Detecting preferential seepage along casings installed in fractured rock aquifers

Sandra K. Richard, Romain Chesnaux, Alain Rouleau

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

Session: General Hydrogeology I

ABSTRACT: Installing a well in a fractured rock aquifer requires a good sealing of the casing into the bedrock to avoid preferential seepage along the casing from the granular aquifer to the fractured rock aquifer. An improperly sealed casing can generate a direct hydraulic connection between both aquifers (Chesnaux et al 2012). Such connections can have important consequences regarding groundwater quantity and quality (Chapuis et Sabourin 1989). Cross-contamination issues between aquifers can indeed occur in the presence of a hydraulic gradient between the two aquifers. This study proposes a methodology to detect a preferential flow along an improperly sealed casing based on interpreting the hydraulic responses to a pumping test and a variable-head permeability test. Three experimental sites in the Saguenay-Lac-Saint-Jean region in Quebec (Canada) are considered: PZ-35, PZ-104 and PZ-S18. The PZ-35 site includes the PZ-35-R well installed in a crystalline gneissic rock aquifer and a pair of piezometers PZ-35-DD and PZ-35-SD installed as nested piezometers located 5 meters away from PZ-35-R in the upper granular layers (Figure 1a). The second experimental site includes the PZ-104-R well installed in a limestone aquifer and a pair of piezometers PZ-104-DD and PZ-104-SD installed as nested piezometers located 4 meters away from PZ-104-R in the upper granular layers (Figure 1a). The last site includes the PZ-S18-R well in a limestone aquifer and the PZ-S18-D piezometer located 2.6 meters away from PZ-S18-R in a fine sand granular formation, separated from the bedrock by a till unit 6.1 m thick (Figure 1b). Figure 1. Simplified aquifer scheme: a) of the PZ-35 and PZ-104 sites; b) of the PZ-S18 site. Pumping tests were performed in the PZ-35-R, PZ-104-R and PZ-S18-R wells, with water level measurements carried out both in the bedrock well and in the overlaying granular aquifer. The water level response to pumping in the pumping well and in the observation wells is analyzed using the log-derivative drawdown as a function of time. The log-derivative drawdown curve is indeed more sensitive to variations induced by the different hydraulic conditions encountered during the test (Bourdet et al. 1983; Rafini and Larocque 2009). Figure 2 (a,b,c) illustrates the results of the tests conducted in the pumping wells of the sites PZ-S18-R, PZ-104-R and PZ-35-R respectively. Figure 2. Pumping tests results in the fractured rock formations: drawdown and log-derivative drawdown versus time (diagnostic plots) for a) PZ-S18-R and PZ-S18-D, b) PZ-104-R and PZ-104-DD and c) PZ-35-R and PZ-35-DD. At piezometer PZ-S18-D no variation in the groundwater level was observed during the pumping of PZ-S18-R. This suggests that the granular aquifer is not hydraulically connected with the fractured rock aquifer at this site. For the two other sites, the overlaying granular aquifer reacts rapidly to pumping in the fractured rock aquifer, which indicates a ˘ˇˆ˙ˇˆ˙˝ˇ˛ˇˇˆ˙ˇˆ˙ˇ˛ˇ˚˜ˇˆ˙ˇˆ˙ˇ˛ˇˇ˛ˇ ˆ

Please include this code when submitting a data update: GEO2013_336

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: