Modelling leakage from perched rivers using the unsaturated flow model VS2DTI
Jackson, C.. 2005 Modelling leakage from perched rivers using the unsaturated flow model VS2DTI. British Geological Survey, 46pp. (IR/05/019) (Unpublished)Before downloading, please read NORA policies.
The objectives of the modelling work are two-fold. The first aim is to investigate and verify the adequacy of the relationship between water table elevation and river leakage that is used to represent river-aquifer interaction in saturated groundwater flow models, such as MODFLOW McDonald and Harbaugh, 1988). The second objective is to highlight some of the tasks involved in the process of constructing an unsaturated groundwater flow model to contribute to the development of BGS best practice guidelines for numerical modelling. River-aquifer interaction is usually represented in saturated groundwater flow models by a simple Darcian type leakage mechanism, which depends on the head in the aquifer, the river stage and the hydraulic properties of the river. This representation of the river assumes that the channel is rectangular, that it banks are vertical and impermeable and, that a bed separates the surface water from the aquifer. This relationship between groundwater head and leakage from a river in a coupled surface water-groundwater system, which is implemented in MODFLOW and commonly applied in other saturated groundwater flow model codes, is discussed. To determine the accuracy of this simplified relationship, the unsaturated flow model code VS2DTI (Hsieh et al., 2000) is used to simulate a surface water-groundwater system in detail. The process of developing an unsaturated flow model using this software is described. The shape of the curve describing the relationship between water table elevation and river leakage, which is included in MODFLOW and other saturated groundwater flow models, is similar to that which is simulated using the unsaturated flow simulation code, VS2DTI. Under perched conditions the leakage from a river reaches a maximum value as the water table is lowered. However, this maximum leakage rate cannot be assumed to occur immediately as the water table falls below the bed of the river. The assumptions that it is only the aquifer medium that governs the rate of flow to the water table and that this occurs through a saturated column with a width that is equivalent to that of the perched river is simplistic. The approaches to modelling perched rivers that are incorporated into saturated groundwater flow models based on these assumptions are likely to be adequate for regional scale models, in which the detailed flow pattern beneath the river is not of interest, but this will not be the case for the simulation of channel scale river-aquifer interaction processes. The rate of leakage from a perched river is complex and governed by the following features: • The river stage and water table elevation. • The vertical hydraulic conductivity of the river bed and the aquifer. • The relationship between pressure head and moisture content and between hydraulic conductivity-moisture content for the porous media. • The horizontal hydraulic conductivity of the aquifer, which will affect the width of the column of vertically flowing water. • The width and length of the river channel. • The thickness of the river bed deposits. • The shape of the river channel. Further work is requires to understand the process of river aquifer interaction in different hydrogeological settings. This should involve a literature review of the research undertaken to date and the simulations of different types of river-aquifer systems, for example in which the shape of the river channel is more complex.
|Item Type:||Report (UNSPECIFIED)|
|Programmes:||BGS Programmes > Other|
|Additional Information:||This item has been internally reviewed but not externally peer-reviewed|
|Additional Keywords:||GroundwaterBGS, Groundwater, Groundwater modelling|
|Date made live:||24 Aug 2010 15:35|
Actions (login required)