Assessing the Connectivity of Groundwater Wells to Surface-Water Using a Volumetric Capture Delineation Tool

Abstract

Existing groundwater modeling methods for determining the degree of groundwater well connectivity to surface-water either provide weak guidance (i.e., analytical methods), only examine the hydrological impacts of pumping (i.e., mapping method developed by Leake et al. (2010)), or are too computationally expensive (i.e., solute transport modeling methods) (Ceric & Haitjema, 2005). As such, a modeling technique is required that can accurately estimate the amount of water a particular well sources from surface-water at low computational costs. Here, we present some novel applications of a software tool called FlowSource (Black and Foley, 2013) in assessing well and surface-water connectivity. FlowSource is a MODFLOW-based (Harbaugh et al., 2000) software and can completely describe the flow connectivity between specified parts of the aquifer system without the use of particle tracking or advective transport simulation. First, an algorithm was developed to reduce numerical dispersion in FlowSource calculations that result from the assumption of fully mixed groundwater model cells. For any given set of flows along the faces of a three-dimensional rectilinear finite-difference groundwater model cell, the algorithm can evaluate the volumetric flow from each inflow face to each outflow face by either applying mass balance inside the cell or analytically recreating the internal streamtube geometry using the semi-analytical particle tracking method developed by Pollock (1988). The calculations of the algorithm are exact. The algorithm may be applied on a cell-by-cell basis to establish the volumetric flow connections inside the cell before performing the flow connectivity calculations in FlowSource. Its potential to reduce numerical dispersion in FlowSource calculations is demonstrated with a hypothetical example. Second, a novel FlowSource-based modeling tool was developed for assessing well and surface-water connectivity. The tool deploys FlowSource within the framework of the mapping method developed by Leake et al. (2010) (the LRD method). The differences in the results of the novel tool and the LRD method are demonstrated. Numerous applications of the modeling tool are demonstrated using a synthetic model, which include the ability to rapidly: (1) screen for wells that are surface-water dominated and are at risk of contamination from non-point sources, (2) infer the volume of water removed by pumpage from surface-water and, (3) generate diagnostic maps that illustrate how the location and time of pumping affects: (a) volumetric connectivity with surface-water, (b) the connectivity of existing wells to surface-water and, (c) hyporheic flows. Most importantly, the use of FlowSource enables these assessments to be performed without performing any solute transport modeling. Hence, the modeling tool developed here has the potential to aid water managers to inexpensively assess the risks posed to public health and/or the local environment of existing and planned pumping operations so that they can effectively prioritize monitoring and modeling efforts.