An open-population distance sampling framework for assessing population dynamics in group-dwelling species

1. Effective wildlife management and conservation depends on a detailed understanding of population dynamics at large spatial scales. Distance sampling is an efficient tool for landscape-scale monitoring, and the recent development of the open-population distance sampling framework for analysis allows efficient estimation of population parameters for unmarked populations. 2. Applications of open-population distance approaches are relatively rare, and further development is required to generalize this powerful class of models for use in complex sampling scenarios. 3. Here, we extend this class of models to situations where individuals occur in groups and address: the incorporation of group composition information, fluctuations in aggregation among individuals into groups, and non-repeated survey designs. We also describe a new parameter, kappa, representing the change in the rate of aggregation of individuals into groups over time. Using simulations, we assess the performance of this extended framework, and demonstrate its utility through an applied example using a Dall's sheep (Ovis dalli dalli) dataset collected in Alaska from 2009 to 2016. 4. Our simulation study demonstrated that our extended model reliably recovered the data-generating parameter values without bias. In the applied example, we found that an abrupt decline in abundance observed in the Dall's sheep population was driven primarily by reductions in lambs and ewe-like sheep that coincided with the late onset of spring conditions in 2013. Rams were much less affected, suggesting declines in the ewe-like class drove overall population trajectory in this population. 5. This work demonstrates how ecological questions related to the dynamics of group-dwelling species may be assessed in complex sampling situations using an open-population distance sampling framework. The resulting framework is broadly applicable to a variety of species and will be especially useful for evaluating population dynamics in unmarked populations of group-dwelling species at large spatial scales.