Elevated Heavy Metals Near Natural Gas Extraction Sites in the Barnett Shale


In a recent peer-reviewed study published in Environmental Science and Technology, an American Chemical Society journal, our team of scientists from The University of Texas at Arlington sampled 100 private water wells to assess the potential effects of natural gas extraction on water quality in the Barnett Shale.  Our analyses revealed levels of heavy metals above the Environmental Protection Agency’s Maximum Contaminant Limit (MCL) in private well water samples located near natural gas extraction sites.

The Barnett Shale, a 48,000 km2 shale formation located 1500-2400 meters below 17 counties in North Texas, is one of the most heavily drilled shale formations in the United States with over 16,000 natural gas wells drilled in the past decade and more planned.  While this region has a long history of oil and gas activities, the recent boom in natural gas extraction has resulted in increasing concern among citizens about how practices such as horizontal drilling and hydraulic fracturing could affect their private well water quality.  Still, no unbiased scientific study had been put forth to provide any evidence to answer those questions. Our study incorporated a multi-disciplinary approach to measure groundwater quality in the Barnett Shale using both analytical chemistry and geospatial analysis. The study is titled “An evaluation of water quality in private drinking water wells near natural gas extraction sites in the Barnett Shale formation” and can be found at the Environmental Science and Technology website: http://pubs.acs.org/doi/abs/10.1021/es4011724.

To carry out this study, our team sampled private water wells of varying depths within a 13-county area in North Texas. Of our 100 samples, 91 were drawn from “active extraction areas,” sites that had one or more natural gas wells within a five kilometer radius. Another nine samples were taken from sites either inside the Barnett Shale and more than 14 kilometers from a natural gas well, or from reference sites outside the Barnett Shale. We referred to these sites as “non-active/reference areas”.

Private water wells in the United States are unregulated, so it is necessary to receive permission from well owners to sample their water. We collected the majority of our samples from citizens who answered a University press release asking for participants. The study participants included people both in favor of and against drilling for natural gas in the Barnett Shale. Collectively, this study provides a “snapshot” analysis of groundwater quality during the summer and fall of 2011, a period when natural gas extraction activities were already well established in the Barnett Shale. We also compared our data to a historical database of groundwater quality to provide some context for groundwater in this region prior to the expansion of natural gas extraction activities.

Our analytical work focused on determination of harmful compounds thought to be associated with natural gas extraction such as methanol, ethanol, heavy metals (arsenic, strontium, selenium, barium, etc.), and BTEX compounds (benzene, toluene, ethylbenzene, and xylenes).  Using inductively coupled plasma mass spectrometry (ICP-MS), we found arsenic in 99 of the 100 wells sampled. Notably, 29 of the 91 samples within active extraction areas had arsenic concentrations above the Maximum Contaminant Limit of 10 parts per billion. The maximum concentration of arsenic we detected within active extraction areas was 161 parts per billion, a value nearly 18 times greater than both the maximum concentration among the non-active/reference area samples and historical levels. We found selenium and strontium at elevated concentrations, with selenium detected exclusively within 2 kilometers of natural gas wells. Several water wells also contained quantifiable levels of methanol and/or ethanol, chemicals known to be included in hydraulic fracturing fluids.  These alcohols can be formed naturally, but have a very short lifespan in the environment before they disappear; therefore, the levels we found were unusual.  We found the highest concentrations in active extraction areas, although we did detect alcohols in a few of the non-active/reference areas as well.  We found no evidence of BTEX chemicals and barium levels were all under the contaminant limit.

Our data suggest there may be a correlation between natural gas extraction and elevated levels of heavy metals and alcohols in private well water since elevated compounds occur on average less than 2 kilometers away from natural gas wells and these compounds were historically at low levels. We do not have any evidence of direct fracking fluid contamination; however, there are a number of indirect pathways through which the heavy metals and alcohols could be introduced. For example, industrial accidents such as faulty gas well casings or improper wastewater disposal could introduce dangerous compounds from produced or flowback waters into shallow groundwater. Additionally, large withdrawals of groundwater for use in hydraulic fracturing operations could result in a localized decline in the water table. Such decreases can be associated with higher arsenic content in waters drawn from shallower wells.

Another scenario to explain elevated heavy metals could be the mechanical vibrations produced from natural gas drilling activity.  In this scenario, a poorly maintained private water well that has accumulated rust, sulfate, and/or carbonate scale is mechanically disturbed by vibrations from nearby intense drilling activity.  Once the rust and scale in the water well are disturbed, arsenic, selenium, and strontium that were previously bound in oxide complexes could be mechanically liberated and released into the well water.  While our study does not conclusively identify the causes of elevated constituents, it does provide an impetus for continued research on this important topic. As water availability decreases around the globe, it will become increasingly important to understand how various anthropogenic activities impact our diminishing resources.

In addition to performing follow up analyses in the Barnett Shale, we are also conducting a separate study in the Cline Shale in the Permian Basin region of Texas. Using an advanced suite of physical and chemical analyses, we are performing a time series analysis of groundwater quality before, during, and after resource extraction activities. A series of baseline measurements have been collected prior to drilling activity near a number of private wells.  These data will then be compared to measurements taken during and after drilling activities. Our study will provide direct evidence as to whether unconventional drilling techniques affect groundwater quality in the Cline Shale.