Evaluation of alternative groundwater flow models for simulating hyporheic exchange in a small mountain stream

SummaryGroundwater flow models have been used to estimate the amount of exchange flow and the residence time distribution of stream water in the hyporheic zone. However, reliability of these predictions have not been tested. We ask the questions: (1) how reliable are hyporheic groundwater models in typical applications examining hyporheic exchange flows? and (2) how does the reliability change with increased data availability and model sophistication? We developed groundwater flow models of the hyporheic zone for a mountain stream in the HJ Andrews Experimental Forest, Oregon. The models are based on surveyed topography and hydraulic conductivity (K) measurements from both slug tests and a well-to-well tracer test. We developed several models using different methods to estimate two of the most uncertain parameters – K and the depth and shape of the bedrock boundary. We first tested the goodness of fit of each model to the water levels observed in a network of wells and piezometers. Results showed that differences among models in predicted heads were quite small, whereas differences among estimated hyporheic fluxes varied by a factor of two. We then tested the model predictions of tracer arrival times to each well in the network during a stream-tracer injection. Comparison of simulated and observed travel times showed that increased model sophistication did not lead to improved model reliability, because travel time predictions from the homogeneous model were equal to, or better than, the predictions from the heterogeneous models. While general trends in solute breakthrough were correct in the models, K data from even 37 wells in a 15 m by 50 m model domain were insufficient to characterize detailed arrival times accurately. This suggests that geomorphic data may be sufficient to predict water fluxes through the subsurface and approximate travel times. However, for detailed analysis of solute transport pathways and breakthroughs, intensive sampling of the subsurface may be necessary.