Do CO2 Injections Pose Risk of Harmful Earthquakes?
How common are measurable earthquakes in association with oilfield operations? The answer is: exceedingly rare. Nevertheless, another scientific paper has raised the possibility of seismic events occurring as a result of injection of CO2 to stimulate new oil production from depleted oil fields. Since this process, known as enhanced oil recovery (EOR), is a vital component of making carbon capture and storage (CCS) economically viable as a means of addressing global climate change, we must take a close look at the facts. So here’s what we know:
On November 4, the Proceedings of the National Academy of Sciences (PNAS) released a paper on seismicity that may have been induced by injections of gases in a West Texas oilfield. The oilfield studied, near Snyder Texas, has been subject to injection-related production stimulation since 1957. In the present study, authors report minor seismicity recorded between 2006 and 2011 with 18 earthquakes. Of the 18 recorded events, 17 were Richter magnitude 3 (associated with barely or unnoticeable ground shaking) and one was a magnitude 4.3 (ground shaking capable of rattling dishes but not significant harm). To put this in perspective, according to the U.S. Geological Survey (USGS), worldwide there are an estimated 1.3 million earthquakes between magnitude of 2.0 and 2.9, 130,000 earthquakes between 3.0 and 3.9 and 13,000 earthquakes between magnitudes 4.0 and 5.0 annually. None of the seismicity halted injection; instead the operators paid extra attention to optimizing the injection rates.
The study further points out that in the adjacent and well-known SACROC field– in the same town of Snyder, Texas that has been undergoing CO2 flooding for 40 years– that there has been no induced seismicity. In fact, CO2 enhanced oil recovery (EOR) was born in these fields, having been in operation since 1971. Since then, over four decades of experience of CO2management with approximately one billion metric tons of CO2 injected over that period in tens of thousands of wells has produced one and a half billion barrels of oil. But, only three known earthquakes greater than 4.0 magnitude have been recorded during oilfield water flooding, and none known to be associated with CO2 flooding, according to the complete reviewof seismic events associated with energy technology in the United States published by The National Academy of Sciences (NAS, 2012).
It is well known that tiny earthquakes – those that impart an energy release at a depth of a kilometer similar to dropping a gallon of milk on the floor – can be associated with tiny cracks that may form to accommodate fluids injected into the pores of rocks. Such seismicity (known as microseismicity) is only measurable with extremely sensitive instruments, and do not represent precursors to major events nor do they signal movement on known or unknown faults. In fact, in EOR, operators take pains to ensure that rocks are not over pressured and inadvertently fracked because fractures allow CO2 to circumvent the oil-filled pores rather than to sweep the oil out. In fact, fracking is avoided in EOR and carbon storage because it will severely reduce the effectiveness of the spread of CO2 through the formation pores. Instead, EOR takes place in a pressure-depleted reservoir and rebuilds pressure towards minimum miscibility-the point at which CO2 mixes with the oil to most effectively move it out of the rock. This process takes place well below the rock fracture point. In a carbon storage regime, operators will focus on “concurrent storage”, that is, normal operations with added monitoring and accounting–related surveillance. If operators desire to undertake storage alone, then, under current rules, they must operate EPA’s Underground Injection Control Rule that requires remaining well below the frac pressure at 90% of the rock strength.
CO2 injection operations are commonplace in the US. Today 4,000 miles of CO2 pipelines connect to 127 projects producing over 100 million barrels of oil annually utilizing 57 million metric tons of CO2. Furthermore, there are over 100,000 wells undergoing water flooding today and another 13,000 wells undergoing CO2 flooding. After decades of operations, wastewater disposal has also been associated with only eight events that have been actually felt by nearby residents, none of which have been associated with significant damage. Moreover, over 4 billion tons of fluids are injected into the subsurface in over 30,000 wells every year in the United States and minor induced seismicity is limited to a few fields. While the experience with CO2injection for carbon storage projects is small, according to the 2012 NAS study, there are no known historically felt events and none with a magnitude of 2.0 or greater. Why is this? Storage of CO2 in oilfields is accompanied typically by production of water, hydrocarbons and CO2 resulting in a balancing of subsurface pressure. In fact “stacked storage” in oil and gas field using associated brine formations, may prove advantageous in a number of ways including the opportunity for pressure management by fluid production.
Induced seismicity associated with oil and gas operations continues to be an issue of interest to policymakers, though, following a 2012 paper by Stanford researchers Mark Zoback and Steve Gorelick relative to future ability of deep subsurface geologic formations to accept and contain large volumes of injected CO2 captured from power plants. However, MIT researchers Ruben Juanez, Brad Hagar and Howard Herzog penned a PNAS rebuttal to that study pointing out that earthquakes largely occur in crystalline “basement” rocks that lie beneath the many thousands of feet of sedimentary reservoir rocks where oil and gas deposits occur, or where CO2 might be stored. Injections into those sedimentary rocks are very unlikely to trigger an earthquake in the underlying crystalline basement rocks. CATF has also addressed that study on our own website.
How do we avoid causing earthquakes? Despite the vanishingly small risk of damaging earthquakes with CO2 injections, careful site selection, risk analysis, constant surveillance and injection management must be essential components of healthy geologic carbon storage projects, particularly in seismically active areas. Carbon storage sites should be carefully screened, and those posing high seismic (or other) risk should be avoided or management systems employed. Monitoring of CO2injections should include pressure management and tracking of subsurface CO2 plumes relative to geologic structures.
So, the recent PNAS paper provides further understanding of into seismicity associated with subsurface injection of CO2, but it is important to note that in the paper, the authors correctly put their results in perspective, stating: “The fact that no other gas injection sites have reported earthquakes with magnitudes as large as 3, suggests that despite Zoback and Gorelick’s (2012) concerns it is possible that in many locations large volume CO2 injection may not induce earthquakes.”