A spatially explicit indicator for species diversity derived from remote sensing in the mountain rainforest of southern Ecuador


Monitoring species diversity by using field data is both, time-consuming and cost-intensive due to the inaccessibility and extent of most study areas and given the vast number of (often poorly known) species. Ecological theory predicts that the number of co-existing species is related to habitat productivity and heterogeneity. Therefore, habitat proxies derived from remote sensing (RS) have been used to estimate species diversity of selected taxa spatially based on the preference of species for specific habitats. Habitat proxies such as topographical, spectral and textural metrics determine terrain characteristics, the spectral response of the vegetation and its surrounding spatial heterogeneity. In combination with in-situ data this information can be used to up-scale field observations, such as local (‘point’ or plot scale) assessments of species diversity, to landscape or regional scales. The derived maps can be used in conservation planning and action. Thus, RS has a high potential for monitoring biological indicators in space and time. However, the success of RS to estimate measures of biodiversity may vary considerably among taxa. Here, we describe the steps to derive indicators for species diversity using RS data, and exemplarily, show differences in the applicability of RS using three different moth groups of very high local species richness in the San Francisco valley in southern Ecuador

Christine Wallis
Christine Wallis
Postdoc @ TU Berlin

Remote sensing of biodiversity