Characterization of Accidental Spills and Releases Affecting Groundwater in the Greater Wattenberg Area of the Denver-Julesburg Basin in Northeastern Colorado

Abstract

Public concern about the potential effects of unconventional oil and gas development on water quality has grown in recent years. In this study, we considered accidental spills and releases that occurred from 2007 to 2014 in the Greater Wattenberg Area (GWA), an area of intensive oil and gas extraction located within the Denver-Julesburg Basin in northeastern Colorado. Our objective was to quantify the occurrence rates of (1) all spills and (2) spills determined by the operator to have caused groundwater impacts. Additionally, for spills affecting groundwater, we analyzed characteristics including facilities and equipment involved, causes, and resolution times in order to identify recommendations for improved regulatory or operational practices. Spills and releases were identified from publicly-available reports collected by the Colorado Oil and Gas Conservation Commission (COGCC). Spills were identified as having potentially impacted groundwater based on an operator-reported field present on the reports. Groundwater impacts were verified for this subset of spills by reviewing the narrative description of the incident and the supporting documents (e.g., consultant reports, laboratory data) available in the COGCC database. Additional information collected included the causes of the spills, the facility types, and the time required for resolution of spills. To determine spill occurrence rates, the number of annual spills was compared to several annual indicators of oil and gas development, including the volumes of oil and water produced and the number of active producing wells in the GWA. The annual spill count remained relatively stable from 2007 to 2011, but increased each year from 2011 to 2014. From 2007 to 2014, spill occurrence rates decreased or remained steady when compared to oil and water production volumes, respectively; however, the occurrence rate increased compared to the number of active producing wells. These trends suggest that fewer spills occurred for a greater volume of fluids produced and handled in the GWA, but that more spills occurred per active well. For all three normalization metrics, we observed that the occurrence rate of groundwater-impacting spills remained steady or decreased. The percentage of spills impacting groundwater compared to total spills decreased by a factor of two from 2010 (54%) to 2014 (27%). Among groundwater-impacting spills, the most common facility types were tank batteries and lines (including flow lines, gathering lines, load lines, and pipelines), and the most common cause was equipment failure. Based on our observations of spills in the GWA from 2007 to 2014, we have suggested several strategies for regulators and operators to improve spill reporting practices and to reduce the likelihood of surface spills that result in persistent impacts to shallow groundwater.

Towards Quantifying the Likelihood of Water Resource Impacts from Unconventional Gas Development

“…While not specifically stating the rate of surface water contamination or the exact volumes involved (hence rendering the quantitative assessment of "consequence" impossible), several other studies have found similar spill rates for formations throughout the United States. In the Niobrara Shale formation in northeastern Colorado, for example, the spill rate remained relatively constant from 2007 to 2011 at 1 spill recorded for every 100 wells, while from 2012 to 2014 the rate of spills doubled to 1 in 50 (Armstrong et al 2017). Synthesizing spill data from shale and tight gas wells in four U.S. states from 2005 to 2014, Maloney et al (2017) showed that spill rates were increasing in Colorado, New Mexico, and North Dakota over time, while Pennsylvania spill rates peak and decrease over time as discussed above.…”

“…Within the area, there were 18,000 active wells, giving a likelihood of groundwater impact of 1 in 250 per well per year. In another analysis of Colorado shale well spill rates, Armstrong et al (2017) found that the rate of groundwater contamination from surface spills increased from 2007 to 2014 (although this increase was slower than the increase in production volume); the rate of surface spills was estimated at between 1 in 50 and 1 in 100 with 30 to 50% of those spills affecting groundwater. For 3533 wells drilled in the Marcellus Shale, the Pennsylvania Department of Environmental Protection issued 1144 NOVs, of which 4% were for "major" land spills (over 400 gallons) (Considine et al 2013).…”