High-Resolution, Two-Dimensional Groundwater Flow Models based on Direct-Push Electrical Conductivity and Hydraulic Profiling Tool Logs

Abstract

Sedimentary heterogeneity has an important influence on groundwater flow. An accurate prediction of groundwater flow is dependent on an understanding of heterogeneity. Characterizations of heterogeneous conditions in unconsolidated aquifers are often generalized and result in misrepresentations of vertical and lateral sediment variation. Direct-push electrical conductivity (EC) and hydraulic profiling (HPT) log data has been shown to provide a high-resolution interpretation of hydraulic conductivity. When compared to traditional driller’s logs, high-resolution geophysical logging has the potential to change views of groundwater flow and contaminant transport. The purpose of this research was to develop groundwater flow models based on direct-push high-resolution EC and HPT data, and demonstrate the influence of modeled high-resolution heterogeneity on groundwater flow. Groundwater models based on conventional driller’s, EC, and HPT logs were generated to compare groundwater flow. Groundwater flow was simulated with the commonly used groundwater modeling program Visual MODFLOW 4.3. Flow models simulating gradient were run to establish groundwater flow direction and rate. Behavior of equipotential lines responded in agreement to the assigned hydraulic conductivity providing confidence in the modeling approach. Conventional driller’s and HPT flow models indicated that groundwater flow was restricted because of the presence of massive low-K¬ zones. The EC groundwater flow model indicated a more complex flow model because of the presence of preferential flow paths. In the conventional driller’s, EC, and HPT groundwater flow models, local gradients, influenced by lateral variations, were observed within the regional gradient. The sensitivity of heterogeneity was tested using the EC model. The influences of hydraulic gradient, groundwater withdrawal and injection wells, and infiltration recharge basin were evaluated to observe their effect on groundwater flow. The regional gradient simulations indicated the presence of local gradients suggesting varying flow rates through the cross section. The withdrawal well models showed that the presence of flow-restrictive lenses limited the proximity of influence on groundwater flow, while influencing local gradient. Injection simulations indicated that if the head value was not increased above the maximum head of the regional gradient flow direction would remain the same, but the rate of flow up-gradient from the injection boundary would decrease, while the down-gradient side of the boundary would increase. The recharge basin model demonstrated that the local gradient near the recharge boundary would change, decrease on the up-gradient side and increase on the down gradient side. As depth of recharge is increased the gradient on the up-gradient side of the basin decreased. The groundwater flow models confirmed that high-resolution log data provides more detailed information about groundwater than traditional driller’s logs do. The identification of distinct heterogeneous sediment conditions allowed for the observation of sedimentary features that would often go unrecognized, and demonstrated their influence on groundwater flow. This research demonstrates that the implications of being able to observe the influence of preferential flow paths and flow-restricted zones can support predictions of solute movement. This high-resolution groundwater modeling approach offers the potential to redefine how groundwater moves in other areas of the Equus Beds aquifer and in other unconsolidated aquifers, and to improve aquifer management strategies.