Accounting for connectivity alters the apparent roles of spatial and environmental processes on metacommunity assembly

Context Understanding the relative contributions of spatial and environmental processes on community assembly is a central question in ecology. Despite this long-standing interest, our understanding of how landscape structure may drive spatial processes of community assembly remains poorly understood in part because of the challenge of tracking community assembly across landscapes and quantifying key aspects of landscapes that may impact assembly processes. Objectives We examined the roles of spatial and environmental processes on structuring assemblies of ants in 72 cleared patches embedded within a forested landscape. Methods To examine the role of spatial processes, we contrasted the use of geographic distances between patches and effective distances estimated from connectivity modeling accounting for matrix vegetation structure hypothesized to be important for ant community assembly. To examine the role of environmental processes, we quantified patch age and abundance of a key competitor and invasive species, the fire ant Solenopsis invicta. Results We found evidence for the importance of both spatial and environmental processes in structuring ant communities. When spatial processes were quantified as geographic distance, environmental variables were the predominant factors accounting for variation in ant community dissimilarity among patches. However, accounting for matrix resistance with circuit-theoretic connectivity modeling resulted in higher accounting of variation in ant community dissimilarity than geographic distance and changed the predominant variables accounting for that variation from environmental to spatial processes. Conclusions These findings show that accounting for connectivity through the matrix can be decisive in determining the primary drivers of community assembly.

Automated summary

The study found that both spatial and environmental processes play a role in structuring ant communities. When considering landscape structure, spatial processes had a greater impact on ant community dissimilarity compared to environmental variables.