Seismologist: Fracking Injection Wells Linked to Earthquakes
Dr. Elizabeth Cochran discusses research showing that injection wells from fracking and gas and oil production is likely causing an uptick in U.S. earthquakes
JESSICA DESVARIEUX, TRNN PRODUCER: Welcome to The Real News Network. I’m Jessica Desvarieux in Baltimore.
Many of us are familiar with the term hydraulic fracturing, or commonly known as fracking. But less known is how scientists have linked it to earthquakes. It’s become so prevalent in places like Oklahoma that the seismic activity has a name of its own–frackquake.
But what’s the correlation between earthquakes and tracking? Seismologist researchers from the University of Oklahoma say that it is the disposal of waste water in the form of fluid injections and underground disposal wells, which are common practices in fracking and other unconventional oil and gas production.
Now joining us to give us more details is one of those researchers, Dr. Elizabeth Cochran. She’s a seismologist with the U.S. Geological Survey in Pasadena, California.
Thanks for joining us, Dr. Cochran.
DR. ELIZABETH COCHRAN, SEISMOLOGIST, US GEOLOGICAL SURVEY IN PASADENA, CA: My pleasure.
DESVARIEUX: So, can you tell us about the study you worked on with the University of Oklahoma and its link to wastewater injection and the 2011 earthquake sequence? What was so (pardon the pun) groundbreaking about what you found?
COCHRAN: Yeah. So we studied this sequence of earthquakes that occurred in Oklahoma. It actually included two magnitude 5 events and a magnitude 5.7 event–that’s the largest one. And what was really interesting about this sequence was that what we were able to show was that these earthquakes occurred in very close proximity to several injection wells. In this case, in Oklahoma the injection wells were disposing of what’s known as produced water. So this is water that comes out along with the oil during production. So it’s water that’s naturally occurring in the formation. And it can actually be as much as eight times the volume of the actual oil produced. And so, as the oil comes up, they separate off the oil and this water that comes up at the same time, and they have to dispose of that in some way. And the typical way of doing that is by reinjecting it down what’s known as a wastewater disposal well. So these are wells that are drilled down into the ground to about 1 to 2 kilometers depth, about a mile deep, and they pump in this produced water.
And so what we looked at for the Oklahoma sequence was where the earthquakes occurred relative to these disposal wells. And we also looked at whether there was any link between changes in seismicity and changes in production or change in the volume of fluids being disposed of. And so what we found was that the earthquakes were occurring within very close proximity to the depth of these wells, within about 500 meters. And that’s basically to the resolution we can accurately estimate the location of those earthquakes.
And also what was happening with the actual injection parameters was that they had been injecting fluid into these wells since about the 1990s, and so it was somewhat of a surprise to see that, you know, 20 years later we see a series of earthquakes occur. But if you look in detail at the injection pressures, what happened was initially they could inject wastewater without any pressure. It would just basically go straight down the well, and they didn’t have to put any pressure to make it go into the formation. But those pressures gradually rose over the 20 year period until essentially they’d have to keep increasing the pressure at which they’d force the water down in order to continue injecting the same volume of water. And so we think that was showing that essentially this formation, which had been previously drilled and produced and now is being reinjected into, was essentially filling up, that it was a closed space where they were pumping a lot of water down into, and essentially it got to the point where the formation was full, and that caused increases in core pressure, which may have led to these events along the existing fault systems there.
DESVARIEUX: Alright. But we should mention not all scientists are in agreement with your findings. The Oklahoma Geological Survey concluded in response to your survey that, quote, “The interpretation that best fits current data is that the Prague [Oklahoma] Earthquake Sequence was the result of natural causes.” Natural causes. What would be your response to that?
COCHRAN: Yeah, that was just a statement they issued on their website. And this was not in a peer-reviewed article. And so we really don’t have much of a response to that, because we’re not sure what they’re basing that observation on.
Since that time, there have been other swarms of events in Oklahoma that the Oklahoma Geological Survey has linked or has suggested maybe a link to injection. And so it’s not–you know, the evidence behind that statement is not particularly clear to us.
DESVARIEUX: Dr. Cochran, can you give us a specific example of where there’s been a more definitive link?
COCHRAN: Well, as I was just stating, we have done this study of the Prague, Oklahoma, sequence and feel like there’s a good link between the events that occurred there with injection.
There are other studies also that link wastewater injection to earthquakes. A recent study from Ohio actually showed that they were injecting wastewater from hydrofracking process. And this was water that was produced during hydrofracking of the Marcellus Shale in Pennsylvania and was trucked into a disposal well in Ohio. And it was closely monitored. And this is a study by Won-Young Kim at Columbia. And he showed that there was direct correlation between the seismicity and the injection parameters. So when they saw slight lulls in production, they saw some quiescence of the seismicity, so they saw fewer events. And then, when production would come back up, they–or, sorry, when injection would come back up, they saw the seismicity increase as well. That’s a case in contrast to Oklahoma that was looking at water directly related to hydraulic fracturing process, whereas in Oklahoma we’re looking at sort of naturally occurring fluids that were just pumped out as part of the oil production process.
DESVARIEUX: Alright, Dr. Cochran, I do want to go back to the Oklahoma Geological Survey and some of their conclusions that they came up with after reviewing your study. They also had this to say: quote, “The Prague Earthquake Sequence, as well as other current and historically active seismic areas in Oklahoma, would benefit from further study, including improved earthquake monitoring and acquisition of formation pressure data.”
If I understand correctly, Dr. Cochran, the gas and oil industry wastewater disposal is regulated by state governments, and therefore regulations can vary by state. In your experience, how transparent and forthcoming have oil and gas companies been with providing data?
COCHRAN: Yeah. I think that, as you mentioned, varies quite significantly, depending on the oil producer or the oil company involved, as well as what the state regulations are. What we found in Oklahoma was that the Oklahoma Corporation Commission is responsible for collecting monthly injection pressures from the various oil companies. So that data is available.
However, what we would like to see as seismologists is actually much finer data. So we would like more production or injection data on, say, a daily or even hourly rate so we can better link production to any seismicity that is occurring in the area. So we, for example, for this study that we had of this sequence near Prague, Oklahoma, we went to the specific oil company involved in that and asked for additional information, but was not able to receive that information. And so it would be very helpful if we did have greater access to that sort of data.
DESVARIEUX: Alright. Dr. Cochran, thank you very much for joining us.
COCHRAN: Thank you.
DESVARIEUX: And thank you for joining us on The Real News Network.
DISCLAIMER: Please note that transcripts for The Real News Network are typed from a recording of the program. TRNN cannot guarantee their complete accuracy.