Analysis of plant spatial structure can reveal plant associations with environmental heterogeneity, species-species interactions, and dispersal limitation. Recent work in mapped forests suggest multiple processes likely contribute to community structure, with evidence supporting the influence of abiotic variables, dispersal, recruitment, and competition on plant spatial patterns. While this work has yielded important insight, it is difficult to clearly tease apart effects of environmental heterogeneity from other factors in these observational studies where scales of heterogeneity are unknown or inferred. Here we use a soil heterogeneity experiment to examine how known patterns of soil heterogeneity influence plant spatial structure after five years of grassland community assembly. Soil heterogeneity was manipulated by splitting the vertical soil profile into three soil-types that were randomly assigned to 40 X 40cm patches comprising 2.4 X 2.4m plots. Homogeneous plots were created by mixing all soils before filling each patch. Following the soil manipulation, thirty-four tallgrass prairie species were sown into all plots and allowed to grow for five years after which the presence and density of all plants were mapped using a 5 X 5cm grid. Point-pattern spatial analysis was used to analyze univariate and bivariate patterns for the fifteen most abundant species.
Results/Conclusions
Thirteen of fifteen species were spatially aggregated in both plot types, while one species was aggregated only in homogeneous plots. Scales of aggregation ranged from within 10cm of a focal plant to over 50cm. Most species had similar scales of aggregation between plot types, but several aggregated at second scales in only one plot type. Four out of five times this additional aggregation occurred in heterogeneous plots and corresponded with the approximate size of soil patch structure (35cm), suggesting soil conditions alter plant spatial structure. Analysis of bivariate pairs indicated that nineteen species pairs in heterogeneous and fifteen pairs in homogeneous plots were either positively or negatively associated. However, only three associations were shared between both plot types, suggesting differences in soils may alter the identity of species-species associations. Interestingly, most non-random bivariate patterns were positive, suggesting that facilitative interactions may be important in this community regardless of soil conditions. Taken together, differences in univariate and bivariate patterns between plot types suggest soil heterogeneity influences community structure and that even in the absence of dispersal limitation and soil heterogeneity, species interactions are likely a driving force in these commonly observed patterns of plant aggregation.