Disentangling Effects of Vegetation Structure and Physiology on Land–Atmosphere Coupling

Terrestrial vegetation is a key component of the Earth system, regulating the exchange of carbon, water, and energy between land and atmosphere. Vegetation affects soil moisture dynamics by absorbing and transpiring soil water, thus modulating land–atmosphere interactions. Moreover, changes in vegetation structure (e.g., leaf area index) and physiology (e.g., stomatal regulation), due to climate change and forest management, also influence land–atmosphere interactions. However, the relative roles of vegetation structure and physiology in modulating land–atmosphere interactions are not well understood globally. Here, we investigate the contributions of vegetation structure and physiology to the coupling between soil moisture (SM) and vapor pressure deficit (VPD) while also considering the contributions of influential hydro-meteorological variables. We focus on periods when SM is below normal in the growing season to explicitly study the regulation of vegetation on SM–VPD coupling during soil dryness. We use an explainable machine learning approach to quantify and study the sensitivity of SM–VPD coupling to vegetation variables. We find that vegetation structure and physiology exert strong control on SM–VPD coupling in cold and temperate regions in the Northern Hemisphere. Vegetation structure and physiology show similar and predominant negative sensitivity on SM–VPD coupling, with increases of vegetation dynamics leading to stronger negative SM–VPD coupling. Our analysis based on Earth system model simulations reveals that models largely reproduce the effect of vegetation physiology on SM–VPD coupling, but they misrepresent the role of vegetation structure. This way, our results guide model development and highlight that the deeper understanding of the roles of vegetation structure and physiology serves as a prerequisite to more accurate projections of future climate and ecosystems.