Predicting the distribution of Tawny Owl (Strix aluco) at the scale of individual territories in Denmark

Abstract

Wildlife management strategies can be improved by accurate predictions of species distributions and sound estimates of population sizes. Species distribution models that relate occurrence or abundance of a species to environmental predictors can be applied for both purposes. We build a generalized linear model relating the distribution of breeding Tawny Owls (Strix aluco) to remotely sensed environmental data to estimate the current distribution of breeding territories throughout Denmark. Additionally, we apply three different methods to calculate population estimates from the model predictions. The most parsimonious model includes only coverage of deciduous forest and its squared effect as significant predictors. Correspondence between predictions of breeding territories and atlas data on breeding Tawny Owls is high (73%), indicating that our model captures an essential part of this species’ habitat requirements. The three population estimates suggests the number of breeding pairs in Denmark to be ~4,800, ~7,200, or ~20,500, respectively. This case adds to the existing body of literature that illustrates that species distribution models can be useful to predict the distribution of species and for estimating population size.