According to a spate of recent scientific studies from the United States and Australia, the shale gas industry has generated another formidable challenge: methane and radon leakage three times greater than expected.
In some cases the volume of seeping methane, a greenhouse gas that traps heat 25 times more effectively than carbon dioxide, is so high it challenges the notion that shale gas can be a bridge to a cleaner energy future, as promoted by the government of British Columbia and other shale gas jurisdictions.
"If natural gas is to be a 'bridge' to a more sustainable energy future, it is a bridge that must be traversed carefully," warned one 2014 study published in Science. "Diligence will be required to ensure that leakage rates are low enough to achieve sustainability goals."
The new studies quantifying methane leakage, an age-old problem for the oil and gas industry, illustrate the need for extensive air monitoring before and after gas drilling to determine how hydraulic fracturing may impact natural leakage rates.
In addition, methane seepage seems most pronounced in areas where industry has been fracturing seams of coal, a resource much richer in methane than shale rock, or where there are existing underground coal mines.
The studies underscore how little science has been done on the impact of high-pressured hydraulic fracturing combined with horizontal drilling. The brute-force technology creates fractures in deep rock formations by injecting streams of water, chemicals and sand with such force that the activity can cause small earthquakes.
"I think these studies point out the need for more monitoring before, during and after development on the ground and in the air to definitively point out the methane sources," energy and coal specialist David Hughes said.
Many instances of under-reported emissions
One of the first studies to sound the alarm on the under-reporting of methane emissions by industry and regulators was partly supported by the U.S. National Oceanic and Atmospheric Administration last year.
Published in Geophysical Research Letters, it took a snapshot of methane levels in a producing shale gas field in Utah over one day in February 2012.* It found a leakage rate of six to 12 per cent in the well-fractured basin. A previous study found a four per cent leakage rate in a heavily fracked shale gas field in Colorado.
The finding shook both industry and government, because it had been long assumed that leakage rates from natural gas production averaged around 1.5 per cent.
Such a significant loss of methane to the atmosphere "negates any short-term climate benefit of natural gas from this basin for electricity generation compared to coal and oil," concluded the scientists.
The findings also supported work by Cornell methane expert Robert Howarth, who argued in 2011 that the greenhouse gas footprint of shale was significantly larger than conventional gas due to methane emissions from venting, leaks and flow back fluids during the drilling process.
Then came a 2013 Harvard-led study in which real time monitoring again confirmed the significant under-reporting of methane emissions by the U.S. Environmental Protection Agency.
Using atmospheric measurements from airplanes and communication towers, the study found methane in the atmosphere above the heavily fracked oil and gas fields of the south-central U.S. at rates 2.7 times higher than expected.
Concentrations of propane, another indicator of pollution from oil and gas activity, were also high in the air above Texas and Oklahoma where fracking or the injection of fracking waste has triggered swarms of earthquakes.
The Harvard study was followed by another comprehensive study published in Science and funded in part by a foundation set up by George Mitchell, one of the pioneers of modern hydraulic fracturing.
It also found a big difference between measured methane emission rates versus "official" estimates. Stanford University researcher Adam Brandt and 12 other scientists reported that a small number of high-emitting sources such as leaky pipelines, faulty wellbores, polluting gas plants, and venting storage tanks might account for the high rates of methane.
The scientists didn't think that hydraulic fracturing was the dominant source of methane seepage, but noted that rates of leakage varied greatly from basin to basin.
"Improved science would aid in generating cost-effective policy responses," they concluded.
A 2013 study by the University of Texas and U.S. Environmental Defense Fund contradicted these findings by finding methane emissions lower than estimated by the U.S. EPA in shale fields. But industry picked the sites that were studied and the times that data was collected.
Coal plays a role
Last month, another U.S. study reported that an airplane survey over a two-day period found large plumes of methane above shale gas well pads over southwestern Pennsylvania two to three times order of magnitude greater than expected during drilling operations.
The scientists suggested that just a few shale gas wells may be super methane leakers and account for the large spikes of methane in the atmosphere.
(The concept of extremely leaky wellheads in particular oil and gas fields isn't new: one 1996 study published by the Society of Petroleum Engineers found that 9,000 out of 20,000 wells in the Lloydminster heavy oil region of western Canada were leaking methane so prolifically through the wellbore, soils or aquifers that the problem would cost hundreds of millions of dollars to fix.)
The researchers also hypothesized that drilling activity near active underground coal mines may inadvertently trigger the escape of methane from the coal by opening new pathways.
Unlike other hydrocarbon-bearing rocks, the softness of coal creates complex fracture networks that release large amounts of methane. In the past, unvented methane has caused horrific mine explosions.
Andrew Revkin, the respected New York Times environment blogger, noted that the results strongly suggest "that regulators should require monitoring of local air chemistry before, during and after drilling of gas wells."
Revkin also asked Louis Derry, a Cornell University geologist, to comment on the study's implications in regards to coal.
"It was possible that a small number of wells contribute heavily to methane leakage in a producing field -- a largely fixable problem," Derry said.
But "the real message of this study may be that gas fluxes from coal operations have been underestimated, and that they are mostly responsible for the hotspots," he said.
He added that drilling in a coal-rich area may "require special precautions to prevent transient leaks" in addition to pre and post-air monitoring.
Australia's approach to leakage
Australian studies have also found significant problems with methane and radon leakage from fields producing coal-bed methane in southern Queensland and northern New South Wales.
In 2012, Isaac Santos and Damien Maher at Southern Cross University measured concentrations of methane in the atmosphere above fractured coal-bed methane fields with a spectrometer mounted on a vehicle.
Like subsequent U.S. studies, they found methane emissions in producing fields were three-and-a-half times more than expected.
The scientists offered two explanations for the findings: leaky industry infrastructure or seepage through soils.
Any geological area that has gas deposits will have natural seepage, but no baseline studies were done by government, eager for hydrocarbon revenue, prior to the launch of the billion-dollar industry.
Australian scientists are now trying to figure out whether drilling activities such as fracking are activating large methane escapes by releasing methane and other gases such as radon into groundwater, creeks, soils and the atmosphere.
Coal is a major source of methane, so it's not surprising that levels are higher over coal fields, noted David Hughes.
"But saying these levels are related to later [coal-bed methane] or shale development or initial mining or natural outcrops of coal again needs the data before, during and after," added the geologist and consultant.
"I think the recognition of this need is dawning on governments, but I'm unsure of the level to which it has actually been implemented."
'There is no will to fix anything': scientist
A 2013 follow-up study measured radon gas concentrations at monitoring stations both inside and outside the Kenya/Talinga gas fields north of Tara in southern Queensland. Radon is a naturally occurring gas in the soil.
The scientists found radon levels in the atmosphere three times higher than average in areas with a high density of coal seam wells.
"It has been known for years that radon anomalies can be observed during earthquakes," explained Santos in a press release. "As the soil structure expands or contracts and cracks before and during an earthquake, it creates conduits for the release of soil radon into groundwater and the atmosphere."
"We hypothesize that an analogous process is happening when the soil structure is altered during coal seam gas mining through processes such as drilling, hydraulic fracturing and alteration of the water table."
Damien Maher, one of the study's co-authors, said the findings suggested the radon leaks are not only coming from wellbores but through new man-made pathways that industry had not accounted for.
"Fixing the infrastructure is relatively easy. Fixing up the changes in the soil structure is much more difficult," said Maher.
Karlis Muehlenbachs, a University of Alberta expert on tracing stray gases from oil and gas fields, doubts that oil and gas regulators have the gumption or the resources to do baseline atmospheric monitoring, let alone fix the continent's leaky natural gas production system.
"There is no will to fix anything," Muehlenbachs said.
For years Muehlenbachs has proposed that governments mandate baseline monitoring, through isotopic fingerprinting, of methane, ethane and propane from producing wells, abandoned wells, natural seeps and water wells in order to protect groundwater prior to drilling and fracking.
No regulator has yet implemented his protocol.
*Story corrected May 9. Month and year of study was incorrectly identified as February 2013.