Master Thesis

Effects of soil heterogeneity on root water uptake under drying conditions and varying transpiration rates

Support: Andrea Carminati, Mohsen Zare

Water flow in soils is heterogeneous at many scales. In a given representative elementary volume these heterogeneities can be described by an effective hydraulic conductivity. Standard root water uptake models commonly rely on this representative hydraulic conductivity and assume that the hydraulic conductivity of the soil cylinder around the roots is equal to that measured in soil samples representative of the bulk soil.

Water flow in soils is heterogeneous at many scales. In a given representative elementary volume these heterogeneities can be described by an effective hydraulic conductivity. Standard root water uptake models commonly rely on this representative hydraulic conductivity and assume that the hydraulic conductivity of the soil cylinder around the roots is equal to that measured in soil samples representative of the bulk soil. Here we argue that this effective conductivity is different from that controlling root water uptake for both hydrological and biological reasons: 1) When in drying soil the hydraulic conductivity becomes limiting, roots extract water from the most conductive zones, where the fluxes are highest and the most rapid depletion occurs. Hence, for transient flow conditions, typical for root water uptake, the emerging relation between flow and gradient in water potential cannot be properly represented using a single effective conductivity. 2) Roots control their hydraulic contact with the soil via root hairs, mucilage and hydropatterning (the preferential growth of roots towards regions where water is more easily available), thereby altering their capacity to extract water from the soil.

We propose a model with two distinct concentric compartments for describing the radial flow of water from the soil to the root surface. The two compartments include a highly conductive domain proximal to the root, corresponding to the most conductive zones where water is extracted, and a distal domain with a hydraulic conductivity equal to that of the bulk soil. The radial flux is divided by the fraction of the root surface that is in contact with highly conductive zones. Compared to a homogenous model with a single compartment, water flow in the two-compartments model could sustain root water uptake under higher transpiration rates even when the root surface in contact with the soil is largely reduced. The cost, however, is that once depleted, these highly conductive zones rewet very slowly. This results in an apparent hysteresis in the relationship between root water potential and root water uptake during diurnal cycles in transpiration rate. The model is capable to include alteration of the root-soil contact due to the presence of root hairs, mucilage secretion and hydrotropism.

The proposed model challenges existing models of water flow in soils and sheds a new light on the puzzling hysteresis that has been long observed in the relation between transpiration and leaf water potential.