Compositional variation in grassland plant communities

Human activities are altering ecological communities around the globe. Understanding implications of these changes requires that we consider composition those communities. However, can be summarized by many metrics which in turn influenced different processes. For example, incidence-based strongly reflect species gains or losses, while abundance-based minimally affected abundance small uncommon species. Furthermore, might correlated with predictors. We used a globally distributed experiment to examine variation within 60 grasslands on six continents. Each site had an identical experimental and sampling design: 24 plots × 4 years. expressed compositional each site—not across sites—using abundance- magnitude dissimilarity (Bray–Curtis Sorensen, respectively), measures relative importance replacement (balanced turnover, richness at two scales (per plot-year [alpha] per [gamma]). Average among all plot-years was high similar spatial pretreatment year, but lower years untreated plots. both types metrics, most due rather than nestedness. Differences sites overall within-site were related several Environmental heterogeneity (expressed as CV total aboveground plant biomass unfertilized site) important predictor for metrics. Biomass production turnover alpha diversity not other Continentality (measured annual temperature range) strong Sorensen dissimilarity. Metrics moderately correlated: knowing provides little insight into whether is driven Overall, our understanding enhanced considering multiple simultaneously. Monitoring programs explicitly incorporate implications, when designing strategies analyzing data, will have stronger ability understand systems quantify impacts human activities.