2006 Philadelphia Annual Meeting (22–25 October 2006)

Paper No. 2
Presentation Time: 1:30 PM-5:30 PM

HIGH-RESOLUTION MONITORING OF ACCIDENTAL DRAINAGE OF A CARBONATE AQUIFER, MONROE COUNTY, WEST VIRGINIA


CHECK, Daniel, Office for Terrestrial Records of Environmental Change, Department of Geology, University of Akron, Akron, OH 44325-4101, SASOWSKY, Ira D., Dept. of Geosciences, The University of Akron, Akron, OH 44325-4101 and QUICK, Tom J., Department of Geology, University of Akron, Akron, OH 44325-4101, dcheck@uakron.edu

In 1997, drilling of a residential well resulted in catastrophic, ongoing leakage from one or more aquifers into Scott Hollow Cave, at a depth of >100 m below land surface. The cave system is in the Appalachian Plateau of southeastern West Virginia, within the Hillsdale Limestone and Maccrady Shale. Initial spot measurements of solutes showed that the leaking water had low nitrate, high sulfate, and was saturated with respect to calcite. This indicated long aquifer residence time, and minimal recharge from overlying agricultural lands. Spot measurements of discharge showed a trend of decreasing flow. This, in concert with loss of yield from surrounding wells, suggested that the drainage was causing substantial dewatering of one or more aquifers.

To evaluate the long term effects on the aquifer(s), we undertook high resolution monitoring of the leaking water. We installed a custom built orifice weir, a transducer stage recorder, and conductivity and temperature sensors. The parameters are collected at 20-minute intervals on dataloggers. In addition, we conduct grab sampling for pH and major ions at selected times, to allow conductance to be used as a proxy for overall water chemistry.

Discharge has varied between 3.9 and 8.1 m3/hr over the period of investigation. The data show a rich structure not apparent from the initial spot interval sampling. Three different modes of variance occur. First, on a scale of months, a gentle downward slope (0.3 m3/hr/month) is seen, showing continuous drainage of the aquifer. Second, superimposed on this is a coarsely-sinusoidal undulation on the scale of days, and with magnitude +/- 0.25 m3/hr. Finally, on the scale of hours, frequent downward spikes of magnitude <0.5 to 3 m3/hr are seen. These we interpret as reaction to the pumping of residential wells in the area. The rapidity of the response in these spikes shows involvement of a confined aquifer. It is likely that most of the leakage is sourced from an underlying artesian aquifer, and that one or more overlying aquifers are now depleted due to drainage in to the cave. Challenges to understanding the hydrology include: unresolved structural complexity of the rock units (due to limited mappable exposures and lack of borehole data), construction of monitoring facilities in the cave environment, and remoteness of the leakage (for spot sampling).