Bayesian multi-species N-mixture models for unmarked animal communities

We propose an extension of the N-mixture model that enables the estimation of abundances of multiple species as well as the correlations between them. Our novel multi-species N-mixture model (MNM) is the first to address the estimation of both positive and negative inter-species correlations, which allows us to assess the influence of the abundance of one species on another. We provide extensions that permit the analysis of data with excess of zero counts, and relax the assumption that populations are closed through the incorporation of an autoregressive term in the abundance. Our approach provides a method of quantifying the strength of association between species' population sizes and is of practical use to population and conservation ecologists. We evaluate the performance of the proposed models through simulation experiments in order to examine the accuracy of both model estimates and coverage rates. The results show that the MNM models produce accurate estimates of abundance, inter-species correlations and detection probabilities at a range of sample sizes. The MNM models are applied to avian point data collected as part of the North American Breeding Bird Survey between 2010 and 2019. The results reveal an increase in Bald Eagle abundance in south-eastern Alaska in the decade examined.

Automated summary

We propose a novel multi-species N-mixture model that can estimate the abundances of multiple species and the correlations between them. This model is the first to consider both positive and negative inter-species correlations, allowing us to examine the influence of one species' abundance on another. We validate the accuracy and applicability of our model through simulation experiments and apply it to avian point data from the North American Breeding Bird Survey.