Air Impacts of Unconventional Natural Gas Development: A Barnett Shale Case Study

Many atmospheric pollutants have been linked to the lifecycle of unconventional natural gas. Attributing air emissions to particular segments of the natural gas life cycle can be difficult. Further, describing individual and community exposure to air pollutants is complex since contaminants can vary spatially and temporally, based on proximity to point sources, magnitude, transport and dispersion of emissions. Here we will present data from the Barnett Shale formation near Dallas/Fort Worth, TX with the goal of providing a better understanding of the extent to which population exposure to air toxics is associated with emissions from natural gas production operations in this region. The Barnett Shale formation covers nearly 13000 km² and is located west of Dallas/Fort Worth, TX. This formation contains natural gas, natural gas condensate, and light oil. Samples were collected in April-May 2010 in two phases with the purpose of Phase 1 being to characterize emissions from major gas production facilities in the area, while Phase 2 involved more intensive monitoring of two residential areas identified in Phase 1. One of the residential areas was downwind of a gas well and two condensate tanks and the other area was close to a compressor station. Phase 1 sampling involved our mobile monitoring system, which includes real-time estimates of volatile organic compounds (VOC), using a portable photoionization detector monitor; continuous NO, PM2.5 mass, and a GasFindIR camera. Phase 1 also included 1-hr integrated canister VOC samples and carbonyl compound samples, using DNPH impregnated Sep-Pac Si cartridges. These samples were analyzed by GC/MS and high performance liquid chromatography with a photodiode array detector. Phase 2 sampling included 7-day integrated passive samples for NOx, NO₂ and SO₂ using Ogawa passive samplers, and BTEX (benzene, toluene, ethylbenzene, and xylenes), 1,3-butadiene, and carbonyl compounds (formaldehyde, acetaldehyde, and acrolein) using Radiello samplers. In addition, weekly PM2.5 samples were collected on Teflon and quartz filters that were analyzed for mass and elements (Teflon filters), for organic and elemental carbon (OC and EC) by thermal/optical reflectance (TOR) method and for polycyclic aromatic hydrocarbons (PAH) using a gas chromatography/mass spectrometry (GC/MS) technique (quartz filters).VOC emissions from condensate tanks were largely low molecular weight hydrocarbons, however these tanks were enhancing local benzene concentrations mostly through malfunctioning valves. PAH concentrations were low (in pg m^-3 range) but the average PAH concentration profiles (higher fraction of methylated PAHs) indicated an influence of compressor engine exhausts and increased diesel transportation traffic. These measurements, however, only represent a small 'snap-shot' of the overall emissions picture from this area. For instance during this one month study, the compressor station was predominantly downwind of the community and this may not be the case in other times of the year. Long-term study of these systems, especially in areas that have yet to experience this type of exploration, but will in the future, is needed to truly evaluate the air impacts of unconventional natural gas development.