Modeling root water uptake: root water compensation and hydraulic redistribution

Root water uptake by plants mediates the exchange of water, carbon and energy between land surface and atmosphere and is important in hydrological, climatological, agricultural and ecological studies. Field measurements show that root water uptake could be significantly affected by root water compensation and hydraulic redistribution. We thus use a root water uptake model able to describe these two mechanisms and show their importance when vegetation is growing in shallow water-table environments or duplex soils. The model is based on the Richards equation for the water-flow in soils, with a term for root water uptake being a function of the water potential difference between root xylem and the soil. We describe the flow in the xylem using the Darcy's equation. The model is used in three studies aimed at highlighting the role that root water compensation and hydraulic redistribution might have on the overall root water uptake. In the first study, the model in one dimension is applied to a site near Sydney, Australia, to investigate how native trees growing on duplex soils are able to sustain transpiration rate despite long periods with little or no rain. The model was able to reproduce sap-flux data and the pattern of soil, root and leaf water potential for several months. Scenarios with different root depths showed that trees were able to adjust their water-uptake rates from different soil layers based on soil moisture availability; thus, root water compensation appears to be a key mechanism to maintain sustained transpiration rates. In a second study we investigated the contribution of root water compensation and hydraulic redistribution to root water uptake in shallow water-table environments. We compared the results of our model with a more commonly used root water uptake model. In the third study, we extended the 1D-model to two dimensions, thereby being able to simulate horizontal hydraulic redistribution and the interaction between species with different root systems. In the 2D-model, the roots were assumed to be a continuum in soil and the root systems were described in terms of xylem conductivity fields. Scenarios are presented to show that the 2D-model is able to reproduce observed flow patterns through roots in parts of the soil with different degrees of moisture. The studies presented in this thesis show the further development and use of a modeling approach that is gaining increasing interest in the recent literature. These studies present a realistic description of the role that root water compensation and hydraulic redistribution play in plant water use. The 2D-model introduces a representation of the root system that allows for modelling vertical (hydraulic lift) and horizontal hydraulic redistribution of water in soil, and an efficient description of the interaction between species with different root systems.