Field experiments and observational studies are the two most commonly used approaches to assess the effects of climate change on ecosystems. Although the two methods differ substantially, and their limitations have been broadly discussed, there are surprisingly few comparisons of experimental and observational results. Furthermore, for drivers where experimental results abound, such as drought, synthesis studies tend to focus on experiments. Yet, in a recent analysis we found that experiments severely underestimate the effect of drought on aboveground biomass. We plan to build on these results, and test the generality of this difference between experiments and observational studies, as well as explore underlying mechanisms.
The objective of the proposed work is to compare experimental and observational approaches in global change research by (a) synthesising meta-analyses on the ecological effects of global change, (b) investigating the potential role of soil moisture movement in the underestimation of drought effect in drought experiments, and (c) continuing a multifactor drought experiment and an observational, long-term monitoring study at the same site, and comparing their results on grassland vegetation dynamics.
Because field experiments are widely applied in global change research, and their results are often used even for broad upscaling and modelling (e.g. to estimate changes in global carbon stocks), we believe it is highly important to document the limitations of experiments, while also acknowledging and making use of their strengths.