Artificial recharge of confined prairie aquifer

Abstract

Artificial recharge of a confined aquifer was successfully conducted through a 30 m deep small diameter well.

The study site selected was located north of Saskatoon, Saskatchewan, over the Dalmeny aquifer system. Farms and towns in the area obtain water from wells developed into the aquifer and from manmade and natural surface water sources. The aquifer covers an 800 km2 region between the North and South Saskatchewan rivers.

Treated fresh water was injected into the aquifer formation by maintaining the water level in the well standpipe above the static water level of the aquifer. The groundwater quality exceeded 2000 mg/L total dissolved solids and was aesthetically displeasing to the well owner because of dissolved iron concentration in excess of 10 mg/L. The injected water contained less than 0.5 mg/L iron, and less than 1000 mg/L total dissolved solids.

Water injected into the aquifer was subsequently recovered by pumping the injection well. The recovery efficiency ranged from 44% to 85% of the theoretical recovery volume, as measured by inorganic chemistry and electrical conductivity. 15% to 36% of the injected water mixed with the natural groundwater and its recovery could not be measured by the adopted methods.

The specific capacity of the aquifer during injection was calculated to be less than 1 m3 day-1 m-1. The calculated specific capacity of the aquifer during recovery was estimated to be greater than 4 m3 day-1 m-1 in 1987, but decreased to less than 2 m3 day-1 m-1 at the end of the field trials.

A preliminary site assessment procedure was established through review of previous work and the findings of the field program.

The field program was limited because the well was over ten years old and had begun to show signs of decreased capacity prior to this research, and because no monitoring wells could be installed in the study area. Supplementary information relating to artificial recharge was obtained through laboratory and mathematical models.

Tests conducted using a laboratory scale sand model demonstrated that density induced flow occurs after fresh water is injected into brackish water. The fresh water front migrated from a vertical to a horizontal position over a period of 100 minutes. This observation, applied to the field site, may partially account for less than 100% recovery efficiency under artificial recharge.

The fresh water storage in a brackish aquifer was estimated mathematically as a function of hydraulic gradient, aquifer parameters and water density. The model indicated density induced flow may contribute to decreased recovery efficiency of stored fresh water, but the model capabilities was limited to demonstrating the process, rather than application and analysis of field data sets.

The net increase of dissolved calcium in the injected water was calculated to be less than 1 mg/L. The increased calcium concentrations of the recovered water quality over time was attributed to the mixing of brackish water with the injected water, not to calcite dissolution from the aquifer matrix.

Artificial recharge and recovery efficiency of fresh water into a confined aquifer in Saskatchewan was estimated assuming changes in water quality due to dispersion. Some additional losses occurred that may be due to density differences between the injected and original groundwater, or due to the regional groundwater gradient.