Chesapeake: 11 Percent of Water Wells Contain Methane Before Drilling

Gas producer shares data in partial response to spring study that found methane in water near shale gas wells.

The State Journal
7 August 2011
By Pam Kasey

Chesapeake Energy–funded laboratory tests find dissolved methane in about 11 percent of northern panhandle drinking-water wells before drilling for gas in the Marcellus shale ever begins.

Two wells of 1,312 tested in Brooke, Ohio, Marshall and Wetzel counties turned up with potentially dangerous levels of methane.

Chesapeake released its data to the State Journal as follow-up to a May study that showed methane contamination of drinking-water wells in northeast Pennsylvania and nearby New York state. Methane concentrations in the study were higher nearer active shale gas wells, with some concentrations at dangerous levels, and the methane bore a chemical signature that resembles gas from the Marcellus depths.

That study, conducted by researchers at Duke University and published in the Proceedings of the National Academy of Sciences, recommended research into the potential health effects of methane in drinking water.

Although the industry disputed the study, its arguments do not address the study’s central findings.

Methane in W.Va. Water

The state Department of Environmental Protection requires oil and gas operators to test the drinking-water well of any landowner or resident within a 1,000-foot radius of a proposed gas well who requests testing. If there are no requests, the operator has to sample a well or spring within 1,000 feet or, if none, within 2,000 feet of the proposed well.

Chesapeake offers free water quality testing to anyone within a 2,500-foot radius, according to spokeswoman Maribeth Anderson.

Hired consultants collect the water samples and send them to third-party laboratories for analysis, Anderson said.

For residents whose results come back with dissolved methane above 0.026 milligrams per liter, or mg/L, Chesapeake mails literature from the Environmental Protection Agency and other third parties and directs them to sources of more information.

For those whose results are above 20 mg/L, a Chesapeake employee or a consultant hand delivers the results and the company offers to vent the well free of charge. The federal Department of the Interior recommends hazard mitigation above 10 mg/L and has determined that dissolved methane is a hazard at 28 mg/L.

Of the 1,312 West Virginia water wells Chesapeake sampled, it detected methane in 11.1 percent: 11 percent in Brooke, 17 percent in Marshall, 8 percent in Ohio and 18 percent in Wetzel county. In Bradford County in northeast Pennsylvania, about 25 percent of wells have methane, Anderson said.

The two West Virginia drinking-water wells that topped the 20 mg/L threshold were in Ohio and Marshall counties. One, Anderson said, was determined by chemical analysis to be from shallow coalbed methane; she did not explain the other.

Duke Data Revisited

The 60 drinking-water wells the Duke researchers tested across northeast Pennsylvania and nearby in New York state were not pre-drilling, but categorized rather by distance from active shale gas wells.

The researchers found dissolved methane in 51 of the 60 wells sampled: 85 percent.

Methane concentrations were 17 times higher on average in drinking-water wells that had active shale wells within a 3,000-foot radius than in those without; 12 came in above the 20 mg/L threshold Chesapeake uses for warning residents and nine had concentrations of methane above the federal 28 mg/L hazard level.

Importantly, the dissolved methane was chemically consistent with gas found in deep reservoirs such as the Marcellus and Utica shales.

Industry representatives countered the study at the time.

They argued, in part, that the study did not include samples taken from the water wells before gas well drilling took place.

The Duke researchers acknowledged that criticism and said they would be releasing a before and after study later.

Many drinking-water aquifers hold dissolved methane before Marcellus activity begins, the industry argued.

That point is illustrated by Chesapeake’s northern panhandle data — but it doesn’t address the Duke researchers’ hypotheses about the source and pathway of the gas they found.

Still an Open Question

The Duke researchers hypothesized that the likeliest pathway of the shale gas methane in drinking-water wells near active shale gas wells was leaking well casings, although they felt it was possible that hydraulic fracturing aggravated existing seismic fractures and allowed methane to travel up through rock.

Chesapeake agreed through Anderson that a well that is not properly sealed can be a conduit for the migration of methane, but said that proper well construction practices, maintenance and regulatory oversight in modern gas production have limited the potential for this to occur.

The company contributed other ideas.

Methane might migrate naturally, for example.

“This natural migration of methane has been documented by academic institutions and government agencies over many years in areas with and without drilling activity,” Anderson’s e-mail response read. Industrial activities besides gas drilling could cause migration, including mining, landfills, construction and drilling of water wells, her response continued.

And environmental variables could affect gas migration, too, it said: changes in atmospheric air pressure and in aquifer levels due to natural cycles of flood and drought.

“A tremendous amount of engineering, planning and design work goes into the development of any natural gas well, the design and construction of which is formulated to be specific to the subsurface geology of the area and conditions encountered and data collected and analyzed as a well is drilled,” Anderson’s e-mail read.

“As the most active driller of wells in the United States with over 11,500 wells drilled in the past twenty years, Chesapeake utilizes its vast experience and the best modern technology to design and construct its wells,” it said.