Phenotypic response of plants to simulated climate change in a long-term rain-manipulation experiment: a multi-species study.

Many species will need to adapt to the observed climate change in order to persist. However, research about adaptation or phenotypic plasticity in response to climate change is rare. In particular, field studies are lacking that impose artificial selection for a sufficiently long time to elicit changes in phenotypic and genotypic structure of populations. Here, we present findings for an 8-year field experiment with 16 annual plant species that tested potentially adaptive phenotypic responses to precipitation change. In both a Mediterranean and a semi-arid site, annual precipitation was manipulated (±30%) and phenotypic response was recorded. We measured flowering time as a key trait related to climatic conditions and biomass and survival as fitness correlates. Differences in traits among treatments were compared to trait shifts between sites, according to space-for-time approaches. In the drier site, phenology was accelerated, but within that site, experimental drought delayed phenology, probably as a plastic response to delayed ontogenetic development. Biomass was smaller in the dry treatments of that site, but it was also reduced in irrigated plots in both sites, indicating more intense competition. The shifts from limitation by drought to limitation by competition corresponded to patterns along the gradient. This also implies a larger negative impact of climate change in the drier site. Our results suggest that experimental selection in the field caused directional responses in most species, but these were not necessarily adaptive. Furthermore, competitive release imposed by climate change may revert direct negative effects of rainfall change in determining plant performance.