Fingerprinting of gas contaminating groundwater and soil in a petroliferous region, Alberta, Canada by Barb Tilley and Karlis Muehlenbachs, In Proceedings from International Network of Environmental Forensics conference, Cambridge, UK, July 25-27, 2011
Public attention to possible water well contamination by CBM activity in Alberta originated in the hamlet of Rosebud, Alberta. A very high density of gas wells occurs within the vicinity of this hamlet, up to as many as eleven producing gas wells within one square mile block. The first complaint was submitted in 2004, which was soon followed by complaints from most of the other landowners in the hamlet. The need for water well testing is evident when the geology is taken into consideration. Figure 9 is a schematic showing the general geology and setting of CBM production wells and domestic water wells in Alberta. Multiple coal seams occur near the surface and the shallowest coal seams are often used as aquifers. The various levels of well casings designed to protect against gas migration are illustrated. However, the numerous surface casing vent flow gases that we have analyzed indicate that the release of hydrocarbon gases to the shallow environment occurs regardless. Gas production is from multiple coal seams that are fractured and the gases from several coal horizons are generally commingled. The carbon isotopic composition of gases from water wells (seven from Rosebud itself, 17 from presumed pristine water wells) are compared to oil field gases (production gases from 3 different depth intervals, shallow CBM wells, and surface casing vent flows (SCV)) within a 40 km radius of the hamlet of Rosebud (Figure 10). At first glance the Rosebud water well gases are not identical to the resource gases. However, they may be impacted by mixing. Mixing of presumed pristine well water gas with 3 different resource gases (CBM gas at 163-399m, production gas from 526-614m, and production gas from 614-682m) has been modelled using the mass balance equations of Jenden et al.3 The range in each of the mixing curves on Figure 10 represents a variation in the estimated proportion of ethane in the pristine water well between 0.02 and 0.005%. The wide range in δ13C values of SCV gas indicates that there is a history of problem gas flow from a wide range of depths in this area. Comparison of the δ13C values of methane and ethane for gas from the problem water wells in Rosebud with the calculated mixing curves of Figure 10 shows that gases of waters from all seven wells have mixed with either CBM or the deeper production gases. Gases from well waters J and D fall on the mixing curve with the deeper production gas at 614-682m. A deep gas, rather than a shallow CBM, impact on well D is supported by well D’s history of gas loading many years before any CBM development began in the area. Water wells K and L fall on the mixing curve for CBM-water gas, with well L apparently the most contaminated. Water well P lies closest to the mixing curve for production gas from 526-614m, whereas water wells S and R lie above and below the CBM-water gas mixing curve. Following the mixing curves, the isotopic differences between waters S, R, K-L and J-D suggest different contaminant sources. These sources cannot be identified unambiguously. They are certainly not from the deepest production zones (1205-1417m), nor the shallowest CBM horizons (<160m).
Environmental Forensics Proceedings from International Network of Environmental Forensics conference Cambridge, UK, July 25-27, 2011, Editors Robert D Morrison and Gwen O’Sullivan, ISBN: 978-1-84973-372-4, 2012