Parameterization of two‐dimensional approaches in HYDRUS‐2D: Part 1. Simulating water flow dynamics at the field scale

Abstract

AbstractParameterization of theeffective properties as input in hydrogeological models for simulating water flow is a time consuming and costly process, and it is characterized by high uncertainty because of the spatial variability of soil properties. Integrating visible−near‐infrared spectroscopy (vis−NIRS) models and a developed pedotransfer function, a hydrogeological tool was suggested by Varvaris et al. in 2019 for estimating the hydraulic properties used in an equilibrium approach. In the current study, the proposed concept was further expanded for assisting on a rapid and inexpensive parameterization of effective parameters in nonequilibrium models. In our approach, the soil‐water retention curves for the wet and hyper‐dry ends were predicted using existing vis−NIRS calibration models for obtaining the parameters for the Campbell retention model and the Campbell and Shiozawa model, respectively. The HYDRUS‐2D software package was used to simulate tile drainage discharge in a loamy structured field using as initial input hydraulic parameters the output from the suggested concept. The estimated parameters were validated based on measured hydraulic properties in the laboratory. Two‐dimensional models incorporating nonequilibrium flow were used to evaluate the ability of each approach to predict the tile drainage discharge: a single‐porosity with modified van Genuchten function, a dual‐porosity model, and a dual‐permeability model. All approaches gave a fairly good fit to the hydrograph features for both calibration and validation datasets with dual‐porosity to present the highest predictive ability. Finally, the suggested initial parameterization concept proved its potential to be used as an alternative for covering the absence of measured data of soil hydraulic properties.