A multi-method approach to quantify hydraulic redistribution

Hydraulic redistribution (HR) occurs in temperate ecosystems and may increase forest resilience to drought. However, its magnitude and influence on the stand-level water budget remain poorly understood. One major challenge in field studies is to distinguish HR from other processes that lead to real or apparent increases in soil water content at night. In this study, we aimed to identify and quantify upward water fluxes due to HR by combining multiple methodological approaches in a mixed broad-leaved forest in Germany. During three consecutive growing seasons, we monitored soil water dynamics using continuous soil water content and soil water potential sensors installed at multiple locations in the tree stand and in root-exclusion control plots, where roots were cut to prevent HR. Additionally, several infiltration events were performed by inducing 200–800 L of deuterium-enriched water at 1.8–3.5 m depth, and its movement was traced with continuous in situ and destructive isotope analyses plus geoelectrical monitoring. We hypothesized that (1) nocturnal increases in soil water content and soil water potential would be larger in the stand plots than in the control plots due to HR, and (2) we would detect the tracer in upper soil layers if water moved from the irrigation depth to the topsoil through HR. Our multi-method analysis confirmed the occurrence of HR in this temperate forest, although its magnitude was low. Initial results showed that nocturnal increases in soil water potential were more frequent in the stand plots than in the controls, but the associated changes in water content remained below 1 vol. %. The tracer rarely appeared in observed trees and was not detectable in the topsoil. Geophysical monitoring results showed the injected tracer was rapidly distributed (approximately 10 m within only a few hours) along preferential flow paths below 2 m depth. The reason no tracer uptake was observed in the trees could be that they either primarily rely on shallower water sources, or that the clayey soil prevented the infiltrated water from reaching more distant trees during our observation windows. Future spatially and temporally high-resolution geoelectrical monitoring experiments are planned to noninvasively map and upscale these HR fluxes, link them to nutrient and carbon cycling, and improve predictions of forest resilience to drought.