Open AccessBuilding better wildlife‐habitat models
Author(s) -
Beutel T. S.,
Beeton R. J. S.,
Baxter G. S.
Publication year - 1999
Publication title -
ecography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.973
H-Index - 128
eISSN - 1600-0587
pISSN - 0906-7590
DOI - 10.1111/j.1600-0587.1999.tb00471.x
Subject(s) - habitat , wildlife , ecology , proxy (statistics) , wildlife management , abundance (ecology) , quality (philosophy) , geography , environmental resource management , environmental science , computer science , biology , machine learning , philosophy , epistemology
Wildlife‐habitat models are an important tool in wildlife management today, and by far the majority of these predict aspects of species distribution (abundance or presence) as a proxy measure of habitat quality. Unfortunately, few are tested on independent data., and of those that are, few show useful predictive skill. We demonstrate that six critical assumptions underlie distribution based wildlife‐habitat models, all of which must be valid for the model to predict habitat quality. We outline these assumptions in a meta‐model, and discuss methods for their validation. Even where all six assumptions show a high level of validity, there is still a strong likelihood that the model will not predict habitat quality. However, the meta‐model does suggest habitat quality can be predicted more accurately if distributional data are ignored, and variables more indicative of habitat quality are modelled instead.