Genetic inference as a method for modelling occurrence: a viable alternative to visual surveys

Abstract

Management and conservation require a comprehensive understanding of species distributions and habitat requirements. Reliable species occurrence data are critical in the face of climate change and other anthropogenic activity, but are often difficult to obtain, particularly for wide ranging species. This directly affects ecological models of occurrence and habitat suitability and, in turn, conservation and management decisions. We used generalized linear mixed-effects models to identify ecological determinants of occurrence for four macropod species (across a region of tropical northern Australia) using a non-invasive genetic scat approach with and without additional observation records from visual surveys. We show that genetically derived occurrence data, alone, can be used to develop informative ecological models that describe the inter-specific habitat requirements of macropods. Furthermore, we show that genetic scat surveys of macropods are cheaper and less time consuming to conduct, and tend to provide more occurrence records (and less false absences) than visual surveys. We conclude that indirect surveys using molecular approaches have an important role to play in modelling species' occurrence, and developing future management practices and guidelines to aid species conservation.