2007 GSA Denver Annual Meeting (28–31 October 2007)

Paper No. 5
Presentation Time: 9:15 AM

ESTIMATING SPATIAL AND TEMPORAL GROUNDWATER DISCHARGE THROUGH CLAYEY STREAMBED DEPOSITS USING TEMPERATURE AS A TRACER


CONANT Jr, Brewster1, MCGUINTY Jr, Dalton2, ROBIN, Michel J.L.2 and BUSTROS-LUSSIER, Elyse2, (1)Department of Earth Sciences, University of Waterloo, 200 University Ave. W, Waterloo, ON N2L 3G1, Canada, (2)Department of Earth Sciences, University of Ottawa, Marion Hall, 140 Louis-Pasteur, Ottawa, ON K1N 6N5, Canada, bconantj@uwaterloo.ca

Accurately characterizing spatial and temporal variations in groundwater discharge through low hydraulic conductivity streambed deposits is difficult to achieve using conventional approaches like seepage runs, streambed piezometers, and seepage meters. These techniques can be labor-intensive and may lack the spatial resolution and accuracy necessary to fully characterize groundwater discharge. Techniques involving temperature as a natural tracer are cost effective and have the potential to overcome these limitations. For this reason, temperature tracing techniques were used to characterize groundwater/surface-water interactions on different scales for rivers with silty and clayey streambed deposits in the Raisin River watershed in eastern Ontario as part of the Watershed & Environmental Resource Assessment Project. An initial survey of over 100 km of the rivers using an electrical conductivity and temperature drag-probe showed almost no measurable (i.e., high) groundwater discharge zones. Spatial variability in discharge was then characterized at three locations on a reach scale (each 20 to 28 m long). Streambed temperature measurements made on a 1- to 4-m grid pattern at a depth of 0.5 m were converted to fluxes using the Turcotte and Schubert analytical solution for the 1-dimensional heat flow equation. Again, no significant high flux discharge zones were identified. Calculated fluxes were low and relatively uniform and ranged from 1.2 to 23.8 Lm-2d-1 and were higher than the 0.5 to 4.5 Lm-2d-1 range observed using seepage meters. The apparent discrepancy between flux ranges may be a result of measurements being made at different times or transient temperature effects. At three locations, vertical profiles of temperature in the top 1.5 m of the streambed was measured hourly for over a year and is currently being simulated using the USGS VS2DH numerical heat flow model to better resolve the magnitude and temporal variation of fluxes. Hourly water level data for the rivers and streambed piezometers showed upward-flow hydraulic gradients (even during storm runoff events) and indicated relatively uniform groundwater discharge conditions over time. Temperature tracing methods provided valuable information on groundwater discharge at different scales and over time for the low hydraulic conductivity materials.