Combining visible-near-infrared and pedotransfer functions for parameterization of tile drain flow simulations

Estimation of soil hydraulic parameters is essential when generating a hydrogeological model for simulating water flow dynamics in an agricultural field. However, estimation of the input parameters through direct measurements is time consuming and costly, and the spatial variability presents an uncertainty. Therefore, we proposed a rapid and inexpensive concept (integration of visible-near-infrared spectroscopy [vis-NIR] and a pedotransfer function [PTF]) to estimate hydraulic properties considering catchment scale. An existing vis-NIR-predicted Campbell retention function was used for estimating the Campbell b parameter and the water content at -1000 cm H 2 O soil-water matric potential (log|-1000| = pF 3). A PTF was developed for predicting the saturated hydraulic conductivities using the vis-NIR-predicted Campbell b and the effective porosity, defined as the difference in volumetric water contents at pF 0.3 and 3. The concept was evaluated by developing a hydrogeological model in HYDRUS-2D software for simulating the tile drainage discharge from a clayey agricultural subcatchment in Denmark, using as input hydraulic parameters the output from the suggested approach. The suggested approach simulated the main attributes of the flow hydrograph with a reasonable degree of accuracy (R2 and RMSE values of 0.86 and 1.25 L s -1 , respectively). A sensitivity analysis was performed to determine the response of the model to changes in values of predicted parameters when predicting the drainage discharge, and it showed that small variations (<10%) would not affect the predictive ability of the model.