Understanding how plant communities will respond to future climate change is critical. For a species to persist, it must not only survive the direct effects of climate changes but also the indirect effects of altered densities and identities of competitors. We don’t know what the indirect effects of biotic interactions are for many communities making it difficult to predict how plants will be affected by future climate. Using an annual grassland community in coastal California which is predicted to have increased precipitation volatility, we explored how changes in precipitation alter competitive dynamics via direct effects on focal species as well as by the changing strength of interactions between species. We grew six annual species (including two grasses and four forbs) in different pairwise competition backgrounds that were subject to both an ambient and a reduced precipitation treatment that was created via temporary rainout shelters. We tracked over 2,000 individuals recording their fecundity and competitors. We then assessed the coexistence consequences of the rainfall changes for each species pair using an annual plant population dynamics model that was fit using field-parameterized estimates of germination, fecundity and per capita competition from the experiment.
Results/Conclusions
Our rainout shelters produced a 20% reduction to seasonal precipitation, but this had weak direct effects on the seed production of plants growing without competition. Only one species, Vulpia microstachys, suffered significantly lower seed production in the decreased water treatment. In contrast, the precipitation manipulation had substantial effects on both inter and intraspecific competitive interactions. Specifically, water availability changed intraspecific and interspecific interaction coefficients enough to substantially alter stabilizing niche and fitness differences between many species pairs. The changes induced with the rainout shelters were large enough to shift the predicted outcome of competition between species pairs from coexistence to competitive exclusion (5 of 15 pairs) or vice versa (also 5 pairs). These findings indicate that modest changes in rainfall that have little direct effects on plant performance can nonetheless have substantial impact on community composition via changes in the strength of plant-plant interactions. Our work highlights that it is essential to account for changes in biotic interactions when predicting the consequences of climate change on plant communities.