Remotely sensed habitat quality index reliably predicts an umbrella species presence but not demographic performance: A case study with open pine forests and red-cockaded woodpeckers

Abstract

Remote sensing data can be a powerful and cost-effective method for determining the extent, composition, and structure of ecosystems across large areas. To use this tool for effective conservation of individual species, we need to test the assumption that remotely sensed habitat indices correspond to both patch suitability (i.e., presence) and value (i.e., demographic rates) for the species reliant on those habitats. We built an open pine habitat quality index (HQI) from remotely sensed spectral data to identify the condition of pine stands. We correlated the HQI with presence, group size, and fledgling production of an avian species (red-cockaded woodpecker, Dryobates (=Picoides) borealis, RCW) associated with longleaf pine (Pinus palustris) using Bayesian logistic regressions and structural equation models. The HQI was correlated with RCW cavity tree and foraging area presence, with 10.8 and 16.0 increased odds of RCW presence with each unit increase in the HQI, respectively. However, the HQI was not correlated with RCW demographic performance. Given that RCWs are an umbrella species and are currently limited by habitat, that the HQI successfully identifies areas of higher quality habitat for RCW is useful for conservation planning. Improving the specificity of the remotely sensed index could improve the link between the HQI and RCW demographic performance, but might limit its application to RCW, instead of the entire open pine endemic community. Our study suggests that linking generalized habitat indices to species habitat suitability is possible and strengthens the justification for their use in cost-effective, large-scale conservation of imperiled ecosystems.

Keywords

• Remote sensing
• Habitat use
• Forest quality
• Ecosystem conservation