COS 124-6 - Using spatially explicit random walk models of animal movement as a tool for the outline of conservation areas

Thursday, August 9, 2012: 9:50 AM
B114, Oregon Convention Center
Bernardo B. S. Niebuhr, Programa de Pós-Graduação em Ecologia e Conservação, Universidade Federal do Paraná, Curitiba, Brazil, Ernesto P. Raposo, Departamento de Física, Universidade Federal de Pernambuco, Recife, Brazil, Gandhimohan M. Viswanathan, Departamento de Física, Universidade Federal de Alagoas, Maceió, Brazil, Marcos G. E. da Luz, Departamento de Física, Universidade Federal do Paraná, Curitiba, Brazil and Marcio R. Pie, Departamento de Zoologia, Universidade Federal do Paraná, Curitiba, Brazil
Background/Question/Methods

Habitat fragmentation is an important issue concerning species management and conservation. On the one hand, the basis for the study and decision making related to the outline of conservation areas comes in great part from island biogeography and metapopulation theories; however, the models derived from these theories generally ignore explicitly animal movement patterns. On the other hand, most research on movement ecology do not consider environments with a heterogeneous distribution of habitat. The aim of this study is to develop a model of animal movement in fragmented environments in order to answer the central question of the SLOSS problem: is a single large habitat patch more efficient than several small ones? The model assumes the animals follow truncated Lévy walks and search for circular habitat patches. The total habitat area remains constant and the number of patches is changed. While traveling among patches, there is a probability of dying that depends on the distance traveled; after finding a patch, an animal stays in its interior for a time proportional to its area. By counting the number of individuals that survive until a time T over many simulations, we aim at finding optimum search strategies in each configuration of habitat fragmentation.

Results/Conclusions

In the search for habitat fragments, strategies that correspond to a smaller value of the Lévy exponent μ were more efficient, once they assume a higher probability of long walks during the search in comparison to Brownian walks, thus minimizing the distances traveled among patches and maximizing the rate of survival. When only the dispersion among patches was focused, for all kinds of search strategy the minimization of the distances walked among fragments emerged from a higher habitat fragmentation (several smaller areas). However, when the time inside the patches was considered, the optimal situations occurred for lower levels of fragmentation (few larger areas). These results suggest that a higher number of habitat fragments could be an advantage for animal survival if only the dispersion processes were taken into account; in spite of that, when one considers more complex dynamics of intra- and inter-specific relations inside and among the patches, a spatial set with a few large patches is rather indicated for the survival and consequently for the management of animal species.