Image texture predicts avian density and species richness.

Area of Science:

• Ecology
• Remote Sensing
• Wildlife Conservation

Background:

• Traditional ecological field measurements of habitat are limited in scale for broad conservation efforts.
• Remote sensing offers efficient methods for large-area habitat characterization.
• Bridging the gap between fine-grained habitat data and broad-scale inference is crucial for conservation.

Purpose of the Study:

• To evaluate the efficacy of remotely sensed image texture as a predictor of avian density and species richness.
• To compare the predictive power of remote sensing data versus field-measured vegetation structure.
• To assess the utility of remote sensing for mapping habitat quality across grassland, savanna, and woodland ecosystems.

Main Methods:

• Calculated image texture from aerial and Landsat TM satellite imagery.
• Measured field vegetation structure (foliage-height diversity, horizontal structure).
• Correlated vegetation structure metrics with avian density and species richness in different habitat types.

Main Results:

• Air photo image texture best predicted grassland bird density (grasshopper sparrow) and overall avian species richness.
• Satellite-derived NDVI best predicted woodland bird density (ovenbird).
• Remotely sensed vegetation structure often outperformed field measurements in predicting avian metrics.

Conclusions:

• Remotely sensed image texture serves as a valuable surrogate for vegetation structure in ecological studies.
• Remote sensing provides a scalable and effective tool for assessing wildlife habitat quality and biodiversity across large areas.
• This approach holds significant promise for informing conservation and management strategies.