The interactions between species glue the community together. From the structure of the network of such interactions we can infer community dynamics and stability. However, little is known about how the number and patterning of species interactions scale-up with area. Here we explore how species, interactions, and the resulting network structure change across the landscape. We also explore how such spatial variability is explained by the topological properties of the network of habitat patches. Our approach is based on a metacommunity model, where space is represented by a spatial network of habitat patches exhibiting extinction/colonization dynamics.
Results/Conclusions
As demographic stochasticity increases, not only a larger area is required to record all species and interactions, but such an area increases faster for interactions that for species. We therefore need larger area if we want to preserve not only species, but also interactions. The explanation behind this finding yields on the fact that interactions are more heterogeneously distributed across the landscape. However, it is necessary to account for network structure beyond just counting interactions to address questions such as the stability of local assemblages. Spatial heterogeneity regarding network structure, and the distribution of species and interactions, strongly depends on the structure of the spatial network. The nested structure of the local assemblages changes across the landscape, thus suggesting strong among-patch variability in network robustness.