1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 The Upstate New York Society for Risk Analysis Webinar Series, Scientific Studies on Impact of Natural Gas Extraction from Marcellus Shale on Water Resources. Prepared by Peter McClure, Senior Toxicologist, SRC, Inc. Syracuse, New York October 2014 With the development of horizontal drilling and hydraulic fracturing, U.S. reserves of natural gas in shale formations have become commercially viable sources of energy with national energy independence benefits and potential benefits to the environment, compared especially with coalbased energy generation. The depth of gas-containing shale formations can vary across regions, but most are thousands of feet below the surface. Modern gas extraction from shale formations entails multiple processes including: well pad construction; vertical and horizontal drilling of wells; hydraulic fracturing of shale formations with pressurized suspensions of water (~90%), sand (i.e., proppant) (~9%), and other chemicals (~0.5-2%) to release gas trapped in shale; capturing waste water and released gas from the wells; and treating or disposing of waste water. Each time the shale is fractured, ~3-5 million gallons of water are used, and considerable water flows back to the surface. The wastewater, which contains fracturing chemicals and potentially other naturally occurring chemicals extracted from the shale (such as chloride, bromide, other metals and radionuclides), is typically stored in ponds at the site. More recently, wastewater has been reused on site for additional fracturing events. Wastewater is transported from the site by trucks for treatment or disposal by injection into deep underground disposal wells. Collected gas is transported from storage tanks at well pads by truck to pipeline distribution centers. Gas extraction from U.S. shale formations has rapidly grown over the past 15 years. Public concerns have been growing in parallel due to potential groundwater and air contamination, limited regulation, and possible health impacts. 1
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 Applying the risk analysis processes of toxicity assessment (i.e., hazard identification and doseresponse assessment), exposure assessment, and risk characterization holds promise to assess potential health risks to gas extraction workers, populations living close to gas extraction sites, and terrestrial and aquatic ecosystems. Risk assessments can then aid in making decisions to minimize potential risks. However, current information is insufficient to conduct reliable regional or site-specific health risk assessments for gas extraction by hydraulic fracturing. Knowledge gaps include: the identity and concentrations of chemicals in drilling fluids, fracturing fluids and wastewater; toxicity of chemicals and mixtures used and produced in various phases of gas extraction; frequency and magnitude of spilling events at well sites; and the extent of gas leakage from wells to the surrounding atmosphere and ground and surface water. This series of teleseminars was organized to present examples of research efforts and plans to fill knowledge gaps and better assess potential health and environmental risks presented by the expanding U.S. gas extraction industry. Slides for all six presentations are available at the http://www.sra.org/upstateny. January 2014: Jeanne Briskin of the U.S Environmental Protection Agency (EPA) provided an overview of EPA s Study of the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources. The study s purposes are to assess whether hydraulic fracturing can impact drinking water resources and identify the severity and frequency of any impacts. The study includes 17 research projects. Impacts on drinking water availability in the semi-arid upper Colorado River and the humid Susquehanna River basins are being modeled under different water usage scenarios. Data submitted to the national hydraulic fracturing chemical registry FracFocus (fracfocus.org) are being analyzed with respect to water use, proppants, and chemicals used. FracFocus was created to provide public access to lists of chemicals used in individual wells. For chemicals identified, toxicity assessments are planned based on available data, but Dr. Briskin emphasized that conducting a risk assessment for hydraulic fracturing is not part of this EPA study. Databases of spills in a number of states with hydraulically fractured gas wells are 2
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 being analyzed to understand frequency of spills on or near well pads and components of wastewater. Water treatability studies and source apportionment studies are also being conducted to determine discharges from hydraulic fracturing wastewater, coal-fired power plants, mines, and road salt. A draft final report [planned for late 2014 (epa.gov/hfstudy)] is expected to identify exposure pathways of greatest concern and identify specific knowledge gaps and information needs to support future risk assessment efforts. February 2014: Robert Jackson of Duke University described research activities to assess stray gas in residential water wells in proximity to hydraulic-fractured gas wells and gas leaks from gas well pads and pipelines in urban areas. In samples from 141 residential drinking water wells in northeastern Pennsylvania, no evidence was found for elevated concentrations of salts, metals or radioactivity, but evidence of elevated concentrations of methane, ethane, and propane were found in some of the wells located within 1 km of active gas wells. For example, methane concentrations in 11/59 sampled water wells within 1 km of gas wells were > 28 mg/l, the upper end of the range recommended by the U.S. Department of Interior for mitigation. In some cases, carbon isotopic signatures, hydrocarbon ratios (methane to ethane and propane), and the ratio of the noble gas 4 H to methane suggested contamination from poor well construction, but other sources of methane contamination could not be ruled out, such as older abandoned oil and gas wells, subsurface coal bed methane, or bacterial decomposition. One research approach for better determination of the source of gas contamination would involve studies of water quality measures, before, during, and after drilling, fracturing, and extraction phases. Dr. Jackson briefly noted another study conducted with the U.S. Geological Survey that found no evidence of gas contamination in shallow groundwater samples in the Fayetteville Shale gas production area in Arkansas. The results suggest that geological characteristics may influence the potential for contamination. Dr. Jackson commented on the clear benefits to air quality in switching from coal to natural gas for electricity generation (e.g., less sulfur dioxide, mercury, nitrous oxide, and particulate emissions). He also provided brief overviews of studies examining spatial distributions of methane losses from natural gas pipelines in Boston and Washington. Dr. Jackson noted that the detection of many leaks in both of these urban areas highlights an immediate opportunity to tighten the U.S. natural gas supply chain. 3
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 March 2014: Trevor Penning of the University of Pennsylvania presented an overview of processes involved in extraction of gas from shale formations and recommendations for research to improve understanding of the possible risks to human health and the environment. Challenges of applying risk assessment methods were discussed including: The diversity of chemical additives (and their potential health effects) in fracking fluid; Evidence of elevated concentrations of a diversity of pollutants in flow-back fluids and wastewater including metals and salts, volatile organic chemicals, and radioactivity; Reports of contamination of ground water in gas extraction areas with salts, metals, and gas; and The potential for air pollution in the vicinity of well pads due to diesel exhaust from extensive trucking activity, flaring of gas, and leakage of natural gas and other volatile organic chemicals from wells and storage units. Dr. Penning outlined research recommendations essential to assess potential public health risks from modern gas extraction and identify reasonable approaches to risk mitigation. Research recommendations were made in the areas of water contamination, air pollution, risk communication, and epidemiology; these are outlined more specifically in Dr. Penning s slides. April 2014: Reynold Panittieri of the University of Pennsylvania presented the results of a study to test the hypothesis that increases in health care utilizations are associated with gas well density and well water quality in northern Pennsylvania counties and zip codes. The study included comparisons with health care utilizations in adjacent southern New York counties, located within the Marcellus Shale, but in which hydraulic fracturing gas extraction is not permitted. The study found that hospitalization rates differed between counties in northeastern Pennsylvania and southern New York. From 2007-2011, increases in rates of hospitalization for some causes (e.g., general surgery, neurology, oncology) were found in Pennsylvania counties, compared with southern New York counties. For 2007-2011, an association was observed between hospitalization rates and gas well density in northern Pennsylvania counties. Dr. Panittieri mentioned that an ongoing study in collaboration with scientists at Columbia University is collecting information on well water quality variables (e.g., concentrations of 4
120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 metals and radioactivity) in the same region and examining if associations exist with gas well density and location, and hospitalization rates. May 2014: Sheila Olmstead of the University of Texas and Resources for the Future discussed impact and policy implications for shale gas development regarding Pennsylvania rivers and streams. A database of surface water concentrations of chloride ion and total suspended solids collected in various sites in southeastern and southwestern Pennsylvania between 2000 and 2011 was analyzed by regression analysis to examine possible relationships with proximity to shale gas wells and wastewater treatment plants accepting shale gas waste water. Results of the analysis indicated that shale gas development processes contribute to contamination of streams with chloride (associated with wastewater treatment plants accepting shale gas wastewater) or total suspended solids (associated with well pad density). Dr. Olmstead pointed out that hydraulically fractured gas wells in Pennsylvania are expected to increase in number from about 7,000 in 2013 to about 60,000 in 2030, raising concerns that the scale of contamination will increase. With respect to policy implications, Dr. Olmstead proposed that: 1) increased volume of shale gas wastewater could be dealt with by increased treatment and disposal costs at wastewater treatment plants, which are regulated under the Clean Water Act; and 2) contamination of streams with suspended solids from well pads could be solved by eliminating the exemption of oil and gas construction sites from Clean Water Act regulations involving erosion and runoff controls. Dr. Olmstead briefly described an ongoing project to statistically analyze a Pennsylvania Department of Environmental Protection database of chemical characteristics of shale gas waste water and drilling fluid waste sent to wastewater treatment facilities in Pennsylvania between 2008-2011. Though the results were not ready for presentation in this talk, the study is expected to be useful in evaluating current and future wastewater treatment technologies and capacities. June 2014: Pouné Saberi of the University of Pennsylvania discussed the results of a pilot field survey of health perception in patients with medical complaints in a primary care medical office in a northern Pennsylvania county with hydraulic fracturing gas extraction activity. One hundred and fifty-nine patients were asked to participate and 72 responded. The survey asked 5
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 what kinds of symptoms were experienced, when they began, and what the patients thought was the possible cause. Five causes were listed with the option for the patient to specify other causes: high tension power lines, living near highways, aging from free radicals, natural gas extraction activity, antibiotics in food, or other causes. Sixteen of the 72 respondents (22%) expressed concern about health impacts of gas extraction activity: 9 (12.5%) attributed symptoms to gas extraction activity, 5 did not attribute symptoms to gas extraction, but expressed concern, and 2 were asymptomatic but expressed concerns. The nine subjects with symptoms attributed to gas extraction reported symptoms across a number of organ systems including mental health (sleep difficulty and anxiety): head, ears, throat; neurological (e.g., dizziness, trembling of hands), gastro-intestinal, and cardiovascular (palpitations). Dr. Saberi noted that only 1 out of 6 medical records showed concordance between the patient-reported symptom and medical provider documentation and that a characteristic symptom profile associated with gas extraction was not evident in this limited survey. Dr. Saberi concluded that the pilot study suggests there is substantial public concern about adverse health effects of natural gas extraction in northern Pennsylvania, and that these concerns may not be recorded in medical records. She recommended that, to address public concerns, further efforts should be made to determine possible associations between hydraulic fracturing gas extraction activities and health effects. In summary, the six researchers in this 2014 teleseminar series highlighted the need for additional data and analysis to support future risk assessments for potential human and environmental health effects associated with hydraulic fracturing. More extensive exercises of cycles of research and analysis (analytic deliberative process) are necessary to support public discourse about perceived risks associated with hydraulic fracturing and regulatory decisionmaking to minimize risks. 6