The datasets in this resource are associated with: King, Evan Robyn. Tracing water through a forest root zone. Diss. 2023
Abstract: Recent critical zone studies have highlighted the important role that unsaturated weathered bedrock plays in the storage of plant-available water, particularly during dry periods when incoming precipitation is limited. Unlike for soils, our knowledge of unsaturated water flowpaths within weathered bedrock, which may extend many meters into the subsurface before reaching the water table, remains relatively unknown. In this study, we employed water stable isotopes to trace the fate of waters entering a steep, weathered bedrock-dominated hillslope in Northern California. We used a subsurface vadose zone monitoring system (VMS) that contains sets of flexible sensors and samplers within inclined sleeves to sample waters at discrete intervals down to 16.6 m depth to fresh bedrock. Additionally, we sampled several other water fluxes and reservoirs at the hillslope, including storm samples and tightly-held matrix waters. Previous studies at the site revealed a dynamic, seasonally wetting and drying subsurface in response to a Mediterranean-type climate of long, dry summers and cool, wet winters. Dynamic storage estimates and drilling campaigns show that roots may extend to 16 m depth and likely play a role in the transmission of waters to groundwater and stream. We report the results of a tracer experiment, whereby a deuterated-water tracer was injected into the hillslope in May 2019 to simulate the last large storm of the wet season. We sampled waters transiting the unsaturated zone and monitored precipitation inputs for the three years following the tracer application to confidently detect the signal of the tracer to 4.7 m depth, with tracer signals sustained at a single depth interval for up to 21 months. We propose that mixing between dynamic and nondynamic waters within the weathered bedrock zone allows the persistence of the tracer signal through several dry seasons. We compare VMS-extracted waters with cryogenically extracted waters to show that isotopically distinct pools may exist within the hillslope. Finally, we explored how rooting depth may influence tracer transmission by simulating flow in the upper 10 m of our hillslope in HYDRUS-1D. We find that rooting depth may determine the extent to which the tracer is mixed with a nondynamic reservoir and the proportion of tracer that is extracted via transpiration.