projects > influence of hydrology on life-history parameters of common freshwater fishes from southern florida > 2001 Proposal
Project Proposal for 2001
Project start date: October, 1999 (FCSC); January, 2000 (FIU)
Project co-chief: William F. Loftus
Project co-chief: Leo Nico
Program(s): Integrated Natural Science Program - South Florida Study Area
Program element(s)/task(s): Element 1: Preparation of Information Synthesis Task 1.6: Ecological modeling and Success Criteria
WORK PLAN - Year 3
Project summary: In the effort to restore the Everglades, fishes have been recognized as good indicators of restoration success. Our long-term fish database was used in the fish simulation model developed to evaluate future restoration scenarios and estimate historical fish communities. Much life-history data in the model had to be estimated because no data were available. We intend to use previously collected specimens and new collections to study recruitment, age, and growth in representative species of large- and small-bodied fishes. The data will improve the fish model, and explain patterns of community change observed in the long-term fish database, giving us more confidence in the tools used to restore this internationally recognized ecosystem.
Project objectives and strategy: In the Everglades and Big Cypress wetlands, natural environmental conditions regularly vary on seasonal, annual, or multi-annual time-scales, to which the wetland fishes respond by shifts in relative abundance and standing stocks (Roman et al. 1994; Trexler et al. 1996). Human activities have affected the natural variability by altering the seasonality and areal extent of flooding in the wetlands, by drainage, or by over-inundation (Gunderson and Loftus 1993). The human-induced changes have resulted in loss of diversity, changes in composition and abundance, and possible alterations in energy-flow pathways (Loftus et al. 1990). Although these changes in population size and community structure are ultimately related to marsh hydrology, proximate factors that explain some of this variation probably include inherent reproductive adaptations to ambient conditions, such as the seasonality of recruitment, the age/size at first reproduction, growth rates, and other important life-history parameters.
Fishes are essential to the successful functioning of wetlands in southern Florida through their roles as prey and predators. Any changes that reduce the population sizes, community composition, or availability of aquatic animals will affect all facets of the ecology of these wetlands. Therefore, fishes have been recognized by the multi-agency groups responsible for guiding the Everglades restoration process as a key indicator group by which to measure restoration success. Communities of small-bodied fishes have been studied in EVER since 1977 by using throw traps in spikerush marshes at many Everglades sites (Loftus and Eklund 1994, Trexler et al. 1996). The data analyses allow evaluation of seasonal and long-term dynamics, shifts in relative abundance and size-structure, and produce correlations of fish abundance to water depth, hydroperiod, and plant community structure.
Despite an increasing emphasis on fish studies, important gaps in baseline knowledge remain. Life-history parameters of important fish species remain to be studied at sites along hydroperiod gradients. Some life-history characteristics are known to be rather plastic in response to the environment. We intend to study the effect of hydroperiod on recruitment, size/age structure, growth, and fecundity, which, in turn, determine fish population dynamics. At present, data on fish reproduction, age and growth, and other life-history characteristics are confined to a few species from a limited area of long-hydroperiod marsh in central Shark River Slough. As we continue the analysis and synthesis of data from the long-term fish collections, life-history information will help explain patterns of fluctuations in the time series. Accurate life-history data are also very important in building credible simulation models like ATLSS. Without empirical life-history data from a range of environments, the model will be simplistic and inadequate.
STUDY DESIGN - In this study we propose to collect information for common large- and small-bodied species by taking advantage of existing or newly funded fish studies in south Florida. These include the throw-trap program for small-fish monitoring, and a new electrofishing study of larger native and introduced species in canals. From the electrofishing study, Dr. Nico and a graduate student will focus on the collection of large-bodied species.
For the small-bodied species, Dr. Joel Trexler of Florida International University and his students will concentrate on the following species that inhabit marsh habitats and, about which, little life-history information is known. Age and growth information has been collected for some of the species (Haake and Dean 1983), but reproductive characteristics require study. These species, in total, represent the most numerous species in the marshes.
Age and Size Relationships to Estimate Growth Rate - We will establish age-to-size relationships for three marsh fish species: sailfin molly (Poecilia latipinna); flagfish (Jordanella floridae); and spotted sunfish (Information specifically on marsh populations of spotted sunfish is needed). These relationships will be estimated at one representative short (at Shark Valley) and one representative long hydroperiod location (at Shark River Slough). The proposed work will include counts of daily rings on otoliths, and size-frequency analyses to create an age-at-size table for each species, and to estimate growth in two different seasons. These studies will allow us to construct life-tables for the species under different conditions in the Everglades.
We will conduct experimental rearing of these fishes to a known age in field cages in anticipation of otolith removal and interpretation. In past efforts, this approach has demonstrated a very high fidelity of daily ring deposition in sailfin mollies up to the age of 21 days. This result needs to be repeated and expanded to other species. For several reasons, ring deposition becomes less easily interpreted as fish age. These include that the thickness of ring width decreases with age making them increasingly difficult to discriminate, disease and physiological stress unrelated to age also affect deposition. With livebearing fishes it is possible to place neonates into the cages and raise fishes of known age. However, with spotted sunfish and flagfish it may be necessary to take fish of unknown age and mark their otoliths with tetracycline. This is accomplished by emersing the fish in a tetracycline solution that intercalates with the calcium deposited at that time. The rings deposited under these conditions glow under ultraviolet light. The fish is maintained in a cage for a period of time after marking and prior to otolith removal and analysis. The number of rings deposited after the tetracycline dip may then be callibrated with the time known to have passed. This is especially useful to calibrate aging in adult fishes.
Reproduction/phenology - We will document reproductive phenology and output in six marsh fish species: least killifish (Heterandria formosa); bluefin killifish (Lucania goodei); golden topminnow (Fundulus chrysotus); eastern mosquitofish (Gambusia holbrooki); sailfin molly; and flagfish. We will attempt to use existing archived collections of fishes from the extensive spatial and temporal marsh study collections for reproductive analyses of fecundity, size of maturity, seasonality of reproduction, and sex ratios. If sample sizes are inadequate, supplemental samples will be collected. A companion effort to sample larval fish in the wetlands will provide data for recruitment patterns and effort. The preserved collections will be supplemented during routine sampling five times per year by fresh collections of fishes, which will be frozen or preserved in alcohol for otolith (earbone) analysis. Estimates of reproductive biology will follow Trexler (1985) for livebearing species, and Heins et al. (1986) and Heins (1995) for egg-laying species.
Data from several years are needed because each year in the Everglades can be idiosyncratic. Field data will be collected for a minimum of two years, and supplemented with previously collected samples to expand the temporal scope of the study.
Non-indigenous species: Spotted tilapia (Tilapia mariae) - an abundant and widespread species.
Samples of these species from canal habitats, where they are very abundant at drier times of year, will be made by Nico during periodic electrofishing trips to south Florida from Gainesville. Specimens will be preserved in formalin for gonadal examination, with a second sample will be saved in ethanol to avoid damage to the calcium-rich hard tissues used for aging. We anticipate that it will be more difficult to collect adequate sample sizes of large fish at high-water times when they disperse from canals into wetlands. Dr. Nico plans to put in more effort at those times, and to supplement his collections with previously preserved specimens from archived existing collections. Collections will be made quarterly throughout the year. The analytical procedures will be similar to those described for the small-bodied species, except that none of the large species appear to be fractional spawners. This makes the estimate of lifetime fecundity somewhat less problematic. For these long-lived species, we do not think it feasible to use daily otolith rings for aging because it becomes difficult to do so because the rings become indistinct. We will investigate this further in the pilot study, but may follow the usual procedure for large fishes of reading annular rings. With that, of course, we will have to validate the timing and frequency of "annualar" ring formation through marking of captive fishes.
Potential impacts and major products: USGS-BRD and its cooperators are using a system of empirical data collection and simulation modeling to apply information on fish-community patterns in guiding the restoration process. We are describing the functional responses of present-day community density and standing stock to environmental conditions across the landscape. The empirical data also help to build simulation models with which to estimate fish-community parameters under modeled pre-management conditions, allowing us to establish restoration targets. By applying the fish models to restoration alternatives and predicting fish-community responses, we can choose the alternatives that result in biotic characteristics that approximate historical conditions. The iterative process of evaluating and testing the fish-community simulation model in ATLSS also helps identify important data gaps to guide future research. One of the most obvious gaps is the absence of good life-history data, critical to model performance, for most of the fishes. The benefits to restoration would arise by having more confidence in improved tools, like the ATLSS models, that are used to evaluate alternatives for ecological effects of the Central and Southern Florida Project Restudy, C-111 Project, and Modified Water Deliveries Plan to Shark Slough.
In addition to the application of the life-history data to modeling and to interpretation of the data time-series, these data represent new information about the adaptations of many of these species in wetland habitats that form the southern extent of their geographic ranges. These also represent the first life-history data for some of the most abundant introduced species in Florida, and may identify vulnerable life stages for controlling these species. The publications resulting from this work will be scientifically significant for those reasons.
The benefits to restoration include more confidence in improved tools, like the ATLSS models, that are used to evaluate alternatives for ecological effects of the Central and Southern Florida Project Restudy, C-111 Project, and Modified Water Deliveries Plan to Shark Slough. At present the life history information built in to the simulation model for aquatic animals is incomplete and needs supplementation with better data for more species. A number of Performance Measures and Success Criteria have been developed to assess Restudy actions. These Measures are referable to the Conceptual Models developed for the Restudy to illustrate the stressors and response variables in each major habitat that would be affected by restoration actions. Data from the present study will provide the information needed to examine the responses of aquatic communities to restored hydrological patterns and other environmental parameters. The data, and the models that incorporate them, should also help explain the community patterns of recruitment observed in empirical data collections, and better predict future patterns resulting from restoration actions.
Hypothesis: Hydroperiod of a wetland affects the growth pattern and reproductive output of female small-bodied marsh fishes. We are testing the idea that this effect will be manifested by differences in the size-to-age relationship, life expectancy, and fecundity of female fishes in long- versus short-hydroperiod sites. Our work plan includes gathering data to test for these hypothesized differences in one set of small-bodied species in the first project year and a second set in the second year (FY-00). We are also working with five species of large-bodied fishes collected at a southern Everglades canal site that provides access to the marsh in the wet season and refuge from drying the rest of the year. Our data will demonstrate this seasonal use of canals in the Everglades system and the relation of reproductive phenology and growth to that seasonal pattern. As with the small-bodied species, we are studying three species in the first year, continuing with two additional species in FY-00.
In the third and final year of the project, we will complete the work with species collected during the two previous years and analyze the complete data sets for large and small-bodied fishes. We are requesting continued support personnel as graduate students at FIU, and FCSC technicians, to complete the work in this final year.
In the third year, we will complete the laboratory analysis of additional and existing samples to assess the biology of the flagfish and bluefin killifish, and of the spotted tilapia and yellow bullhead. The life-histories of the flagfish and bluefin killifish will be characterized from existing samples collected at two sites in Everglades National Park that were selected in project year 1. The locations represent different hydroperiod conditions. The analysis of previously collected samples of other large- and small-bodied species will be completed and the results written for publication
Age-Size relationship: Fishes will have been collected two times at these two sites for otolith analysis. The number of fish examined will have been determined by the pilot study, but will be consistent with the methods of Haake and Dean (1983). Maximum life expectancy will be estimated by the estimating the ages of large specimens of those species from our collections. In the large-fish segment, we will analyze the results of our work with Florida gar and warmouth sunfish, while completing the work begun in year two with spotted tilapia and yellow bullhead aging. Statistical analysis of the data will begin once all data have been collected and error-checked.
Reproductive Biology: Our estimates of reproductive phenology and fecundity in egg-laying fishes have required the dissection of female fish to determine the number and developmental stage of ova. In addition, we have been recording the weight of female fish and of their ovaries for calculations of reproductive allocation. Female flagfish are batch spawners that produce a brood of eggs in discrete spawning events, while bluefin killifish are fractional spawners that release a small number of eggs on a routine basis. These two different modes of reproduction have posed challenges to life- history description. The time interval between breeding bouts and the relationship between egg counts from preserved females and actual number of eggs laid must be determined for flagfish. For bluefin killifish, we must estimate the daily egg production, and the distribution of female reproductive lifespans. To satisfactorily obtain these parameters for these two fishes, we are testing the need and feasibility of rearing these fishes in mesocosm tanks and harvesting eggs from nests (flagfish) or artificial spawning substrates (bluefin killifish). Techniques will follow those proposed by Heins (1995). We feel that we can achieve success in this work because both species are kept in aquarium tanks by hobbyists who routine spawn them. In the large-fish segment, we will perform analysis of reproduction with spotted tilapia and yellow bullhead this year by dissection, gonadal examination, and preparation of a gonadal-somatic index for each species. We will complete similar work begun in previous years with Florida gar, spotted sunfish, and warmouth.
PRODUCTS At the end of this final project year, we anticipate producing manuscripts destined for publication on the reproduction and phenology of large-bodied fishes associated with canals, and the same for the small-bodied marsh fishes. We plan separate publications resulting from this study that will report the age and growth data for the large- and small-bodied fishes in canals and marshes, with relation to habitat seasonality. We will compare our results to the patterns reported for these species in other southeastern US waters, and to patterns identified for fishes that inhabit other seasonal wetlands.
The two graduate students will produce Masters Theses, based on this work. Because the work leading to a thesis requires a minimum of 2.5 years, one of these products will appear in the middle of this final year, the other after the end of the project. We will produce yearly updates of our project Fact Sheets. The final data and model rules resulting from the collections will be fed into existing ATLSS simulation models to improve their performance and credibility.
FY-2001 Outreach We and the graduate students plan to attend meetings of scientific professional societies to present results of the study. We will also participate in relevant local and regional meeting related to Everglades restoration. These data are also being sought by the team leaders for RESTORE who wish to use them to refine Performance Measures during the restoration effort. We plan to work closely with that team to make our data and insight s available, even before the final products are ready. We plan to deliver a seminar at Everglades National Park to inform clients of research results and implications. We will also attend the annual meeting of the Greater Everglades Restoration Science Forum.
FY-2001 New Directions No new directions are planned or required.