LAUSR

Climate change is predicted to have profound consequences for multispecies coexistence, and thus, patterns of biological diversity. These consequences will be mediated by direct and indirect impacts of environmental change on species’ vital rates and interactions. While the impacts of environmental change on individual species has received much attention to date, the consequences for coexistence mediated by changes in the strength and direction of multispecies interactions are not as well understood. To investigate how coexistence dynamics may be sensitive to environmental change, we conducted a field experiment in a diverse semi-arid annual plant system. We imposed a water manipulation treatment in two sites that vary in aridity and associated rainfall. Focusing on four common annual plant species in these sites, we quantified the fecundity (seed production) of individuals in response to a gradient of intra- and interspecific competitor densities and aridity. We then used these fecundities to parameterize an annual plant population model and examine the influence of aridity and species identity on resultant coexistence dynamics (as a function of stabilizing niche differences and fitness inequalities). While the responses of some vital rates and competitive impacts to watering varied somewhat predictably across sites, coexistence metrics encapsulating changes in these vital rates and interaction strengths did not. Fitness inequalities among our focal species were driven largely by differences in sensitivity to competition, which were almost always much greater than the magnitude of stabilizing niche differences. These findings were surprising given observational evidence suggesting that these species do coexist at local scales in these natural communities. Synthesis. Our study is one of the first to explicitly consider the influence of environmental variation on the individual components of coexistence outcomes. We show that environmental change has the ability to influence coexistence not only through direct pathways (i.e., vital rates), but also indirect pathways (i.e., species interactions). Despite the consistency of many of the responses of these individual components to environmental variation, their combined influence on predictions of both current and future coexistence remains unclear.