Using EC for baseflow separation in a discontinuous permafrost watershed: Crooked Creek, Alaska

N. Utting, W. Wade, C. Thompson, T. Wolley, R.N. Enos

In the proceedings of: GeoMontréal 2013: 66th Canadian Geotechnical Conference; 11th joint with IAH-CNC

Session: Groundwater-Surface Water Interactions II

ABSTRACT: Accurate estimation of baseflow is important for improved calibration of surface water and groundwater models used for stream water budgets and determining the recharge and discharge components of groundwater systems. Baseflow is commonly determined using baseflow recession techniques or low pass filter calculations. These methods assume that a component of the stream flow is derived from surface or interflow runoff while the remaining flow is from the discharge of groundwater. In cold climates, and particularly in permafrost terrain, where the annual hydrograph is dominated by high flows during the annual spring melt period and low flows derived from baseflow during the winter period, this assumption is not valid. One solution to this problem is to estimate the baseflow component in stream flow using the unique electrical conductivity (EC) signatures of groundwater and overland run off. In this study we use EC to estimate the baseflow contribution to surface water along American Creek and Crooked Creek, two perennial streams located in south western Alaska, U.S.A. The assumptions of traditional baseflow methods do not reflect observed flow conditions in either stream. The site is located in discontinuous permafrost where prolonged snow covered periods at sub-zero temperatures limit runoff contributions to the surface water regime during the winter. Quarterly streamflow discharge and water quality data have been gathered at American Creek and Crooked Creek since 2005. The EC of creek water shows a seasonal trend with lower values in summer and higher values in winter. The high values in winter are comparable to the EC measured in groundwater monitoring wells, which is expected as winter low flows should be almost exclusively fed by groundwater. In contrast, the EC of creek water during high flow periods (i.e. spring snowmelt) should be dominated by surface or near-surface runoff. These end values of EC were then used to conduct a two component separation of creek flow. The resulting baseflow estimates were found to be comparable to the baseflow values predicted from a 3-dimensional MODFLOW groundwater model with a recharge rate of 139 mm/yr, or 28% of annual precipitation. EC was found to be an effective way to verify baseflow predictions in an environment where the assumptions of traditional models are not realistic.

Please include this code when submitting a data update: GEO2013_624

Access this article:
Canadian Geotechnical Society members can access to this article, along with all other Canadian Geotechnical Conference proceedings, in the Member Area. Conference proceedings are also available in many libraries.

Cite this article: