The Florida panther (Felis concolor coryii) represents one of the most challenging problems in conservation. This population is the remnant of a population that once occurred throughout the Gulf and Atlantic coastal plains of southeastern United States. The total population of panthers in southern Florida probably includes no more than 50 adults, and this subspecies was listed as endangered in 1973. The Everglades/Big Cypress system is not high quality panther habitat compared with upland areas farther north, but its relative freedom from human impact makes it currently the only available area for the panther population. The question is whether this area is adequate to sustain a population of panthers in the long run.

Models can be used to help determine the carrying capacities of various landscapes within southern Florida for the panther, and assess the usefulness of protection of land not currently protected that may increase the overall carrying capacity. Therefore, as part of the ATLSS Program, a model has been developed that will allow us to determine the potential long-term impacts (for example, over 30 years or more) of spatially-explicit modifications in habitat (particularly hydrology) on the panther population (Comiskey and others, in press). The spatio-temporal dynamics of habitat coupled with small population size and long-distance movements of panthers, implies that an individual-based modeling approach offers the best hope to utilize the extensive available empirical data on panthers to produce a model which appropriately tracks the effects of alternative hydrologic scenarios. Additionally, such a model offers the capability of providing cost-effective methods to help guide any potential captive release program, by comparing the effect of alternative release programs on the model population.

Individual panther success (for example, survival and reproduction) in South Florida is closely linked to a panther's ability to obtain large prey items, notably white-tailed deer (Odocoileus virginianus seminolus) and feral hogs. Thus, in order to produce a panther model, it was essential to develop an individual based model of white-tailed deer to determine how hydrologic changes would affect these key prey resources, and link this to the panther model. White-tailed deer, the only large (native) herbivore in the region, has a significant localized impact on vegetation, so including it as a major component of the project is justified independent of the importance of deer to panthers. Accounting for the impacts on vegetation of feral hogs in the system would also be appropriate, however at present hogs are represented only as a static density. The main objective of this modeling component is to allow for prediction of the relative impacts of alternative hydrologic scenarios over a 30-year timeframe on the spatial and temporal (for example, seasonal) distribution of panther and deer across South Florida, and to produce relative comparisons of mortality, reproduction, individual movement patterns and territory size across the landscape for both species.

The model operates on a daily time step, although within this time step, deer and panther movement are simulated, taking account of local water conditions, forage and prey availability. Spatially, the model makes use of vegetation data to calculate forage availability on a 100 m scale, but tracks deer and panther locations on the daily time step at 500 m scale.

The primary inputs to the Florida panther/white-tailed deer model are daily hydrology data at the 100 m scale (available through a pseudotopography model, see below), a vegetation map (from which three classes of dynamic forage maps are constructed), a landuse map, a map of feral hog density, and a road map. The primary environmental factors driving the model are, therefore, hydrology and vegetation, which vary temporally.

The first version of this model is now complete and produces highly detailed spatial information on deer and panther distribution pattern changes over time that appear reasonable based upon consultation with experts. Validation efforts include detailed comparisons of deer distributions with historical data, comparison of aggregated variables such as age-dependent mortality, age-structure, body weight distribution and birth rates with available data, and comparison of model individual-movement patterns with radio collar data.

Back to Project Homepage