Spatial heterogeneity is a primary mechanism explaining species coexistence and extant patterns of diversity. Despite strong theoretical support and ample observational evidence, few experimental studies in plant communities have been able to demonstrate a causal link between environmental heterogeneity and plant diversity. This lack of experimental evidence suggests that either fine-scale heterogeneity has weak effects on plant diversity or that previous experiments have been unable to effectively manipulate heterogeneity. We tested the effect of soil heterogeneity on plant diversity using a novel soil manipulation in which we redistributed portions of soil profile—which were expected to vary in soil texture, organic matter, nutrients, and pH, and root activity—into either homogeneous or heterogeneous spatial arrangements. In each 2.4x2.4 m plot, we excavated the upper-three 15 cm strata from the soil profile and either redistributed the strata into individual 0.4x0.4 m patches within heterogeneous plots or combined all three prior to redistribution into the aforementioned patches within homogeneous plots. In the spring following the soil manipulation, we removed early emergent weeds with herbicide and then sowed seed from thirty-four native species into each plot and quantified plant density by species in each patch.
Results/Conclusions
After two growing seasons, species richness was significantly higher in heterogeneous relative to the homogeneous plots. In the heterogeneous plots, thirteen species had higher establishment in a specific patch type representing one of the three soil strata. Conversely, no species had greater establishment in the mixed strata that comprised the homogeneous plots relative to the heterogeneous strata. This pattern of species sorting suggests that fine-scale heterogeneity creates opportunities for plant establishment due to niche differences, which translates into increased plant diversity at the plot scale. Furthermore, in the homogenous plots where fine-scale heterogeneity was minimized, species richness was significantly related to cumulative plant density while this relationship was not found in the heterogeneous plots. These patterns suggest that—as species sorting increases—plant richness becomes increasingly dependent on the match between the suite of species and fine-scale heterogeneity rather than the cumulative number of individuals.