PARENT SESSION
Tuesday, August 8, 8:00-11:30 am
COS 16 - Disturbance ecology
L-2, Lobby Level, Cook Convention Center
Presiders: S Pathikonda and S Riffell

Do bird-habitat models accurately predict effects of large-scale disturbance?

Seavy, Nathaniel^{*,1, 2}, Alexander, John¹, ¹ University of Florida, Gainesville, FL² Klamath Bird Observatory, Ashland, OR

ABSTRACT- Species distribution models describe the probability of occurrence as a function of habitat characteristics. Usually based on correlational studies, these models are widely used to guide management decisions. However, relatively few studies have tested the ability of these models to accurately predict effects of large-scale disturbances, such as fire. With a data set of nearly a thousand bird and vegetation surveys in a southern Oregon watershed, we used logistic regression to model the probability of occurrence of bird species as a function vegetation composition (broadleaf-conifer composition) and structure (total vegetation volume). We then tested these models using a before/after comparison of a large portion of our study area that was burned by wildfire. Our ability to predict differences between burned and unburned areas was poor immediately after the fire, but steadily improved over the 4 year study period. These results suggest that bird-habitat models are useful tools for predicting effects of large-scale disturbances, but their predictive power may vary as a function of time-since disturbance.

Key words: bird-habitat models, disturbance ecology, model validation