A sparse observation model to quantify species distributions and their overlap in space and time

Camera traps and acoustic recording devices are essential tools to quantify the distribution, abundance behavior of mobile species. Varying detection probabilities among device locations must be accounted for when analyzing such data, which is generally done using occupancy models. We introduce a Bayesian time-dependent observation model camera trap data (Tomcat), suited estimate relative event densities in space time. Tomcat allows learn about environmental requirements daily activity patterns species while accounting imperfect detection. It further implements sparse that deals well will large number potentially highly correlated variables. By integrating both spatial temporal information, we extend notation overlap coefficient between time study niche partitioning. illustrate power through an application eight sympatrically occurring duiker Cephalophinae savanna – rainforest ecotone Central African Republic show most pairs little overlap. Exceptions those one very rare, likely as result direct competition.