Abstract
Rainout shelters are research tools commonly used to simulate drought conditions in climate change experiments, allowing researchers to investigate the effects of reduced precipitation on ecosystem processes. However, these structures can introduce unintended microclimatic effects, such as changes in soil temperature and solar radiation, which can affect the interpretation of drought simulations. This study, conducted at a long-term ecological research station in a Mediterranean shrubland ecosystem, assessed the intended and unintended microclimatic effects of different rainout shelter designs-striped and closed-on soil moisture, temperature, and solar radiation. Using sensors, we monitored microclimatic conditions beneath two rainout shelter design types simulating multiyear extreme drought with a 66% rainfall reduction. While both striped and closed rainout shelter types effectively reduced soil moisture, the magnitude of reduction was nonlinear and lower than expected (18%-40%). Additionally, rainout shelters increased soil temperatures in winter and decreased them in summer, likely due to wind-blocking and shading. Solar radiation was reduced on average by ~20%, with maximum reductions of up to 40% under shelters. This study demonstrates how rainout shelters regulate soil moisture, but also unintentionally alter temperature and solar radiation, revealing seasonally contrasting thermal effects and nonlinear soil moisture reductions which may complicate the interpretation of drought manipulation experiments. This study also underscores the importance of tailoring experimental set-ups to specific site conditions, such as soil texture and natural rainfall variability, to enhance experimental aims and ecological relevance. By addressing key methodological challenges, this study complements rainfall manipulation experiments and improves predictions of ecosystem responses to climate change.