In this seminar Res Altwegg, of the Department of Statistics at UCT will speak on some of the latest research that his collaborators and he carried out about modelling species distributions when detection is not perfect.

Some of the most pressing problems in nature conservation play out at large geographic scales, which has led to the new field of Conservation Biogeography.

At this scale, incomplete detection and the difficulty of studying dynamic processes are two key issues. We used dynamic occupancy models adapted to bird atlas data for analysing patterns and processes in the distributions of bird species in southern Africa between 1987 and 2015. We found both clear legacies of past climatic conditions in the current species assemblages, and evidence that species ranges dynamically changed in apparent reaction to recent bush encroachment and changes in land use. Range changes led to opportunities for new ecological and evolutionary interactions between brood parasites and their host species. We also found evidence that pairs of closely related species competitively limit the range of one another. Finally, we found that protected areas are critical for the persistence of some species. Our work shows the value of citizen science projects for our understanding of large-scale biodiversity processes but also highlights the need to account for the variable observation process that affects such data sets.