A joint model of species interaction dynamics for multistate processes incorporating imperfect detection

The dynamics of species interactions are critical to understanding community ecology. Although species co‐occurrence models have long been available, recent methodological advances support extensions to incorporate temporal population dynamics with variation in the observation process. Contemporary dynamic co‐occurrence models often focus on simple presence/absence despite evidence that the influence of interactions may extend beyond species presence to behaviors influencing fitness such as mating success or recruitment. Here, we propose a joint model of species interaction dynamics that extends to multistate processes for each species, while limiting the attendant combinatorial expansion in the number of parameters. In our model, one or more states of a subordinate species are assumed to depend on the state of a dominant species, but a reciprocal dependency does not exist. This model results in a more concise representation of dominant‐subordinate species interactions compared with other approaches, which should aid in model setup, fitting, and interpretation. The basic framework we outline can be fit in either Bayesian or maximum‐likelihood modes of inference, can include interaction lag effects, and deals efficiently with cases where data are missing for one species but not the other. We evaluate interactions between the northern spotted owl ( Strix occidentalis caurina ) and barred owl ( Strix varia )—the latter species is considered to be dominant—using long‐term empirical data collected in Oregon, USA . Our results suggest state‐specific interactions between the two species where barred owls are associated with declines in spotted owl pairings, but not necessarily occupancy.