Groundwater surveys around natural gas drilling sites usually begin at a homeowner’s water well or kitchen faucet. The tests provide a snapshot of water quality at a specific site and time.
Now researchers with the U.S. Geological Survey, Pennsylvania State University and the University of Utah are finding that streams running through areas of gas development can be a portal for detecting the migration of gases through aquifers on a broader scale.
In a paper published recently by the journal Environmental Science and Technology, the researchers studied methane levels in northeastern Pennsylvania streams to look for both natural conditions and possible signs of leaks from natural gas wells.
At one stream in Lycoming County they discovered evidence pointing to just such a leak.
High levels of methane bubbling up in Sugar Run had characteristics of gas that is generally found in deep underground layers of rock, possibly from the Marcellus Shale. After the researchers finished their sampling in 2013, they learned that the state Department of Environmental Protection had determined that at least five residential water wells had been affected by stray methane from gas drilling and issued a violation notice for defective casing or cement in a nearby shale gas well.
The leaking well had been drilled near and under Sugar Run just upstream from the sampling sites where the researchers recorded the high methane levels, which led them to suggest that the shale well is a likely source of the bubbling.
“We can’t prove it because we don’t have data before shale gas well development, but everything is really conclusive,” said Susan Brantley, a Penn State geosciences professor and one of the paper’s authors.
DEP spokesman Daniel Spadoni said the gas migration case has been ongoing since January 2012 and gas continues to impact the stream and water supplies. DEP has issued two violation notices and is evaluating further enforcement options, he said. The affected property owners have received water treatment systems and settlement agreements.
“The department is committed to ensuring that the gas well is remediated and the impact to the environment is eliminated,” he said.
The scientists’ monitoring technique could be a valuable complement to water well testing in regions with shale gas development or an efficient way to evaluate background conditions in areas where drilling is likely to begin.
Rivers are natural collectors for groundwater, so if methane is percolating upward it will often be channelized into streams, which are more representative of water quality over a broader area and much easier to access and sample than residential wells, where it can be difficult to secure a homeowner’s permission for tests, Ms. Brantley said.
Kip Solomon, a geology professor at the University of Utah and another of the paper’s authors, likened a watershed to a funnel and a stream to the funnel’s spout where chemicals picked up by the water over distances can be spotted in one place.
An individual water well, on the other hand, is not necessarily representative of water quality in the wider aquifer or even water quality at the well next door. Concentrations of compounds dissolved in the water can change based on the well’s depth, seasons, weather and other variables, like how many loads of laundry the owner did that day.
At the same time, water well sampling is crucial for oil and gas operators and environmental regulators, because that’s the water that people drink. If oil or gas operators affect a water supply, they are required by state law to restore or replace it.
The DEP has determined that 256 water wells have been contaminated or diminished by oil and gas operations since the end of 2007, according to the most recent update to DEP’s running tally. By far, the most common pollutant regulators cited in the contamination cases is methane, which was reported as elevated above safe concentrations or background levels in 119 of the damaged water supplies. A dozen more water wells were characterized by regulators as having been affected by the migration of stray gas.
Methane is not toxic to drink but it can be hazardous when it escapes from water and concentrates in confined spaces, creating a danger of fire or explosion. It is also a potent greenhouse gas when it is emitted to the atmosphere.
Methane is known to be ubiquitous in northeastern Pennsylvania water supplies naturally, and distinguishing a man-made condition from a natural one typically requires intensive investigations. The methane found naturally in water wells in the region is sometimes thermogenic, which means it was formed by heat and pressure deep underground, just like the methane in Marcellus Shale natural gas or pockets of gas found in rock layers closer to the surface.
Syracuse University hydrogeologist Donald Siegel, who recently collaborated with Chesapeake Energy on a study of its methane sampling results from pre-drilling surveys of private water wells in northeastern Pennsylvania, said that methane concentrations in the region are not systematically higher in water wells that are closer to oil and gas wells.
The percentage of shale gas wells in Pennsylvania with construction problems leading to gas migration “is a trivial number given the number of wells that have been drilled,” he said. “If you did a random sampling, the chances that you would hit one of these things might be pretty small.”
Victor Heilweil, a USGS hydrologist and the lead author of the stream monitoring paper, said the stream sampling approach can help reveal scattered problems when they do occur by taking advantage of what streams reveal about a wider watershed.
“We get a very integrated look at what the groundwater quality conditions are like from the stream sampling,” he said, “whereas with the homeowner wells, it’s kind of like a needle in a haystack whether the wells are in the right place.”
Mr. Siegel, who was not involved in the paper but reviewed it, called the stream monitoring approach “very clever.”
“It might be a nice sentinel method of seeing if there’s a problem,” he said, as long as researchers take care to differentiate man-made methane seeps in the region from naturally occurring ones, some of which are vigorous enough to catch fire and maintain a flame.
Penn State and USGS scientists are hoping to do more stream monitoring and work with volunteers in watershed and fishing groups that already do regular stream surveillance to collect samples for methane analysis as well.
“It doesn’t matter whether you’re worried about brine that might get into a stream or methane that might get into a stream, you really have to know what was in the stream before the alleged incident,” Ms. Brantley said.
“This is the kind of tool that can teach us the answer to that question.”
Laura Legere: email@example.com