The need for research to help understand the causes and mitigate induced seismicity was one of several topics at several recent Washington, D.C., workshops that explored what we know about the subsurface and its response to energy extraction technologies – and what research is needed for more efficient and safe energy development.
The workshops will help the U.S. Department of Energy (DOE) organize a new program that integrates applied and basic subsurface energy research in the areas of nuclear, geothermal, carbon dioxide storage, and oil and natural gas production and waste disposal.
The program is SubTER, for Subsurface Technology and Engineering Research and Development.
Industry involvement in the workshops allowed planners to learn about industry technology needs and interests, which is important to assuring the usefulness of research results. Industry involvement also is important because research projects will depend on industry’s willingness to provide subsurface and operational data.
Several years ago, induced seismicity was of sufficient public concern that the National Research Council conducted a detailed analysis. Its 2012 report, Induced Seismicity Potential in Energy Technologies, found that felt seismicity was associated with a small percentage of geothermal, enhanced oil recovery and wastewater injection wells, but was documented for only one hydraulic fracturing operation.
The study concluded, “The process of hydraulic fracturing a well as presently implemented for shale gas recovery does not pose a high risk for inducing felt seismic events.”
Concern about induced seismicity has grown since 2012, as hydraulic fracturing, wastewater disposal and associated seismic events have mushroomed.
The vast majority of felt seismicity comes from wastewater disposal wells – and felt seismicity is now tied to hydraulic fracturing treatments in British Columbia, Ohio and Oklahoma.
However, it still represents an extremely small percentage of wells.
Selected research findings included:
♦ William Leith, U.S. Geological Survey senior science adviser, reported that the number of earthquakes in the United States greater than magnitude 3 has increased from about 21 per year before 2000 to more than 300 per year from 2010 through 2012.
The increased seismicity correlates with an increase in injection of wastewater in deep disposal wells. Leith noted that hydraulic fracturing does not seem to be related to the increase in earthquakes greater than magnitude 3.
He also noted that earthquakes may be triggered far from injection sources.
“It’s pressure,” he said, “not fluid flow.”
♦ Roberto Suarez-Rivera, director of the Schlumberger Innovation Center, tackled the vexing industry question of why some sections within a horizontal well are more productive than other sections.
He provided examples of how hydraulic fracturing is a “competition between in situ stresses and rock fabric,” and he described lab studies that showed that fractures do not go exactly where planned because of reservoir heterogeneities. Much of this rock variation may be smaller than the scale of subsurface measuring devices.
This and other presentations suggest that heterogeneity at all scales will be a major complication to improving energy recovery technologies and preventing induced seismicity.
♦ K.J. “Kris” Nygaard, senior stimulation consultant at ExxonMobil Upstream Research Company, listed research that would aid industry in characterizing and mitigating the induced-seismic risk:
- Improve our knowledge of subsurface stress and significant fault systems.
- Improve understanding of ground-shaking behavior and seismic wave attenuation characteristics.
- Develop capabilities to differentiate naturally occurring from induced earthquakes.
♦ Thomas H.J. Goebel, an applied seismology consultant at the University of Southern California’s Induced Seismicity Consortium, reported his group has developed a model to differentiate tectonic versus induced seismicity.
They hope to go on to develop a model that works for both plate boundary regions like California and intra-plate regions in Oklahoma.
♦ Austin Holland, state seismologist, Oklahoma Geological Survey, explained what is known and unknown about Oklahoma’s exponential rise in earthquakes.
The vast majority of quakes occurs within or near the top of the crystalline basement and are related to wastewater disposal wells. However, Holland estimated that about 2 percent of hydraulically fractured wells might have earthquakes associated with the completion process.
A preliminary study of earthquakes and hydraulic fracturing treatments at a well in Carter County, Oklahoma, in July 2014, found earthquakes temporally correlated to injection. Most of the 26 earthquakes near the 3.5-kilometer deep well were between 3.5 and 8.5 kilometers depth and occurred within 100 minutes of the start of each of the four hydraulic fracturing injections. The varied locations of the earthquakes, some as far as seven kilometers from the well, require additional study.
Several speakers noted the need for more data. The existing seismic network is inadequate to identify the precise location of induced earthquakes. There also is inadequate data about the injection rates, volumes and pressures of wastewater disposal wells.
Joshua White, of Lawrence Livermore National Laboratory, recommended field laboratories to improve understanding of the subsurface by integrating subsurface characterization, monitoring and modeling.
Unfortunately, field laboratories are expensive to operate.
There may be alternative research data, however, in a few existing subsurface injection projects. For example, the Bureau of Reclamation has operated a well-monitored saline injection well in the Paradox Valley for 18 years, yielding an exceptional data set of injection and induced seismic events.
A few speakers mentioned the Horn River Basin seismicity. British Columbia Oil and Gas Commission and the Geological Survey of Canada publications (2012) define the low-level seismic events in the Horn River basin between 2009 and 2011. These events resulted from fluid injection during hydraulic fracturing in proximity to pre-existing faults. Seismicity was shown to correlate with injection volumes of hydraulic fracturing treatments.
Seismic events ranged in magnitude from 2.2 to 3.6, with only one event felt at the surface. Seismicity increased as stimulation volumes grew and declined with the decline in drilling operations.
The overarching, unanswered question posed by many was: Why is there seismic activity in certain regions, notably Oklahoma, but not in other areas of active drilling, production and wastewater injection, such as the Bakken producing area?
Oil and gas operators, and industry and academic researchers will have an opportunity to learn more about the DOE subsurface research program – and provide comments to DOE managers – at a Town Hall at AAPG’s 2015 Annual Convention and Exhibition, set May 31-June 1 in Denver.