When unusual earthquakes in unusual locations occur, there are always questions:
1. Were they induced?
2. did "fracking" cause them?
3. Would they have happened anyway even without human influence?
...and so on.
A region of Kansas southwest of Wichita has experienced, since early in 2013, a series of unusual earthquakes. At least 100-plus earthquakes have been recorded. The largest of this sequence came on November 12, 2014 and at M4.8, it was somewhat damaging to the village of Milton, KS.
Nearly all of the earthquakes are occurring in three south-central Kansas counties: Harper, Butler, and Sumner.
This has sort of become an irregular series of mine. The difference between this and my past work is I had published scientific background to work with--the substantial increase in seismicity in neighboring Oklahoma has garnered significant public scientific concern, particularly after the November 2011 damaging earthquake.
With Kansas, however, we do not yet have that. Although I can absolutely guarantee that people smarter than me are looking into this and hopefully will publish their findings within the next few months and years, please consider much from this point forward on the potential cause of these earthquakes as highly speculative on my part.
GEOLOGY OF THE REGION
Kansas, of course, is smack dab in the middle of the North American Plate. It wasn't always. The Mid Continental Rift, which extends from Lake Superior into Kansas But the earthquake region in question is in south-central Kansas and geologically, this is called the Sedgwick Basin and the Nemaha Ridge.
You'll remember our friend the Nemaha from Oklahoma. It is part of an uplift that is the buried root of a 300 million year old mountain range. Associated with the Nemaha Ridge is the Humboldt Fault system, a series of normal faults that trends across Nebraska and Kansas in a complex fashion (i.e., they're not just a single fault, but a zone of faults). The Humboldt may be active, despite its ancient age.
Most of these faults are known for their ability to trap oil and gas, as we'll see further down.
SEISMICITY
Kansas is largely a state of low seismicity, however there are a few areas where earthquakes naturally occur in the state.
A 2004 study relocated the epicenters of dozens of microquakes and some larger ones recorded during the late 1970s through the 1980s in the state, demonstrating that the epicenters loosely are associated with the Nemaha Ridge and Humboldt Fault (sorry, Paywall). A large earthquake of approximately M5.2 occurred near Manhattan, KS in 1867, so while very rare, sizeable earthquakes are naturally possible in Kansas.
POTENTIAL CAUSES
Much of this section is highly speculative!
Possibly induced earthquakes have occurred in Kansas in the past, in west-Central Kansas near the town of Palco, in 1989. An abstract from a conference presentation reads:
THE 1989 EARTHQUAKE SEQUENCE NEAR PALCO, KANSAS: A POSSIBLE
EXAMPLE OF INDUCED SEISMICITY
The Palco, Kansas earthquake of June 4, 1989 (4.0 mbLg =, NEIS) began a sequence of
many felt events which continued until at least August 31. June 16, 3.8 mbLg = and July 13, 3.4 mbLg = are other notable events. Western Kansas is an area of low seismic
activity, only 3 event of magnitude 4 and larger are known within 200 km of Palco in the last 110 years (EPRI catalog). The activity was monitored locally by 7 portable analog recorders operated from August 10 to 19 with —60 locatable events observed. The activity is found to originate —7km southeast of Palco in sections 1 and 2 of township 9. The better located events define a zone striking east-west dipping steeply to the north. The seismicity extends from the top of crystaline basement, depth —1.3km, downward 1 km and extends 1 3/4 km east-west. The sense of slip, as defined by the pattern of first motions is primarily dip-slip with the north side down and a small component of rightlateral strike-slip. The activity occurs in the Marcotte Oil Field. Oil extraction, which began in the 1950's, changed original overpressured conditions above the basement to much less than hydrostatic. More recently brine disposal wells have been introduced and pore pressure has undergone a relative increase. One disposal well lies directly above the western, most active, part of the zone of seismicity. The coincidence of a disposal well, recent pore pressure history, extended swarmy nature of the seismicity, and low level of prior earthquake occurrence in this area allow for the possibility that this seismicity could have been induced. Comparing the size of this zone of seismicity with others in the Eastern U.S. suggests that it would not generate events of magnitude greater than —5.
You'll notice some terms that should be familiar. "Disposal Wells." "Pore Pressure." We'll get to those later too.
Oil and gas extraction also produces a great deal of water, water that's often otherwise unusable. It's injected back into the ground as part of secondary oil recovery operations or disposed of in the same way. This water is usually very saline, and may have other pollutants from being ensconced deep within the Earth--remember--the Midwest once was part of a great sea.
Wastewater injection
In Colorado this practice of injecting wastewater back into the ground in the Raton Basin along the Colorado-New Mexico border near Trinidad started a sequence of earthquakes that began in 2001 and continues to this day, with a sizable and damaging quake on August 23, 2011.
Data suggests that the large quake of August 23, 2011 was clearly induced by fluid injection.
Colorado's Oil and Gas Conservation Commission regulates wastewater injection wells, including what goes in, and what comes out, and at what rate. The Raton field, where coal bed methane is extracted, rapidly expanded from 1997 to 2001. There are over 20 active injection wells in this part of Colorado. According to Rubenstein et al (the authors of the paper linked above), a sharp increase in the injection of waste water occurred in Colorado. This would have the effect of raising the pore pressure subsurface. It isn't that the fluids are lubricating the faults. Instead the fluid injection and its pressures change the stress conditions (pore pressure) that are acting on any possible subsurface faults. Because all of the continents have an overall stress and strain field due to their motions relative to each other, a fault that is oriented in a favorable fashion may find its clock advanced significantly forward if the local stress and strain field is changed.
Very precise and exact data is available from Colorado, including water production rates and the rates that it's re-injected back into the ground. Data are not as clear from New Mexico but it's a reasonable assumption that conditions below ground are similar. Ironically Colorado is where people realized this practice can cause earthquakes under certain conditions in the first place, in a now infamous sequence that rattled Denver in the early 1960s.
Here is the abstract (full study here, courtesy of Colorado Public Radio)
We investigate the ongoing seismicity in the Raton Basin and find that the deep injection of wastewater from the coal‐bed methane field is responsible for inducing the majority of the seismicity since 2001. Many lines of evidence indicate that this earthquake sequence was induced by wastewater injection. First, there was a marked increase in seismicity shortly after major fluid injection began in the Raton Basin in 1999. From 1972 through July 2001, there was one M≥4 earthquake in the Raton Basin, whereas 12 occurred between August 2001 and 2013. The statistical likelihood that such a rate change would occur if earthquakes behaved randomly in time is 3.0%. Moreover, this rate change is limited to the area of industrial activity. Earthquake rates remain low in the surrounding area. Second, the vast majority of the seismicity is within 5 km of active disposal wells and is shallow, ranging between 2 and 8 km depth. The two most carefully studied earthquake sequences in 2001 and 2011 have earthquakes within 2 km of high‐volume, high‐injection‐rate wells. Third, injection wells in the area are commonly very high volume and high rate. Two wells adjacent to the August 2011 M 5.3 earthquake injected about 4.9 million cubic meters of wastewater before the earthquake, more than seven times the amount injected at the Rocky Mountain Arsenal well that caused damaging earthquakes near Denver, Colorado, in the 1960s. The August 2011 M 5.3 event is the second‐largest earthquake to date for which there is clear evidence that the earthquake sequence was induced by fluid injection.
I have a strong suspicion that this is what is happening in south-Central Kansas. Again, this is highly speculative (although I would say reasonable) on my part.
I've collected the oil and gas production data from four southern Kansas counties into a spreadsheet. These data are available from the Kansas Geological Survey.
As it happens, Kansas has long been an oil and gas producing state and the industry employs some 100,000 people. The El Dorado Field east of Wichita drove much of that city's early growth (it is still active, although declining). I explored that county's oil production data as Butler County is part of Kansas's current plan for seismic monitoring (more on this later). But I wanted to see what was going on in the "quake zone" of Sumner and Harper Counties, Kansas and see if I could infer anything from what I saw.
In Sumner County, where the November 2014 M4.8 actually occurred, I found that oil and gas production is in a steady decline, a decline continuing since 1971. But what I didn't find was how many of those now exhausted oil wells are now used for wastewater disposal. The aquifer far below this county is saline--it's being investigated for CO2 Sequestration.
Kansas does keep records on the types of wells that exist per county. I used the KGS's search function to search for injection wells, either for enhanced oil recovery or wastewater injection and returned some 460 records. Now, a good number of these are abandoned and plugged--they're not being used at all. But research and observations from Ohio, from Arkansas, from Texas and from Oklahoma strongly suggest that induced seismicity caused by fluid injection occurs within a relatively tight circle around that well and only in certain conditions does it occur farther away. Whether those conditions exist under Sumner County is unknown.
Now, if you've played with the data on my spreadsheet or at KGS you might be saying "but production in Sumner County for both Oil and Gas is declining, so where's the excess water coming from that has to be disposed of." To answer that question, we have to go next door to Harper County, Kansas. Here is where things get very, very interesting.
Both oil and gas production in Harper County, KS have sharply increased since 2010, returning to and surpassing rates seen in the 1970s. Here is a likely source for our saline brine in need of disposal. We can see the same in the county to the west of Harper, Barber County, where production of oil is returning to rates previously seen in the early 1980s. If you note the charts from all three counties, the industry is using more wells to extract oil and gas to return to rates previously seen in the late 1970s and early 1980s. And because I was curious, I explored much of south-central Kansas's data. It is just those two counties that have seen expansive increases in oil production. Even in Butler County, KS, home of the El Dorado Field, production of oil (gas appears tapped out) has been in a slow but steady decline.
Injection Wells are regulated by the Kansas Corporation Commission, and they are of a class of well titled Class II Wells. In Kansas, there are roughly 16,600, 5,000 of which are dedicated to the disposal of saltwater produced during the oil/gas extraction process or hydraulic fracturing fluid. Because of the sharp increase in oil and gas production in Harper and Barber County there would also be a sharp increase in the amount of brine, also known as produced water. A good deal of it would get reused for enhanced recovery operation, but much of it will be disposed of as is standard practice globally. Now, I can't point out a well or group of wells and name them the culprit--yet. I can guarantee that people far smarter than me are going to be taking a very close look. I hope they can find the data (like amount of produced water, rates of injection) that I could not.
(this is a hint, for anyone out there looking for a thesis or a study--here you go!)
CONCLUSIONS?
Induced Seismicity Plan
Kansas does have a plan. Governor Brownback convened a taskforce and tasked them to create a seismic action plan, which they presented in September. Remarkably (from my point of view) the task force could not conclude either way if the current seismicity, despite being located in a small region of Kansas that has seen substantial and sharp increases of production, was induced or not.
Some of the plan is good, such as the construction and deployment of a permanent seismograph network and the purchase of a portable seismograph network for easy and quick deployment. Kansas currently does not have one, and such a network would enable geologists to gain a better picture of what's going on underground. Even micro-quakes, the kind that won't get recorded on distant stations and can't be felt, can paint a picture. This is something I believe all states should have, incidentally, aseisimc or not. 39 US states have at least some seismic risk regardless of human influences.
The plan on whether and how the state would take action if a quake was determined to be induced, or not, read like a bureaucratic mess (and I work in a bureaucracy, so that's saying something). You can read it in its entirety here. Ideas from the Task Force's public hearing on the matter, such as a compensation pool/trust fund did not make it in. This idea comes from the Netherlands, whose gas producing region is also quake prone. While they've slashed production (and we'll see if it stays slashed, given the other place Western Europe gets its natural gas), instead of trying to determine which quakes were the gas industry's fault and which quakes were not they simply have a fund to pay for damages. Not so, here.
I was also struck by public comments that did not want the oil and gas industry to pay for the seismograph network (as if Kansas has any money, although to their credit they have appropriated funds for a bunch portable seismographs ) or any kind of trust fund/compensation pool. You can read the responses to comments here.
Now, I will admit a number of things now:
1. People see patterns everywhere
2. correlation is not evidence of causation
3. we live in the age of the precautionary principle run amok
4. my evidence may be "post hoc ergo propter hoc"
These are fair criticisms, and they seem to be the basis (especially number 2) as to why Kansas's seismic action plan appears, by my eye, to be weak. Could the seismicity be natural? It absolutely could be, but I have a lot of trouble reconciling that with the cursory evidence available.
Smarter people than me are looking into this, I can guarantee it. I personally cannot explain this swarm of seismicity without seeing a human hand in it (although which quakes individually were induced and which ones were later triggered by other quakes in the region, I cannot say--this is an important distinction.) I will note that in exploring the production data, which is only available through July of 2014, that it appears that the two counties where sharp increases in production were seen might not continue this trend. If that's the case, than the production of saltwater in need of disposal will also decline, and if what I suspect about the cause of the earthquakes is correct, so will the earthquakes.
(This diary would not have been possible without the data from the Kansas Geological Survey.)