Six different reduction functions for combined water and salinity stress are used in the macroscopic root water extraction term. The reduction functions are classified as linear additive, non-linear multiplicative, and that which is neither additive nor multiplicative. All these reduction functions are incorporated in the numerical simulation model HYSWASOR. The relation between the experimental relative transpiration and the joint soil water osmotic and pressure heads appears to be linear (with an exception for the salinity near the threshold value). As the mean soil solution salinity increases, the trend becomes more linear. The simulations indicated that for most treatments the newly proposed reduction term provides the closest agreement with the experimental transpiration. Soil water content, and particularly soil solution salinity simulated with this equation agree reasonably with the experimental data: in spite of the observed differences, the trend of the simulated data is good. A reason for the disagreement between the simulated and experimental water contents can be attributed to the influence of roots and the soil solution concentration on the soil hydraulic conductivity. The input soil hydraulic parameters were obtained from soil samples without roots and salinity and assumed constant during the simulations.
Homaee, M., Feddes, R. A., & Dirksen, C. (2002). Simulation of root water uptake. III Non-uniform transient combined salinity and water stress. Agricultural Water Management, 57, 127-144. https://doi.org/10.1016/S0378-3774(02)00073-2