During the past decade or so the oil and gas industry has injected wastewater into deep rocks in eastern Texas, causing Earth’s surface to bulge ever so slightly—and likely triggering a series of tremors there in 2012, a new study suggests. Scientists say the work offers hope that similar analyses of the landscape in other oil- and gas-producing regions could help identify areas at risk of human-caused earthquakes.
The 2012 quakes shook the small town of Timpson, Texas, which lies northeast of Houston near the Louisiana state line. The largest, a 4.8-magnitude quake, and three more magnitude-4 or higher that followed, all originated in a suspicious spot: directly beneath two wells where wastewater generated during oil and gas production in the region is pumped into porous sandstone layers about 1.8 kilometers underground. Oil and gas producers dispose of their wastewater deep underground for a variety of reasons; sometimes pumping fluid into the reservoir helps boost production, and in other cases it’s a convenient method of getting polluted water out of retention ponds on the surface so that it doesn’t inadvertently spill to pollute rivers, streams, or other sources of drinking water.
According to data provided by the companies that owned the wells, between 2007 and mid-2012 the two injection wells nearest the quakes and another two wells fewer than 10 kilometers away pumped, on average, about 890,000 cubic meters of water into the ground each year. (Or, put another way, that’s about one Olympic swimming pool worth of wastewater pumped underground each day.)
Many studies have already noted the link between wastewater injection wells and swarms of nearby tremors, says Manoochehr Shirzaei, a geophysicist at Arizona State University, Tempe. Few doubt that injection wells are the chief reason that Oklahoma has overtaken California as the earthquake capital of the United States’s lower 48. But in the new analysis, he and his colleagues looked to build more than just a circumstantial case linking the four underground disposal wells to Timpson’s temblors.
First, the team used a technique that compared a series of radar images of the area taken from space. Between May 2007 and November 2010, the terrain between the two sets of injection wells rose as much as 3 millimeters per year, on average, the researchers report online today in Science. Although hordes of studies have noted subsidence, or sinking, of the landscape when oil, gas, or water for irrigation are withdrawn from underground reservoirs, this new analysis is one of the first to note uplift as a result of pumping fluid into the ground, Shirzaei says.
The team then went further. Using computer simulations, and the bulging ground as a constraint, the researchers found that over time the wastewater seeped away from the injection point and boosted water pressure within the tiny spaces in the surrounding rocks—a parameter scientists call pore pressure. Eventually, the expanding front of increased pore pressure reached fault zones and triggered quakes between depths of 3.5 and 4.5 kilometers, Shirzaei says. The team’s models suggest the pore pressure in the rocks along the fault zones increased to a level that has been large enough to trigger aftershocks to major quakes elsewhere in the world, he notes.
The group’s combination of monitoring uplift of the landscape over time and modeling the effects of wastewater injection on pore pressure “is a powerful approach,” says Shemin Ge, a hydrogeologist at the University of Colorado, Boulder. “It will help advance our understanding of what’s actually going on in rocks near wastewater injection sites.”
The new findings will also guide where future wells should be located and how much wastewater—and possibly more importantly, how slowly such wastewater—should be pumped into areas near known fault zones, Ge says. Geological context is also important, she says. As Shirzaei and his team noted in one part of their study, some areas seemed immune from tremors, possibly because layers of impermeable rocks lie between the wastewater injection point and susceptible fault zones. In such a configuration, the wastewater can’t migrate to the fault and cause a quake. “Without water, nothing’s going to happen,” she says.
Drilling and maintaining wells to directly monitor water pressure in deep rocks is both challenging and costly, says Roland Bürgmann, a geophysicist at the University of California, Berkeley. Thus, he suggests, injection-induced uplift, detectable from space, could serve as a less-expensive warning sign of increased seismic risk. “We can’t predict such induced earthquakes,” he notes, “but maybe we can better understand the conditions that lead to them.”