Presence-Only Species Distribution Models to Predict Suitability Over a Long-Term Study for a Species with a Growing Population
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
The Wildlife Society, 2015. We used a long-term data set of black-capped vireo (Vireo atricapilla) detection locations collected between May and June of 1990 through 2005 within the Wichita Mountains of Oklahoma, USA, to determine how presence-only distribution models change as a population expands to help focus conservation and management activities. We used MaxEnt to model black-capped vireo habitat suitability for each of 4 years using detection data and several remotely sensed habitat metrics, including soil type, slope, and elevation. We assessed how well each model fit the detection data, what metrics vireos were using, and how well each model predicted occupancy across time. We found that as the vireo population increased, vireos began occupying a wider range of landscape characteristics. As a result, the models predicted more area with high suitability as the population grew. Similarly, we found that a model's ability to predict vireo occupancy in future years decreased with time, with low predictability even 5 years out. The combination of a secluded study site and specific aspects of this species' behavior likely accounted for the poor performance of our models at predicting areas into which vireos would expand over time. We conclude that habitat models should consider population status and be used with caution to predict areas of future occupancy if the population is currently expanding.
