COS 48-3 - CANCELLED - Spatial patterns of seed arrival and soil heterogeneity alter plant community assembly in a tallgrass prairie

Background/Question/Methods

Plant community assembly can have substantial effects on plant community structure and function, yet teasing apart the role of local environmental conditions and propagule arrival on the assembly process remains challenging. We used a field experiment designed to test how soil heterogeneity, patch scale, and the spatial structure of seed arrival influence plant community assembly in a tallgrass prairie. We established 96, 4x4.8m plots each divided into 120 large-scale (0.4x0.4m) or 480 small-scale (0.2x0.2m) patches. Different soil types were utilized to create either homogeneous or heterogeneous soil conditions. Additionally, plots either received uniform seeding by sowing 40 species in each patch, or aggregated sowing by sowing 1 species in each patch. Seeding rates were held constant for each species at the plot level. We utilized Indicator Species Analyses to test for species sorting among patch types in the various treatment combinations.

Results/Conclusions

Over five growing seasons, differences in plant diversity at the plot level were primarily driven by seed arrival and the size of patches that comprised the experimental plots. In each year examined, we found evidence for species sorting for 12-16 species; however, the identity of the species shifted over time. In the earliest year examined, species sorting was detected among patches where all 40 species were sown (uniform seed arrival across the plot). As community assembly progressed, species sorting mostly occurred in patches that received a single species (aggregated seed arrival at the plot scale). Approximately one-third of the species identified in the indicator species analysis were unsown species that arrived from local seed dispersal, or the seedbank and this pattern was consistent across the five years examined. Establishment by these unsown species was likely associated with differential success of the sown species in the aggregated sowing treatments, particularly in the latter three years. Taken together, our results suggest that the spatial pattern of seed arrival has important effects on community assembly in these grassland ecosystems.