V43G-07:
Trace Hydrophobic Organic Chemicals Present in Pennsylvania Groundwater are Correlated with Geogenic Brines rather than Hydraulic Fracturing Active Zones
Thursday, 18 December 2014: 3:10 PM
Brian Drollette1, Kathrin Shregglman2, Emma D’Ambro3, Martin Elsner4, Nathaniel R Warner5, Megan O’Connor6, Osman Karatum6, Avner Vengosh6, Robert B Jackson7, Thomas Darrah8 and Desiree Plata1, (1)Yale University, New Haven, CT, United States, (2)Helmholtz Center Munich, Oberschleissheim, Germany, (3)Le Moyne College, Syracuse, NY, United States, (4)Helmholtz Zentrum Munchen, Neuherberg D-85764, Germany, (5)Dartmouth College, Hanover, NH, United States, (6)Duke University, Durham, NC, United States, (7)Stanford University, Stanford, CA, United States, (8)Ohio State University Main Campus, Earth Science, Columbus, OH, United States
Abstract:
Recent studies demonstrated that deep Marcellus shale brines migrate into shallow groundwater aquifers, presumably via fractures in the subsurface that exist independent of any gas extraction activities. However, whereas many inorganic species are conservative tracers, hydrophobic organic compounds are both sorptive and reactive, and geogenic organic chemicals may not survive transport from deep shales to the subsurface. Here, 40 shallow groundwater samples from private wells in Northeastern Pennsylvania were analyzed for volatile organic compounds (VOCs) and gasoline range organic compounds (GRO), and 17 were analyzed for VOCs, GRO, and diesel range organic compounds (DRO). BTEX compounds (i.e., benzene, toluene, ethylbenzene, and xylenes) were detected in 6 of 40 samples at concentrations orders of magnitude below EPA maximum contaminant levels (e.g., << 5 ppb) and did not spatially correlate with distance to the nearest active hydraulic fracturing well. GRO was detected in 10 of 40 samples at concentrations as high as 8.8 ± 0.4 ppb and did not correlate with distance to the nearest hydraulic fracturing well (p = 0.24) nor in active fracturing zones, which we defined as sample locations less than 1 km from a well (p = 0.60). However, GRO was strongly correlated (p = 0.004) with shallow groundwater with Marcellus Shale inorganic chemical character, as delineated by inorganic chemical analysis. DRO was detected in all 17 samples up to 158 ± 4 ppb and did not spatially correlate with distance to the nearest hydraulic fracturing well (p = 0.74), nor active zones (p = 0.61). Similar to GRO, DRO did correlate with shallow groundwater containing Marcellus Shale character with moderate significance (p = 0.08). These results indicate that: (a) hydrophobic organic chemicals can survive transport from the deep subsurface to shallow groundwaters, and (b) transport of these compounds is not detectably enhanced by hydraulic fracturing activities in Northeastern PA as of the 2012-2014 summers.
