotted seatrout eggs sink to the bottom and are not viable (Alshuth and Gilmore, 1994). Although the majority of larvae were > 1.5 mm (Figure 3) and could have been transported into the area, one recently hatched larvae (1.3 mm) was collected. Still it is highly unlikely that viable eggs can be produced at those low salinities.
        Growth of juvenile spotted seatrout among three subdivisions (Gulf Transition, Western and Central) was not significantly different. The regression model that best describes juvenile growth (25-95 mm) was Length = -10.1414 + 0.8758 Age (n = 346). The regression that best describes larval and juvenile growth (1.2-100.0 mm) was Log_e Length = 0.98406 + 0.04629 Age. Otolith microstructure analysis indicated there was only a significant relationship between increment width and salinity for the #21 increment group. Salinities at five stations (Bradley, Sandy Roscoe Palm and Buoy Keys, and Johnson Key Basin) averaged across weeks ranged between 27-35 ppt. Salinities at two stations (Little Madeira Bay and Little Blackwater Sound were considerably lower (5-10 ppt).
        From our 1995 juvenile collections, a hatchdate distribution was constructed from juvenile spotted seatrout. We plotted the cumulative percent frequency against hatchdate, to estimate the relative degree of spawning throughout the 1995 spawning season. We observed that 25% of the total spawning occurred by May 21st, 50% by July 15th, 75% by mid-August and by the end of October the production of eggs and larvae appears to be almost completed, but based on larval collections, very limited spawning occurs in November. Spawning peaks were observed during the weeks of May 10, and June 21.

        Spotted seatrout bioenergetic model: A bi-phasic relation of metabolism to dry weight was observed, hence a separate analyses for each group (phase I n=103; phase II n=663) was required. Analysis of covariance, using the natural log of dry weight as a covariate indicated that temperature was significant during the second phase of growth. The effect of salinity was temperature dependent, evident by the significant interaction between temperature and salinity during the second phase of growth. When the data were sliced by temperature for this group, salinity was significant (p<0.05) at 30 and 32°C. A summary of the p-values determined through analysis of covariance is below:

        Ad libitum feeding and growth rates were determined at three temperatures (24, 28, and 32°C) and the same five salinities. Response surfaces describing the influence of temperature and salinity on the routine metabolism, feeding and growth are presented to provide a bioenergetic basis for modeling the environmental constraints on growth (Figure 5).
        Spotted seatrout were collected from Florida Bay in 1999 and 2000 to determine energy density and daily ration. After wet weight and dry weights were determined, the energy density of individual fish was determined through bomb calorimetry. The relationship between percentage dry weight (%DW) and energy density (ED) was as follows:

ED (J gWW^-1) = %DW*239.16 – 1356.6 (r²=0.777)

Stomach contents were analyzed to determine the weight dependence of maximum consumption. Preliminary analysis of consumption data indicates the following relationship between maximum stomach content (C_max) and fish wet weight (W):

C_max (g food/g fish) = 0.084*W^-0.2538 (r²=0.820)

The mean weight of food in the stomach and laboratory-derived estimates of gut evacuation will be used to estimate the daily ration for fish in Florida Bay. These ration estimates, along with growth data from otolith analysis (see above) will provide a field validation of the bioenergetics model. The bioenergetics model will then be used to spatially map spotted seatrout growth potential throughout Florida Bay under different water quality scenarios.

        Juvenile and small resident fishes and seagrass habitat: The recent juvenile and small resident fish assemblage (especially 1998 collections) differed markedly than observed in 1994-1995 and resembled the assemblage observed in 1984-1985. There appeared to be a change in dominance from canopy dwelling fishes (1984-1985) to pelagic zooplanktivores (1994-1995) back to canopy dwelling fishes (1998-present). The bay anchovy (Anchoa mitchilli), a pelagic zooplanktivore, was rarely collected in 1984-1985, dominated collections in 1994-1996, rarely collected in 1998, and moderately abundant in 1999-2000. Even when this species was dominant, it was rarely collected in the Atlantic Transition Subdivision. Mean densities of two commonly collected canopy dwelling fishes (Lucania parva, rainwater killifish; Eucinostomus gula, silver jenny) were dissimilar to bay anchovy. The silver jenny, which is a transient species, was abundant in 1984-1985 (except at the Atlantic Transition Subdivision where this species is not commonly collected), but mean densities declined notably from 1994-1996. Presently, mean densities of silver jenny are similar to those observed in 1984-1985, except densities remain low (4.1 1000 m^-2) in the Gulf Transition Subdivision compared to 1984-1985 (82.6 1000 m^-2). Mean densities of the rainwater killifish, which is a resident species, were similar to those observed for the silver jenny. Rainwater killifish were abundant in 1984-1985 (except at Eastern and Atlantic Transition Subdivisions where they are not commonly collected), densities declined from 1994-1996, and densities in the Central and Western Subdivisions mirrored those of 1984-1985. Like the silver jenny, mean densities of the rainwater killifish in the Gulf Transition Subdivision were low in 1999-2000 (0.2 1000 m^-2) compared to 1984-1985 (67.3 1000 m^-2). Similarly, in the Gulf Transition Subdivision, mean densities of other commonly collected species (Haemulon plumeri, white grunt; Lagodon rhomboides, pinfish; Floridichthys carpio, goldspotted killifish; Bairdiella chrysoura, silver perch; Hippocampus zosterae, dwarf seahorse; and Orthopristis chrysoptera, pigfish) that represent diverse life history types were low in present collections relative to 1984-1985.

        The assemblage of juvenile and small resident fishes that presently mirrors that of 1984-1985 appeared to be greatly influenced by the cyclical dominance of pelagic zooplanktivorous clupeiforms (anchovies and sardines), that might be related to chlorophyll a concentrations, which have been reported to be dynamic and spatially heterogenous. The present assemblage that is dominated by canopy dwelling juvenile and small resident fishes can not be entirely explained by the recovery of seagrasses, especially in the Western Subdivision. Our observations indicate that although there is evidence for some recovery relative to our 1994-1996 observations, seagrass standing crop and densities are much lower than observed in 1984-1985. In the Western Subdivision where seagrass die-off was most pronounced we have observed notable increases in mean densities (numbers 1000 ^m-2) of commonly collected canopy dwelling juvenile and small resident fishes in 1999-2000 collections relative to 1984-1985 collections. This is exhibited (1984-1985 vs. 1999-2000) by silver jenny ( 71.2 vs 96.7), white grunt (7.9 vs. 20.7), silver perch (9.0 vs. 26.5), rainwater killifish (91.5 vs. 101.2) and pigfish (9.7 vs. 39.9). In addition, mean densities for the recreationally important spotted seatrout, Cynoscion nebulosus) were 1.1 and 9.3 for 1984-1985 and 1999-2000, respectively. Mean total fish densities in any yearly sampling series were highest in the Western subdivision, where seagrass composition is most diverse and manatee grass (Syringodium filiforme) is most dense. Yet, mean total densities in 1984-1985 (395.3 1000 m^-2) were much lower than observed in 1999-2000 samples (692.1 1000 m^-3). Total mean fish densities (numbers 1000 m^-2) in 1999-2000 collections were also higher than those observed in 1984-1985 (1984-1985 vs. 1999-2000) in Atlantic Transition (28.8 vs. 43.0) and Central (85.5 vs. 196.9) Subdivisions, and lower in East (105.8 vs. 69.7) and Gulf Transition (316.6 vs. 454.1) subdivisions. In neither of these two sampling periods were bay anchovy dominant.

        Recruitment, growth and survival of offshore spawning upper trophic level fishes:
(Family Lutjanidae: snappers)
        Gray snapper occur throughout Florida Bay and are most abundant in the Atlantic region Lane snapper, on the other hand, are most abundant in the Gulf region, common in the West and absent to rare elsewhere. Juvenile great barracuda also occur throughout the Bay. Catch rates were greatest in the Central region and about equal in the Eastern, Atlantic and Western regions. They were rarely collected in the Gulf or Northern regions. We have found that the mean density of juvenile snappers and barracuda along bank, channel and mangrove habitats is greater than that observed in the basins sampled in the juvenile and small resident monitoring program discussed above.
        We have so far examined otoliths from 224 juvenile lane snappers to determine spawning distributions and growth rates of surviving recruits to Florida Bay. Spawning is protracted with a peak in June and July and a second smaller peak during the late fall. We looked at the average growth rates for each of these cohorts and found that the late spring - summer spawned fish grew 75% faster than the fall cohort. Size at a given age was also more variable for the fall fish. Note also that the spring-summer fish first enter Florida Bay at a size of about 20 mm and an age of about 45 days. Upon completion of the processing of our remaining lane snapper we will begin processing our gray snapper samples.

(Family Haemulidae: grunts)
        We examined the distribution, abundance, growth and spawning dates of juvenile grunts in the bay. Sailors choice (Haemulon parra), white grunt (H. plumieri), bluestriped grunt (H. sciurus) and pigfish (Orthopristis chrysoptera) settle in seagrass habitats at comparable sizes ranging from 9 to 17 mm SL. Sailors choice and white grunt are both about 40 days old at settlement. Pigfish are only slightly older, but bluestriped grunts were found to be about a month older. Pigfish grow the fastest at about 3/4 mm per day. Sailors choice and bluestriped grunts both grow about a half a mm per day. White grunt grew the slowest at about a quarter of a mm per day. Back-calculated spawning dates indicate protracted spawning for all species.

(Family Sphyraenidae: barracudas)
        Otoliths from 220 juvenile great barracuda collected in Florida Bay from 1990 to 2000 have been analyzed. Barracuda have a protracted spawning period. Back-calculated spawning dates, corrected for natural mortality, of 375 juveniles collected from 1984 to 2000, ranged throughout the year. Peak spawning occurred during the summer with 50% of the survivors being a product of the June-August spawn. Nearly 15% of the juveniles were spawned during late fall through winter and are possibly of Carribean origin. Barracuda enter Florida Bay at a size of about 20 mm and an age of 27 days which is similar to observations reported for Natal estuaries (Blaber 1982). The distribution of the youngest recruits indicates that barracuda enter Florida Bay through the Atlantic passes and the open gulf. The growth rates of barracuda varied between annual cohorts, ranging from about 1.6 to 2.1 mm d^-1. Average growth rate across all years for fish less than 250 mm SL was 1.9 mm per day. As a comparison, Figure 18 in DeSylva (1963), representing data from 244 juveniles, collected from 1945 through 1956 in south Florida, suggests a growth rate of 1.6 mm d^-1.

Evaluation

        A. Goals and objectives have or will be attained in the near future. The juvenile small resident fish population was monitored from 1994 to present and data have been forwarded to the SEFSC for the development of a statistical model. The monitoring program will terminate in January 2001. Spotted seatrout otolith processing is complete and analysis of growth is near completion. The spotted seatrout bioenergetic model (Ph.D Dissertation) is still in progress. The growth of grunts and barracuda has been completed and snapper growth is near completion.

        B. Although our Florida Bay proposal for 2000-2002 was not funded, bring to closure our juvenile fish monitoring and our offshore-inshore recruitment study in the near future. This will give us two complete years of juvenile sampling with our modified stratified random sampling design as well as two years of recruitment data. We will continue to analyze data from previously funded projects and manuscripts will be prepared on NOAA base funds. Manuscripts on the (1) "The distribution and abundance of ichthyoplankton in Florida Bay" is in preparation; (2) "The life history and ecology of juvenile grunts, lane snapper, and barracuda" will be completed in 2001; (3) "Growth of spotted seatrout" will be completed in 2001; and (4) "The early life history of fishes in Florida Bay" will be initiated after termination of the juvenile fish sampling program. Presentations were given at the Florida Bay Science Conference in November, 1999. A manuscript on spotted seatrout otolith validation has been published (Powell, A. B., E. H. Laban, S. A. Holt and G. J. Holt. 2000. Validation of age estimates from otoliths of larval and juvenile spotted seatrout, Cynoscion nebulosus. Fishery Bulletin 98:650-654). We also have a robust data set on the early life history of the spotted seatrout and are presently communicating with ecological modelers (Dr. Larry Crowder and graduate students, Duke University) to find funds to collaborate in developing a spatially explicit spotted seatrout model for Florida Bay. Thus, regardless of the lack of funding support we intend to complete our obligations under prior funding.

Ahrenholz, D. W. (In print) Periodicity of growth increment formation in otoliths of juvenile gray snapper (Lutjanus griseus) and lane snapper (Lutjanus synagris). Journal of the Elisha Mitchell Scientific Society. Accepted June 2000.

Alshuth, S. and R. G. Gilmore, Jr. 1994. Salinity and temperature tolerance limits for larval spotted seatrout, Cynoscion nebulosus C. (Pisces:Sciaenidae). International Council for the Exploration of the Sea, Council Meeting Papers, ICES-CM-1994/L:17, 19 p.

Bailey, K. M. and Houde, E. D. 1989. Predation on eggs and larvae of marine fishes and the recruitment problem. Advances in Marine Biology 25:1-67.

Blaber, S. J. M. 1982. The ecology of Sphyraena barracuda (Osteichthyes: Perciformes) in the Kosi system with notes on the Sphyraenidae of other Natal estuaries. South African Journal of Zoology 17:171-176.

Boehlert, G. W. and B. C. Mundy. 1987. Recruitment dynamics of metamorphosing English sole, Parophorys vetulus, to Yaquina Bay, Oregon. Estuarine and Coastal Shelf Science 25:261-281.

Collins, L. A. and J. H. Finucane. 1984. Ichthyoplankton survey of the estuarine and inshore waters of the Florida Everglades, May 1971 to February 1972. NOAA Technical Report NMFS 6, 75 p.

Fourqurean, J. W. and M. B. Robblee 1999. Florida Bay: a history of recent ecological changes. Estuaries 22:345-357.

Hare, J. A. and R. K. Cowen. 1996. Transport mechanisms of larval and pelagic juvenile bluefish (Pomatomus saltatrix) from South Atlantic Bight spawning grounds to Middle Atlantic nursery habitats. Limnology and Oceanography 41:1264-1280.

Hare, J. A. and R. K. Cowen. 1997. Size, growth, development, and survival of the planktonic larvae of Pomatomus saltatrix (Pisces: Pomatomidae). Ecology 78:2415-2431.

Holt, G. J., M. A. Banks, J. H. S. Blaxter, J. C. Gamble, and H. von Westernhagen. 1989. Salinity tolerance and development of osmoregulation in larval sciaenids. International Council for the Exploration of the Sea, The Early Life History of Fish. Bergen, Norway, 3-5 October, 1988. Rapports et Procès-Verbaux des Réunions. 191:489 (Summary).

Houde, E. D. 1987. Fish early life dynamics and recruitment variability. American Fisheries Society Symposium 2:17-29.

Houde, E. D., J. C. Leak, C. E. Dowd, S. A. Berkeley and W. J. Richards. 1979. Ichthyoplankton abundance and diversity in the eastern Gulf of Mexico. Report to the Bureau of Land Management, Contract No. AA550-CT7-28, 546 p.

Johnson, D. R. and W. Seaman. 1986. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (South Florida)-- spotted seatrout. U. S. Fish and Wildlife Service Biological Report 82(11.43), 28 p.

Lee, T. N., E. Johns, D. Wilson and E. Williams. 1998. Florida Bay circulation and exchange study. Proceeding, 1998 Florida Bay Science Conference, May 12-14, 1998, University of Miami. Abstract.

Matheson, R. E., Jr., S. M. Sogard,and K. A. Bjorgo. 1999. Changes in seagrass-associated fish and crustacean communities on Florida Bay mud banks: the effects of recent ecosystem changes? Estuaries 22:534-551.

Meekan, M. G. and Fortier, L. 1996. Selection for fast growth during the larval life of Atlantic cod Gadus morhua on the Scotian Shelf. Marine Ecology Progress Series 137:25-37.

Powell, A.B., D.E. Hoss, W.F. Hettler, D.S. Peters and S. Wagner. 1989. Abundance and distribution if ichthyoplankton in Florida Bay and adjacent waters. Bulletin of Marine Science. 44:35-48.

Rothschild, B. J. 1986. Dynamics of marine fish populations. Harvard University Press, Cambridge, 277 p.

Rutherford, E. S., T. W. Schmidt and J. T. Tilmant. 1989. Early life history of spotted seatrout (Cynoscion nebulosus) and gray snapper (Lutjanus griseus) in Florida Bay, Everglades National Park, Florida. Bulletin of Marine Science 44:49-64.

Sarver, S. K.; Freshwater, D. W. and Walsh, P. J.. 1996. Phylogenetic relationships of western Atlantic snappers (family Lutjanidae) based on mitochondrial DNA sequences. Copeia 1996:715-721.

SAS Institute Inc. 1997. SAS/STAT software: changes and enhancements through release 6.12, Cary, NC, 1167p.

Secor, D. H., J. M. Dean and E. H. Laban. 1991. Manual for otolith removal and preparation for microstructural analysis. Belle W. Baruch Institute Technical Publication 1991-01.

Sheridan, P., G. McMahan, G. Conley, A. Williams and Gordon Thayer. 1997. Nekton use of macrophyte patches following mortality of turtlegrass, Thalassia testudinum, in shallow waters of Florida Bay (Florida, USA). Bulletin of Marine Science 61:801-820.

Smith, N. P. 2000. Transport across the western boundary of Florida Bay. Bulletin of Marine Science 66:291-303.

Sogard, S. M. 1992. Variability in growth rates of juvenile fishes in different estuarine habitats. Marine Ecology Progress Series. 85:35-53.

Sogard, S. M. 1994. Use of suboptimal foraging habitats by fishes: Consequences to growth and survival. In D. J. Stouder, K. L. Fresh and R. J. Feller, eds. Theory and application of fish feeding ecology. Belle W. Baruch Press, Columbia, SC 390 p.

Sogard, S. M. 1997. Size-selective mortality in the juvenile stage of teleost fishes:a review. Bulletin of Marine Science 60:1129-1157.

deSylva, D. P. 1963. Systematics and life history of the great barracuda Sphyraena barracuda (Walbaum). Studies in Tropical Oceanography No. 1, Institute of Marine and Atmospheric Sciences, University of Miami Press, Coral Gables, Florida. 179 p.

Thayer, G. W., A. B. Powell and D. E. Hoss. 1999. Composition of larval, juvenile and small adult fishes relative to change in environmental conditions in Florida Bay. Estuaries 22:518-533.

Wiebe, P. H., K. H. Burt, S. H. Boyd and A. W. Morton. 1976. A multiple opening/closing net and environmental sensing system for sampling zooplankton. Journal of Marine Research. 34:313-326.

Gordon W. Thayer

______________________________