USGS - science for a changing world

South Florida Information Access (SOFIA)

projects > across trophic level system simulation (atlss) > snail kite >abstract

Simulation Model of the Snail Kite

By: Wolf M. Mooij and Donald L. DeAngelis

The snail kite (Rostrhamus sociabilis plumbeus Ridgway) is an endangered raptor whose North American distribution is limited to the freshwater marshes of southern and central Florida. The viability of the snail kite population is strongly linked with the hydrology of the system, due to the kite's almost total dependence on apple snails, which require a freshwater habitat. The number of habitat sites is limited and geographically separated. These sites are subject to both natural and anthropogenic drydowns, which can depress apple snail availability and affect the snail kite population.

To attempt to predict the effects of water regulation strategies across the snail kite's range on the population, an individual-based model has been developed for the population as part of the "Across Trophic Level System Simulation" (ATLSS) package of models. The model aims at predicting the viability of the snail kite population under a range of hydrologic scenarios. It is hypothesized that this viability depends critically on the frequency of droughts, but also on the spatial extent of these droughts. A systemwide drought is likely to result in increased mortality, whereas the birds can respond to a local drought by migration. The model allows one to study the effects of both drought frequency at local habitat sites and the amount of correlation of droughts among the sites.

Fourteen of the major wetlands of southern and central Florida that are inhabited by snail kites are discriminated in the model. An additional spatial unit, called peripheral habitat, is created to mimic the scattered pieces of wetlands that provide some suitable habitat for kites, though not of sufficient quality for nesting. Each kite in the population is modeled as an individual, and at a given moment in time each is located in one of these fifteen spatial units. Kites can move from wetlands to wetlands throughout their life cycles. The historical water levels in each wetlands are estimated based on water stage gauges. Possible future water levels under various water regulation scenarios are forecast using hydrologic models. The habitat quality of each wetlands is modeled on the basis of the standardized water levels. When water levels drop below a threshold, L low, the state of the wetlands changes from high to low. When water levels drop below a second, lower threshold, L elow, the state of a wetlands changes from low to elow. The state of a wetlands changes from low to high after a time lag, t low, starting from the moment that the water levels exceed L low. From elow the system moves to a state lag after a time lag t elow, starting from the moment that the water levels exceed L elow. From lag the state changes back to high after a time lag t lag. Besides direct effects of hydrologic conditions on the kite demographics via relative habitat quality, the concept of density dependence is entered by assigning a carrying capacity to each of the 14 wetlands and the peripheral habitat. An estimate of degradation of the wetlands through prolonged continuous flooding is also included in the model.

Each snail kite in the population is simulated on a weekly basis. In a single week a kite may experience such activities as nest initiation, nest failure, successful fledging, movement to another wetland site, or mortality. Each kite goes through a fixed set of life stages. These life stages affect the probabilities with which the processes of breeding, movement, and mortality occur. The nest period is divided into three periods: the incubation stage of 4 weeks, the nestling one stage of 2 weeks, and the nestling two stage of 3 weeks, following which fledging occurs. After that there are fledgling, juvenile, subadult, adult, and senile stages. The maximum age is 20 years. Based on empirical data, parameters are assigned as probability distributions for fecundity, nest initiation, nest failure, movement to a new wetland site, and mortality.

The model shows that high drought frequencies lead to reduced numbers of kites. The effect of habitat degradation after a prolonged period of flooding, however, had no effect within the range of dry-down intervals that was studied, though it certainly would if these intervals were increased sufficiently. The most interesting aspect of the model is that it allows for the evaluation of spatial correlation between droughts. When the spatial correlation between droughts is now, the model shows narrow ranges of predicted numbers of kites. When droughts occur mostly on a systemwide level, the effect of environmental stochasticity strongly increases the unpredictability of the future numbers of kites and in the worst case the viability of the kite population is seriously threatened.

A significant part of the funding for this research was provided from the U.S. Department of the Interior South Florida Ecosystem Restoration Program "Critical Ecosystems Studies Initiative" (administered through the National Park Service) and from the U.S. Geological Survey, Florida Caribbean Science Center. Additional funding for the "Atlas Tropic Level System Simulation" was also provided by the U.S. Environmental Protection Agency and the U.S. Army Corps of Engineers.

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

Back to Project Homepage

U.S. Department of the Interior, U.S. Geological Survey, Center for Coastal Geology
This page is:
Comments and suggestions? Contact: Heather Henkel - Webmaster
Last updated: 04 September, 2013 @ 02:08 PM (TJE)