Abstract
More widespread species tend to be more locally abundant. This hypothesis has received support when considering single species abundance (mean density of individuals across sites) and occupancy (fraction of occupied sites) through time (intraspecific relationship) and comparing different species sampled at a single point in time (interspecific relationships). But while abundance‐occupancy relationships are fairly well supported in observational studies, the underlying factors driving them are less clear. For instance, variation in demographic rates, dispersal, and spatial network structure could all influence resulting abundance‐occupancy relationships. We used a simulation model to explore these relationships in spatial networks of variable size and dispersal connectivity. We simulated population dynamics on spatial networks by starting from entirely neutral communities and then systematically incorporated complexity in the form of (co)variation in species demographic rates and dispersal processes. The effect of spatial network structure on abundance‐occupancy relationships was dependent on the community dynamics and the covariation imposed on demographic and dispersal parameters. Together, we demonstrate the interplay between the spatial network and variation in demographic and dispersal rates, generating testable hypotheses for when abundance‐occupancy relationships would be more likely to be observed, as well as how these relationships may change with habitat fragmentation and shifts in community composition.