Modified conceptual model for compensated root water uptake - A simulation study

- Conceptual inconsistencies of a frequently used compensation model are analyzed.
- Severity of these inconsistencies depends on soil texture and root distribution.
- These problems are solved by a simple modification.
- Modification accounts for maximum transpiration based on lowest stress in profile.
- Modification does neither increase complexity nor the number of parameters.

SummaryModeling root water uptake within the macroscopic approach is usually done by introducing a sink term in the Richards equation. This sink term represents potential water uptake reduced by a so-called stress reduction factor accounting for stress due to high suctions, oxygen deficit or salinity. Since stress in some parts of the soil can be compensated by enhanced water uptake in less stressed parts, several compensation models have been suggested. One of them is the empirical model of Jarvis, which is often applied due to its mathematical elegance and simplicity. However, it has been discussed that under certain conditions and assumptions this model might predict too high transpiration rates, which are not in agreement with the assumed stress reduction function. The aim of this paper is (i) to analyze these inconsistencies and (ii) to introduce a simple constraint for transpiration in a way as if the complete water would be taken form the location with highest uptake rate in the uncompensated case. Transpiration from 50 cm deep soils with hydraulic functions representing different textures, ranging from a clay loam to a coarse sand, was simulated with the original and the modified model using HYDRUS-1D. Root distribution was assumed to be uniform or linearly decreasing with depth. In case of the fine textured soils and uniform root density, the original model predicted transpiration equal to potential transpiration even when the complete root domain was already heavily stressed if the maximum enhancement factor for uptake was 2. These results are not in agreement with the original meaning of the stress reduction function. The modification eliminates the inconsistencies by limiting transpiration to a maximum value based on the highest uncompensated uptake rate in the root zone. It does neither increase the mathematical complexity nor require any additional parameters.