Testing the efficacy of bat monitoring methods for identification and species surveys in KwaZulu-Natal province, South Africa

Abstract

Multi-method sampling approaches are becoming increasingly popular for investigating species occurrence at specific sites, as there is a need to accurately monitor species with limited time and resources. In this study, a multi-method comparative approach was used to survey bat species in the foothills of the Drakensberg Mountain range in KwaZulu-Natal province, South Africa. We used historical museum records and species distribution modelling (SDM) to predict which species would likely occur in our study area. We then compared physical capture (by deploying mist nets) with acoustic surveys (using an Anabat bat detector) to assess the bat species assemblages present. Species distribution models predicted eight bat species to occur from the historical checklist of 28 species recorded in the broader region, as no museum records existed for the specific study area. Species detection by acoustic data yielded the highest number of detected species (n = 11) while active trapping yielded nine species from 54 individuals of four families, namely, Laephotis botswanae, L. capensis, Myotis tricolor, Pipistrellus hesperidus, Rhinolophus clivosus, and Tadarida aegyptiaca with molecular confirmation required for Miniopterus fraterculus, R. darlingi and R. swinnyi. These complementary sampling methods may be necessary for accommodating the limitations of each individual method for a more inclusive assessment of bat species richness in a targeted landscape. The present study could be used as a model approach to assess the biodiversity and demographics of other taxa and in other habitats.