The role of spatial autocorrelation in native-exotic plant species richness relationships

Spatial autocorrelation in many ecological variables can significantly affect any ecological relationship. Several experimental and observational studies have shown positive or negative relationship between native and exotic plant species richness at different spatial scales. However, how spatial autocorrelation affects these relationships is not yet known. We collected native and nonnative plant species richness data from 180 (0.1 ha) modified-Whittaker plots from Rocky Mountain National Park, Colorado, USA. We used principal coordinates of neighbor matrices (PCNM) approach to quantify broad-scale spatial dependence, and conditional autoregressive regression (CAR) for modeling fine-scale spatial autocorrelation. We used a modified variation partitioning approach to quantify how much of the variation in exotic plant species richness was explained by: (i) native plant species richness; (ii) space (i.e., spatial dependence and fine-scale spatial autocorrelation); and (iii) spatially structured environmental processes. After variation partitioning, we found that native plant species richness accounted only for 9% of the variation in exotic plant species richness; however, pure space, and spatially structured environmental processes explained 34% and 17% of the variation in exotic plant species richness, respectively. Our results suggest that spatial autocorrelation and spatially structured environmental processes are relatively more important than native plant species richness alone in shaping the native-exotic plant species richness relationships.