A physically based model of heterogeneous hillslopes: 1. Runoff production

Abstract

A fully three‐dimensional model of variably saturated flow on a hillslope has been used to explore the effects of different random patterns of saturated hydraulic conductivity on a 150 m by 100 m hillslope. Both surface and subsurface runoff production are simulated. The model's simulations suggest that peak discharges and runoff volumes are generally increased by the presence of heterogeneity, increasing with increasing variance and spatial dependence of the underlying random field. Simulations using different realizations with the same random field parameters show that for the case of soils showing no dependence, differences in runoff production between realizations are small at this scale of hillslope. The differences increase with increasing spatial dependence, particularly for the case of low‐conductivity soils producing surface runoff.

Dynamical process upscaling for deriving catchment scale state variables and constitutive relations for meso-scale process models

“…Binley et al (1989a) used a full three dimensional model based on Richards equation coupled with a simple linear routing approach for deriving hillslope scale effective hydraulic conductivities for stochastically generated heterogeneous parameter fields. This worked for weakly heterogeneous systems of high average hydraulic conductivities ranging for 0.05 to 0.2 cm/min (compare their Table 2).…”

A conceptual model of daily water balance following partial clearing from forest to pasture

“…Topog (Vertessy et al, 1993) and WEC-C (Water and Environmental Consultants-Catchment) are two other fully distributed models which are applicable to hill slope and experimental scale (Croton and Barry, 2001;Croton and Bari, 2001). Although distributed hydrological models are applied all over the world, it is now well understood that the basic limitations of these models to represent catchment response with a small number of parameters, is due to their inability to reproduce dynamic variation of saturated areas within the catchment (Beven, 1989;Binley et al, 1989;Beven, 2001). In fact, the dynamic variation of the saturated area, a function of accumulation and horizontal movement of water in the top soil layers, is mainly responsible for the highly non-linear nature of catchment response to storm events Todini, 1996).…”

Multiresponse modeling of variably saturated flow and isotope tracer transport for a hillslope experiment at the Landscape Evolution Observatory

“…However, multiobjective parameter estimation for nonlinear or coupled models with a high number of degrees of freedom is very challenging (Anderman and Hill, 1999;Keating et al, 2010), since classical techniques developed for simpler hydrological models (e.g., Gupta et al, 1998;Fenicia et al, 2007) are not readily extendable, in terms of robustness and efficiency, to more complex models. Traditional challenges, on both the experimental and modeling sides, are associated with soil heterogeneity, variability in parameters, and variably saturated conditions (e.g., Binley et al, 1989;Woolhiser et al, 1996;Neuweiler and Cirpka, 2005). An added source of complexity arises when passing from flow modeling to flow and transport modeling (e.g., Ghanbarian-Alavijeh et al, 2012;Russo et al, 2014).…”