Comparison of Accuracy of van Genuchten and Brooks & Corey models for Simulating Water Flow in Forest Floor using HydroGeoSphere Code

The forest residuals play an important role in runoff rate, soil erosion, and soil infiltration capacity of protecting mineral (surface) soils from the direct impact of raindrops. By intercepting rainfall, the forest residuals serve as a temporary reservoir and allows more time for infiltration into the mineral soil beneath (Kosugi et al., 2001). Hydraulic properties of forest residuals were unknown to some extent and could not be measured with similar methods used for mineral soil. In recent years, several studies on the forest floor have been published (Kosugi et al.,2001; Schaap et al.,1997). The objective of this study was the comparison of accuracy of van Genuchten and Brooks & Corey models for simulating water flow in forest floor using the HydroGeoSphere Code of broad-leaved, needle-leaved and mixed-stand floor. First, saturated hydraulic conductivity, porosity, and water retention curve parameters (van Genuchten equation) which were unknown parameters in the forest floor were estimated by inverse method. Second, estimated hydraulic properties were compared statistically.
Forest floor samples were collected from broad-leaved (beech and others), needle-leaved (coniferous) and mixed-stand (coniferous and broad-leaved) trees in Guilan province, Iran. In the laboratory, a plastic wire-netting, composed of 0.3 mm diameter was attached to the bottom of each core sample to support forest floor. Then the samples were piled up to make long columns of 18.1 cm in inner diameter and about 40.88 cm in height. Artificial rainfall experiments were conducted on top of the columns and free drainage from the bottom of columns was measured in the laboratory. Applied rainfall intensities were randomly changed in the range of 0-0.01 cm/sec. Drainage at the bottom of the tray was collected and measured using an electronic balance. First, a constant intense rain was applied to reach to steady state condition as a constant discharge rate from the bottom was established in order to accurately define the initial condition required for the numerical simulation of unsaturated water flow. After reaching to state steady experiment, transient condition was carried out. In transient condition, the random rainfall experiment was conducted and the transient discharge rate from the bottom was continuously monitored. In this study, we developed an inverse method for estimating parameters based on the Levenberg-Marquardt (Marquardt 1963) minimization algorithm in the C programming language along with HydroGeoSphere (Therrien et al., 2008) as a forward model. The model was used to address two specific issues. First, it was used to estimate the hydraulic conductivity, porosity, and soil water retention curve parameters (van Genuchten and Brooks & Cory equations) which were unknown parameters in the unsaturated porous media. Second the water flow in the forest floor was simulated using van Genuchten and Brooks & Cory equations along with HydroGeoSphere code.
The results of calibration periods showed that the estimated free drainage using the optimized parameters exhibits a good fitting with the observed free drainage for all treatments. The good agreement between simulated and observed free drainage in the validation period for all the forest floor samples illustrated that the estimated hydraulic properties efficiently characterized the unsaturated water flow in forest floor. So one could conclude that Richard's equation along with Brooks & Cory and van Genuchten's retention functions can successfully describe the unsaturated water flow in the forest floors.
Estimated hydraulic properties succeeded to reproduce the observed free drainage in the transient condition, indicating van Genuchten functions along with Richard's equation can be used to simulate water flow in the entire forest floors. The results of the study showed that the forest floor samples have large saturated hydraulic conductivity values like light soils. The results showed that inverse method was not sensitive to residual water content. Also the results showed that HydroGeoSphere code along with van Genuchten's retention function with 0.2753 mimics free drainage better than Brooks & Corey's retention function with 0.3400 but there is no significant difference (P