Tracking Water Dynamics of a Temperate Forest Under Drought and Non-Drought Conditions Using Active and Passive Microwave and Optical Remote Sensing

Abstract

Water status, water dynamics, and ecohydrological resilience of a protected German beech forest during the 2018–2020 multi-year drought are assessed using over six years of multi-frequency remote sensing data, integrating active, passive, and optical sensors with varying canopy penetration depths, highlighting the importance of monitoring forests under extreme conditions. In this study, we investigate Sentinel-1 C-band backscatter (S-1 γ0) and the relative water content estimated from vegetation optical depth (RWCVOD) of AMSR2 (X- and C-bands) and SMAP (L-band) within the soil-plant-atmosphere system (SPAS). In addition, time series are intercompared and examined through correlation and sensitivity analyses applying tailored environmental and newly developed hydrological selection strategies. Our results show that S-1 γVH is most influenced by leaf area index and thus leaf biomass when sensed during dense vegetation (leaf-on), no-frost, and very wet conditions (r = −0.94). In contrast, during sparse vegetation (leaf-off), no-frost, stable dry-down, and extremely dry conditions, S-1 γVH is very sensitive to topsoil moisture (r = 0.91). Due to increased microwave attenuation, resulting in reduced S-1 γVH backscatter, an anti-cyclical behavior (negative correlations) is observed between almost all SPAS-based variables/proxies and S-1 γVH during leaf-on conditions. Conversely, this reverses to a cyclical behavior (positive correlations) during leaf-off conditions. Our results reveal that X- and C-band RWCVOD effectively detect drought onset by capturing fast water content changes in leaves and twigs of the top canopy due to shallow sensing depth, while L-band RWCVOD captures legacy effects after repetitive droughts through slower water content changes in branches and trunks of lower tree compartments.