Predicting Mammal Species Richness from Remotely Sensed Data at Different Spatial Scales

ABSTRACT

Spatial variability in species richness has been postulated to depend upon environmental factors such as climatic variability, Net primary productivity and habitat heterogeneity. The Advanced Very High Resolution Radiometer (AVHRR)-Normalized Difference Vegetation Index (NDVI) has been shown to be correlated with climatic variability, Net primary productivity and habitat heterogeneity. Moreover, Landsat Thematic Mapper (TM) derived habitat diversity indices have been used to reflect habitat heterogeneity. Interannually average NDVI and its variability (standard deviation and coefficient of variation) as well as Landsat Thematic Mapper derived habitat diversity index were correlated with mammal species richness at landscape scale. Species richness related unimodally to interannual average NDVI and positively to variability of NDVI and habitat diversity index. Conversely, at regional scale mammal species richness were correlated with interannually average NDVI and coefficient of variation of NDVI. Species richness related negatively to the latter and positively to interannually average NDVI. Though these relationships are indirect, they apparently operate through the green vegetation cover. Understanding such relationships can help in estimating changes in species richness in response to global climatic change.

Keywords: