Estimation of Groundwater Inflow to the Vermillion River from Observations of Stream Flow and Stream Temperature

Ben Janke, William R. Herb, Omid Mohseni, Heinz G. Stefan

Research output: Book/ReportOther report

Abstract

A model has been developed for estimation of groundwater inflow to a stream reach from observations of stream temperature, groundwater temperature, stream flow rate, and standard weather parameters. The purpose of this model is to provide an estimate of groundwater inflow rate for stream reaches where groundwater inflow is significant. This information is useful for management of fisheries and urban development in watersheds where stream temperature is a concern. In particular, the model was developed for use in the Vermillion River, which has designated trout stream reaches that may be impacted by development in the watershed. The model estimates groundwater inflow rate from a stream reach heat budget, which takes into account atmospheric heat flux, sediment-water heat exchange, and groundwater inflow. The model requires the following data as input: stream temperature at the upstream and downstream ends of the stream reach, stream flow at either end of the reach, standard weather data, and no significant tributaries or inflows between the ends of the reach. The model was applied to six reaches in the Vermillion River watershed. Estimated groundwater inflow rates showed considerable spatial and temporal variability, both seasonally and between the two years simulated (2006 and 2007). In North Creek, groundwater inflow rate was 0.45 to 1.30 cfs/mile in 2007; in the upper Vermillion River main stem for the same period estimated inflow rates ranged from 0.15 to 3.87 cfs/mile. In the middle Vermillion River main stem, estimated inflow rates were unnaturally large and more variable (0.39 to 11.1 cfs/mile); these estimates include significant tributary inflow, which is lumped with groundwater inflow in the model. This, along with the failure of the model for reaches or periods involving high stream flows, is the likely source of the over-predicted groundwater inflow values. Simulations for lower South Creek showed negligible groundwater inflow for 2006; results for lower South Branch were very typical of a groundwater-fed stream, with relatively constant groundwater inflow (around 1.0 cfs/mile) that fluctuated only slightly during periods of heavy rainfall. A comparison of predicted groundwater inflow rates throughout the watershed for both dry (baseflow) and high-flow conditions suggest the presence of shallow groundwater, particularly in the lower reaches of the watershed. The significant variability in groundwater inflow rate predicted by the model can be traced to a number of factors, including data quality and sensitivity of the model to groundwater temperatures, stream shading/sheltering, and especially to stream flow. An extensive sensitivity analysis of the model is presented in this report, as well as an analysis of available data, in particular, groundwater temperature. Limitations of the heat budget approach to modeling groundwater inflow rate are also discussed and criteria for application of the model are developed from the results of sensitivity analysis.
Original languageEnglish (US)
StatePublished - Nov 2009

Fingerprint Dive into the research topics of 'Estimation of Groundwater Inflow to the Vermillion River from Observations of Stream Flow and Stream Temperature'. Together they form a unique fingerprint.

Cite this