From town to town: Predicting the taxonomic, functional and phylogenetic diversity of birds using NDVI

Abstract

Biodiversity mapping in urban areas is imperative for their conservation. Remote sensors produce environmental information, such as the Normalized Difference Vegetation Index (NDVI), an indicator of vegetation cover in urban areas. NDVI can be used to predict the taxonomic, functional and phylogenetic bird diversity in urban areas. Moreover, a predictive model constructed in one city can be used to predict the bird diversity in other cities. The objectives of this study were: 1) to construct and evaluate predictive models between NDVI and taxonomic, functional and phylogenetic diversity of birds in Mar del Plata city, Argentina; and 2) to extrapolate these models to two other cities in the region: Balcarce and Miramar. Generalized additive models were applied to relate bird diversity variations to NDVI. In Mar del Plata, the taxonomic and functional diversity increased with increasing NDVI values, and the predictive models explained 64–81% of the taxonomic and functional diversity variation. The models correctly predicted taxonomic and functional diversity values in additional transects not included in the models, although they had a low predictive power of phylogenetic diversity. The models constructed in Mar del Plata adequately predicted the spatial variation of species diversity (Shannon index) in Balcarce and Miramar, the spatial variation of species richness in Balcarce, and the variation of functional diversity in Miramar. Our analysis revealed that a predictive model of bird diversity based on NDVI patterns created in one city can also depict the expected species diversity in other cities, being a time-saving and cost-effective method to create a tool for urban biodiversity conservation.