Modeling the effect of slope aspect on temporal variation of soil water content and matric potential using different approaches by HYDRUS-1D

Abstract

Slope aspect affects microclimate and soil moisture conditions, and thus has an important effect on the hydrological cycle. Given the lack of precise field data on this effect, two soil temperature, moisture and matric potential monitoring sites on north- and south-facing slopes have been established since 2017 in sandy loam soil under beech forest in a small valley near Ebergötzen, central Germany. HYDRUS-1D software was used with different approaches (including Rosetta, RETC, and RETC–measured θs approaches) for deriving soil hydraulic parameters of the van Genuchten model, to simulate temporal variability of soil water content and matric potential in the two slope aspects. Due to different microclimatic conditions, the south-facing slope was warmer and drier than the north-facing slope (two-years average temperatures 9.71 and 9.36 °C, and two-years average soil water contents 0.17 and 0.26 cm3 cm-3, respectively). The model was satisfactorily calibrated and validated on the two slopes. The calibration for the wet period was better on the south-facing slope (R2 = 0.772) than on the north-facing slope (R2 = 0.690). The simulation performance by HYDRUS-1D was better in the wet period than in the dry period. In the dry period, the model was better calibrated for the north-facing slope (R2 = 0.880) than for the south-facing slope (R2 = 0.691). The better calibration results for the wet period of south-facing slope can be explained by the location of the majority of measured soil water retention data which were in the near-saturation zone on the north-facing slope, and discontinuity of second-order derivative of the van Genuchten equation at saturation point when shape parameter (n) is smaller than 2. According to the model ability and different datasets used for the model validation, the north-facing plot was validated better than the south-facing plot for both wet and dry periods. Given that Rosetta was developed based on a soil hydraulic dataset from temperate climates, and the shortcomings of the hydraulic model, our results indicated that the Rosetta–derived approach was a reliable source for soil hydraulic parameters in both slope aspects for the wet period. However, the RETC–derived parameters were more accurate than Rosetta for the dry period on the north-facing slope.