Certain studies needed to understand the fish community are difficult to carry out in the field, yet require more space than the confines of laboratory aquaria. Mesocosm experiments provide information on feedback mechanisms among trophic levels in food webs that can play a central role in efforts to model community dynamics. These effects would be very difficult to identify in typical correlative field studies. An experimental mesocosm constructed in Everglades National Park allows us to address questions of predator-prey, competitive, and indirect interactions. The mesocosm is also suited to address questions about the role of nutrient inputs into the Everglades in causing shifts in marsh food webs. The mesocosm also provides the opportunity to examine introduced/native fish interactions, and to test potential control methods.

One of the primary unknowns for modeling the dynamics of fish communities in the Everglades is the identification of factors that limit population growth and community dynamics. Two hypotheses are now debated in the literature regarding the limitation of population dynamics of small-bodied species: (1) that predator-prey interactions limit the population growth, or (2) that food limitation sets bounds on the size of populations. The mesocosm will complement field studies by providing opportunities for experimental studies under more controlled conditions than possible in field conditions, but at a larger scale and with more fieldlike conditions than in laboratory aquaria.

Three types of studies are planned for the mesocosms on the 5-year horizon projected for this funding. They are the study of (1) intraguild predation with mosquitofish, (2) the effects of predation by mosquitofish and bluefin killifish on nesting successes of sunfish, and (3) the effects of nutrient level and primary production on population growth of small-sized fish. Those projects are described in the original proposal. Their results will fit closely with completed and continuing research describing Everglades aquatic communities through field sampling, planned field caging studies on grazing ecology along hydrological gradients, and field nutrient manipulation in the dosing flumes. The ultimate goal of this research is to develop a model of ecological regulation across trophic levels for Everglades environments. Our first experiment examines size-structured predator-prey and competitive interactions involving mosquitofish, sailfin molly, and bluefin killifish. We selected this experiment for two reasons: (1) it will provide important information about trophic interactions for the "Across Trophic Level System Simulation" (ATLSS) model, and (2) it will permit us to work out logistical issues for conducting experiments in the mesocosms while gathering useful data. Mosquitofish were chosen as the focus of our first experiment because they are ubiquitous, and are notorious as a predator of larval and juvenile fish and invertebrates. Mosquitofish are omnivores, and so may regulate the growth of periphyton directly, or influence it indirectly by feeding on other grazers. Mosquitofish may also regulate their abundance somewhat through cannibalism. These interrelationships need to be incorporated in a fish communitywide ATLSS model.

Most of this first study has been completed and the preliminary analyses revealed the strong effects that mosquitofish may have on fish community regulation in the Everglades. We demonstrated that this facility will be successful in performing complex factorial experiments. The results showed that sailfin molly and bluefin killifish adults do not prey on juvenile fish, but mosquitofish exert strong predatory influence on small specimens. Because of its abundance in the marsh, mosquitofish may perform a regulatory role on the small-fish assemblage. The experiment also showed that food limitation might affect population dynamics in the community. Slower growth of juvenile fish because of shared food resources may slow growth and increase predation risk.

Sunfish nesting mortality: The next experiment that we are beginning examines the potential effect of small-fish predation on nests and larvae of sunfishes. Preliminary data indicate that sunfish nesting in Everglades National Park experience different rates of nest predation depending on water depth. The hard substrates needed for nesting are limited in there and, so, sunfish nest in approximately the same areas each year. The depth of water at the nesting sites is a function of the rainfall that year. In years of low water, the surface dwelling mosquitofish occur in near proximity to the sunfish nests. At such times, the number of nest attacks per unit time increases, causing the male nest guarder to spend more time chasing predators away (Loftus, personal observation). We have begun by having a graduate student learn how to breed and raise the sunfish in captivity. We will explore how water depth affects nesting success by examining nesting success of spotted sunfish in tanks, with mosquito fish, bluefin killifish, and juvenile sunfishes as potential predators, at different water depths. We will census the number of eggs per nest, number of larvae hatched, and number of juveniles produced. Each tank will be set up with standardized periphyton cover and invertebrate communities such that alternative prey for mosquitofish is present, as in natural conditions. Additionally, we will perform nest observations in the field to estimate predation rates there at different water depths.

Funding for this project was provided by the the U.S. Department of Interior, South Florida Restoration Program "Critical Ecosystems Studies Initiative" (administered by the National Park Service); and, in part, from the U.S. Geological Survey, Florida Caribbean Science Center and Everglades National Park.

REFERENCE

Taylor, R., Trexler, J.C., and Loftus, W.F., 1998, Experimental studies of population growth and predator-prey interactions of fishes in the Everglades National Park: Annual Project Report to the U.S. Geological Survey, FCSC for Coop. Ag. 1445-CA09-95-0112 from FIU.

(This abstract was taken from the Proceedings of the South Florida Restoration Science Forum Open File Report)