One of the most common applications of statistics in environmental science is species distribution modelling, in which the goal is to model the spatial distribution of a species in some region as a function of environmental variables. Data collected on species come in different types, some of which are:

• Presence-only data, which contain the reported locations where a species is present, with no corresponding absence information
• Survey data, which contain species presences and absences recorded at a set of pre-determined sites
• Occupancy data, which contain species detections and non-detections recorded over multiple visits at a set of pre-determined sites

My expertise is in presence-only data, for which a wide number of methods have been developed for species distribution models. In this talk, I will provide an overview of point process models for this purpose and demonstrate a number of advantages of this approach. I will also discuss some current and challenging projects which extend these methods to modelling the distributions of species for which multiple sources of data are available (for example, presence-only data and occupancy data). Applications of these methods will include modelling the distribution of eucalypts in New South Wales, scarab beetles in France, and the Eurasian lynx in the Jura Mountains of France.