Using a species distribution model to guide NSW surveys of the long-footed potoroo (Potorous Longipes)

Current knowledge on threatened species' distributions is essential for effective conservation decision-making. Species distribution models (SDM) are widely used to map species geographic ranges and identify new areas of suitable habitat. This paper uses a SDM to identify regions in NSW that may have suitable habitat for the long-footed potoroo (Potorous longipes) and guide the selection of field survey sites. In NSW, there are grave doubts surrounding the persistence of this critically endangered species and identification of occupied sites is a high priority for its conservation. The SDM, Maxent, had strong predictive performance (AUC: 0.94 to 0.95) and enabled identification of new areas of climatically suitable habitat, beyond areas of known occurrence in NSW and prior survey locations. Importantly, ground-validation of the SDM output was undertaken and showed that projected habitat suitability values were: a) significantly higher at independent presence locations than absence locations (H=55.61, DF=1, P=0.000); and b) correlated with six out of ten microhabitat variables. However, baited camera trapping, undertaken at 58 sites in NSW, did not detect any long-footed potoroos. Refinement of binary regression models found that the combination of connectivity, i.e. larger, connected areas of climatically suitable habitat (χ2=5.51, P=0.019), understorey cover (χ2=6.86, P=0.009) and soil moisture (χ2=7.6, P=0.006) best predicted this species presence. If the long-footed potoroo remains extant in NSW, it is extremely rare. The findings indicate that, in addition to climatic factors, microhabitat features and connectivity are important predictors of presence of the long-footed potoroo and should be incorporated into any future distribution modelling and survey site selection.