Modeling flow and transport in irrigation catchments 1. Development and testing of subcatchment model

Abstract

The development of a physically based model for flow and transport in hillslope transects is presented. In the second paper of this set, Connell et al. [this issue] use this model as the fundamental unit in a catchment application to describe flow within flood-irrigated catchments. The hillslope transect model is based on the assumption that surface and groundwater flow is one-dimensional, flowing down the topographic gradient. This approximation is particularly accurate for flood-irrigated catchments where the land surface is divided into irrigation bays. To maximize the model's utility for catchment modeling, a number of simplifications to the process descriptions are introduced to reduce model complexity and computational overhead. The description of the unsaturated zone makes a distinction between infiltration and capillary rise, allowing infiltration to be a mixture of macropore and matrix flow and capillary rise to be a porous medium process. The capillary rise description exploits the transition in flow behavior as soil dries from being atmosphere-limited to profile-controlled. An efficient Laplace space quasi-analytical solution to Richards' equation is used for this second stage. This quasi-analytical procedure is also used to solve the advection-dispersion equation for unsaturated solute transport and for Darcian groundwater flow. The model is tested through application to a field site where detailed observations of surface and subsurface flow and transport are available. The good agreement with observations in both calibration and long-duration simulations of a wide range of physical processes promotes confidence in the model representation.