Can we observe flow-weighted water potential at the root-soil interface in heterogeneously moist soils?

Plant roots draw soil water locally, increasing the moisture heterogeneity in the soil at the onset of drought. The resulting heterogeneity presents a major challenge for linking observed flux rates to measured soil moisture values using process-based models. As such, soil moisture heterogeneity is a key remaining hurdle to robust, mechanistic predictions of forest canopy fluxes under water limitation and a stubborn source of uncertainty in predictions of the future terrestrial carbon cycle.

Several recent theoretical advances in describing soil-root water flow at plant or larger scale despite heterogeneous moisture distributions (e.g., Hildebrandt et al., 2016; Vanderborght et al., 2021) share one potentially central feature: the conductance- or flux- weighting of water potential at the soil-root interface. Flux-weighted water potential may be a key concept capable of characterising the hydrodynamic state of the soil-plant system in a single value regardless of its instantaneous heterogeneity.

Given the apparent theoretical promise of this concept, we should ask whether we can infer its values from field observations and, if so, what and how to measure. The challenges of directly measuring plant water potential over time are already daunting aboveground. Maintaining a dense network of probes for soil water content and water potential at substantial cost and effort may not yield relevant values, since the potential drop toward the root is nonlinear and largest over the final millimetres of soil. One potentially promising avenue for field observations is afforded by recent advances in optical methods both above and below ground. Separately, key parameters arising from the process-based models will need to be constrained in lab-based experiments. Collaborators within the ongoing HydroScale project aim to develop a complex approach combining traditional and innovative techniques sufficient to infer flux-weighted potentials from field data.