Studying spatial interactions between sympatric populations of large herbivores: a null model approach

Sympatric populations of species with similar ecology are limited by competition for available resources. While quantifying niche overlap between species in interaction offers a useful description of coexistence patterns, the lack of correspondence between niche overlap and competition prevents any functional interpretation. Using an innovative approach for analysing spatial distributions of individuals from two sympatric species, we aim to fill the gap. We applied our models to data collected on sympatric females of roe deer and red deer. Using the null model approach commonly applied in community ecology, we tested in a first model for deviation from a random distribution of female roe deer in relation to female red deer. We took into account constraints generated by both the marked sedentary habits and habitat use (avoidance of mature forest) of roe deer in this null model. In a second null model, we removed the habitat constraints to avoid any lack of power of our tests. We then compared the overlap index calculated from roe deer and red deer locations with the distribution expected under each of these null models. As we failed to reject the null model in both cases, we tested a third null model simulating an identical distribution of roe deer and red deer home ranges and we rejected it. Our results show that the distribution of female roe deer does not depend on the distribution of female red deer, indicating an absence of competitive interactions between the deer species. This conclusion relies on the application of the null model approach, which provides a suitable way of performing a formal test of interspecific competition rooted in explicitly defined hypotheses, and could not have been reached using simple overlap indices as generally performed when assessing competitive interactions. We thus encourage scientists to apply this null model analysis to population ecology.