DOI: 10.1371/JOURNAL.PONE.0151024
关键词: Ecology 、 Habitat 、 Life history theory 、 Range (biology) 、 Geography 、 Sample (statistics) 、 Niche 、 Generalized linear model 、 Econometrics 、 Distribution (economics) 、 Climate change 、 General Biochemistry, Genetics and Molecular Biology 、 General Agricultural and Biological Sciences 、 General Medicine
摘要: A primary assumption of environmental niche models (ENMs) is that are both accurate and transferable across geography or time; however, recent work has shown may be but not highly transferable. While some this due to modeling technique, individual species ecologies also underlie phenomenon. Life history traits certainly influence the accuracy predictive ENMs, their impact on model transferability less understood. This study investigated how life ENMs using historically calibrated for birds. In I used historical occurrence climate data (1950-1990s) build a sample birds, then projected them forward ‘future’ (1960-1990s). The were validated against generated from at time. Internal external validation metrics, as well metrics assessing transferability, Generalized Linear Models identify significant predictors transferability. found ability differs with regard characteristics such range, migration, habitat, rarity versus commonness affects predicted stability overlap hence models. Projected high scores, still sometimes suffered over- under- ranges. influenced while aspects geographic range mechanisms underlying