Global change alters the stability of biological communities by affecting species richness and how species covary through time (i.e., synchrony). There are few large-scale empirical tests of stability-diversity-synchrony relationships and those mostly focus on the terrestrial realm. Moreover, the effect of synchrony is largely unknown when species only covary at either high or low extremes of abundance (i.e., tail-dependent synchrony), a common phenomenon in ecological communities. Here, we synthesized long-term community time-series data (20+ years of species’ abundances/biomass for 2,668 communities across 7 taxonomic groups) from both terrestrial and freshwater realms and explored how the relationships among richness, synchrony, and stability vary across realms. We also investigated the effect of tail-dependent synchrony on stability across 714 freshwater and 1,954 terrestrial communities. For terrestrial communities, we found a positive diversity-stability relationship and that the tail-dependent synchrony was a more important determinant of stability than the traditional measure of overall synchrony (i.e., based on the covariation of all species). For freshwater communities, only overall synchrony explained some variation in stability. Assessing tail-dependent synchrony can improve our ability to understand why stability varies across different ecosystems and thereby our inferences about the causes of human-mediated biodiversity loss.

Nutrient perturbations are a threat to aquatic ecosystems worldwide. Interactions among important species can influence the recovery of aquatic ecosystems from nutrient perturbations, and the persistence in a clear water state for a given regime of nutrient loading. In a factorial manipulation of two foundation species (macrophytes and mussels), we measured 16 large (15000L) ponds at high resolution (15 min interval) over two years with two periods of nutrient perturbation. We found that each foundation species had strong individual effects on multiple ecosystem properties (e.g. phytoplankton densities and metabolism), largely in line with our expectations (i.e. reduction of algal biomass relative to control). However, when both species were present, we saw dramatic increases in algal biomass and other ecosystem parameters, indicating strong non-additive antagonistic effects. Overall, our results demonstrate how interactions between foundation species can cause surprisingly strong deviations from the expected responses of aquatic ecosystems to nutrient additions.