Temporal dynamics of stress signal propagation across ecosystem scales during a hot and dry period: A multi-sensor analysis

Stress signals propagate through ecosystems via interactions between atmospheric aridity and soil water depletion, resulting in physiological stress in plants, altered structural dynamics, and changes in carbon, energy, and water fluxes. Monitoring these impacts is possible using a growing suite of established and novel sensor systems. Available approaches span tree-level measurements such as stem water potential, stand-scale fluxes derived from eddy-covariance towers, and spatially continuous indicators from Earth observation satellites using optical and radar backscatter. Yet, the degree to which such measurements track complementary stress processes, and the extent to which their responses are temporally coupled, remain poorly quantified.

We address this gap through a multi-scale analysis of a distinct hot and dry period in August 2025 (August 7–19) at the temperate forest of the ICOS-associated Forest Research Site DE-Har (Hartheim, Germany), which is characterized by limited soil water storage and rapid soil drying. Using distributed sensors, we tracked the propagation of stress signals across five interacting levels. These include (1) atmospheric demand (vapour pressure deficit, air temperature), (2) soil water status (volumetric water content), (3) plant hydraulics (stem water potential, tree water deficit, sap flow), (4) canopy structure and leaf properties (leaf angle distribution via AngleCam, GNSS-T based vegetation optical depth, plant area index from permanent terrestrial laser scanning, leaf area index from hemispherical photographs, vegetation greenness, Sentinel-1 radar backscatter, Sentinel-2 optical indices), and (5) ecosystem fluxes (net ecosystem exchange, gross primary productivity, evapotranspiration).

Using cross-correlation and lag analysis at daily resolution from May to October 2025, we quantify the temporal sequence in which these measurements respond to the hot and dry period in August 2025. We determine whether certain variables act as leading indicators and to what extent time-lags emerge as stress signals propagate from the atmosphere to ecosystem fluxes. This integrated perspective can reveal which measurements track similar aspects of stress and which provide complementary information that would be missed by any single approach alone. Moreover, this analysis emphasises the potential of novel, scalable sensor techniques such as tracking leaf angle dynamics from video cameras (AngleCam) and GNSS-T-based vegetation optical depth.

Our outcomes provide a temporally resolved view of stress signal propagation in a drought-impacted temperate forest ecosystem, which can inform ecosystem modelling and the design of multi-sensor monitoring networks.