Paper No. 10
Presentation Time: 10:45 AM

STREAM METHANE TRACER TEST FOR EVALUATING GROUNDWATER IMPACTS ASSOCIATED WITH HYDRAULIC FRACTURING


HEILWEIL, Victor M.1, STOLP, Bert J.2, SUSONG, David D.3, KIMBALL, Briant A.4, ROWLAND, Ryan C.1, MARSTON, Tom M.1 and GARDNER, Philip M.5, (1)Utah Water Science Center, U.S. Geological Survey, 2329 Orton Circle, Salt Lake City, UT 84119, (2)U.S. Geol Survey, 2329 W Orton Cir, Salt Lake City, UT 84119, (3)U.S. Geological Survey, Utah Water Science Center, 2329 W. Orton Circle, Salt Lake City, UT 84119, (4)U.S. Geological Survey, 2329 Orton Cir, Salt Lake City, UT 84119-2047, (5)Water Resources, U.S. Geological Survey, 2329 Orton Circle, Salt Lake City, UT 84119, heilweil@usgs.gov

A gaining stream (derived from groundwater discharge) in a watershed can provide an integrated signal of the groundwater system. Our hypothesis is that impacts on groundwater quality from hydraulic fracturing of underlying natural gas reservoirs may be evaluated at the watershed scale by sampling for dissolved methane gas in gaining streams. While conservative ionic tracers such as bromide can remain in the stream water for long distances, dissolved gases such as methane exchange with the atmosphere and are not conservative. The persistence of methane in stream water is a function of the gas-exchange velocity. To determine the gas exchange velocity in a natural stream environment, a methane tracer test was conducted on Nine Mile Creek, a small constant-gradient, well-aerated gaining stream in a natural-gas development area in central Utah. Methane was injected into the stream using gas-permeable silicon tubing at a flow rate of about 350 cubic centimeters per minute (about 0.4 moles per hour). This increased the dissolved methane concentration of the stream by about 15 ppb at the point of injection. A bromide tracer was also added to quantify the locations and rates of groundwater discharge to correct for tracer dilution. Results indicated that methane remained dissolved in the stream water for more than 1.5 km downstream from the injection site. Preliminary modeling of this field-scale data indicates a gas-exchange velocity of about 6 m/d, perhaps the first reported value for methane gas. The test indicates that sampling dissolved gases in streams may be a promising tool for evaluating both natural methane sources and induced methane leakage (associated with hydraulic fracturing) from gas reservoirs. Such an approach has the advantage of using stream-integrated chemical signatures to indirectly monitor groundwater/surface-water interaction and potential aquifer contamination at the watershed scale, rather than relying on point-specific monitoring of individual groundwater wells.