Predictive modeling of microhabitats for endemic birds in South Chilean temperate forests using Maximum entropy (Maxent)

Temperate forests of Chile exhibit high biodiversity, which generates a wide range of habitats for wildlife. These valuable natural ecosystems have been affected by major natural and anthropogenic processes that have reduced habitats, resulting in serious ecological problems, given both the high endemism of certain avian groups in these forests and the complexity of their habitat selection. Continued degradation and ecosystem problems could lead to the extinction of such groups. In spite of this possibility, ecologically valuable wildlife conservation is seldom integrated into forest management decision-making processes. This study aims to integrate wildlife into forest management, identifying potential habitats for two endemic birds of high ecological value, the Black throated Huet-Huet (Pteroptochos tamii), and the Ochre-flanked Tapaculo (Eugralla paradoxa). Both species inhabit an ecotonal area between evergreen and sclerophyllous forests, making them high-quality bio-indicator species for the degree of conservation of temperate forest. The integration of environmental information and a geostatistical model based on the criterion of maximum entropy (Maxent model) identifies the most important variables that explain the presence of each species. Pteroptochos tamii is less restrictive in its choice of habitat than Eugralla paradoxa, requiring merely certain topographical condition (elevation, ground slope and aspect). However Eugralla paradoxa requires not only the same topographical features, but also eco-geographical characteristics such as distance to trails, waterways and ecotones. Maxent analysis showed that for both species, the model most capable of predicting their choice of microhabitat was not random based, but rather one based on topographical and environmental variables. The integration of Maxent and Geographic Information Systems (GIS) tools could help to solve problems of wildlife habitat conservation and forest planning. (C) 2011 Elsevier B.V. All rights reserved.