Field experiments have enhanced our understanding of drought impacts on terrestrial ecosystems-But where do we go from here?

We review results from field experiments that simulate drought, an ecologically impactful global change threat that is predicted to increase in magnitude, extent, duration and frequency. Our goal is to address, from primarily an ecosystem perspective, the questions 'What have we learned from drought experiments?' and 'Where do we go from here?'.Drought experiments are among the most numerous climate change manipulations and have been deployed across a wide range of biomes, although most are conducted in short-statured, water-limited ecosystems. Collectively, these experiments have enabled ecologists to quantify the negative responses to drought that occur for most aspects of ecosystem structure and function. Multiple meta-analyses of responses have also enabled comparisons of relative effect sizes of drought across hundreds of sites, particularly for carbon cycle metrics. Overall, drought experiments have provided strong evidence that ecosystem sensitivity to drought increases with aridity, but that plant traits associated with aridity are not necessarily predictive of drought resistance. There is also intriguing evidence that as drought magnitude or duration increases to extreme levels, plant strategies may shift from drought tolerance to drought escape/avoidance.We highlight three areas where more drought experiments are needed to advance our understanding. First, because drought is intensifying in multiple ways, experiments are needed that address alterations in drought magnitude versus duration, timing and/or frequency (individually and interactively). Second, drivers of drought may be shifting-from precipitation deficits to rising atmospheric demand for water-and disentangling how ecosystems respond to changes in hydrological 'supply versus demand' is critical for understanding drought impacts in the future. Finally, more attention should be focussed on post-drought recovery periods since legacies of drought can affect ecosystem functioning much longer than the drought itself.We conclude with a call for a fundamental shift in the focus of drought experiments from those designed primarily as 'response experiments', quantifying the magnitude of change in ecosystem structure and function, to more 'mechanistic experiments'-those that explicitly manipulate ecological processes or attributes thought to underpin drought responses.Read the free Plain Language Summary for this article on the Journal blog. Read the free Plain Language Summary for this article on the Journal blog.image