Artigo Acesso aberto Revisado por pares

Use of Maximum Entropy Modeling in Wildlife Research

2009; Multidisciplinary Digital Publishing Institute; Volume: 11; Issue: 4 Linguagem: Inglês

10.3390/e11040854

ISSN

1099-4300

Autores

Roger A. Baldwin,

Tópico(s)

Ecology and Vegetation Dynamics Studies

Resumo

Maximum entropy (Maxent) modeling has great potential for identifying distributions and habitat selection of wildlife given its reliance on only presence locations. Recent studies indicate Maxent is relatively insensitive to spatial errors associated with location data, requires few locations to construct useful models, and performs better than other presence-only modeling approaches. Further advances are needed to better define model thresholds, to test model significance, and to address model selection. Additionally, development of modeling approaches is needed when using repeated sampling of known individuals to assess habitat selection. These advancements would strengthen the utility of Maxent for wildlife research and management.

Referência(s)