A new study led by a researcher at Cornell University confirms what we’ve known for some time now: that the massive wastewater injection associated with “fracking” (hydraulic fracturing) has greatly increased earthquake activity. In this case, the area of study is Oklahoma.

An inset to figure 1 of that paper shows the dramatic increase, comparing Oklahoma earthquake counts per thousand square kilometers (in red) to what has been observed in California (in blue):


Note that the highest count for Oklahoma — by a substantial margin — is for 2014, despite the fact that when the data were retrieved for this paper we were only three months into the year. The count by now (with 2014 half over) is higher.

I retrieved earthquake data for the region from longitude 103.1 W to 94 W, latitude 33.5 N to 37 N, which includes all of Oklahoma and part of Texas. I only included earthquakes of magnitude 3 or higher. Here are the annual counts, when including 2014 data through the first few days of July:


It looks reminiscent of … what’s that shape? … a hockey stick. By year’s end, the 2014 count will be even higher.

That’s powerful evidence of a causal relationship. But there’s more. The paper gives the location of the main wastewater injection wells in Oklahoma, so from the location of each earthquake we can compute the distance to the nearest injection well. Do the post-2009 earthquakes tend to be closer to the injection well locations than the pre-2009 earthquakes?

Yes, they do:


By this time, it’s abundantly obvious that yes, wastewater injection associated with fracking is the cause of the dramatic increase in earthquake activity in Oklahoma. And the increase has indeed been dramatic; the aforementioned paper states:

Seismic swarms within Oklahoma dominate the recent seismicity in the central and eastern United States (9), contributing 45% of M3 and larger earthquakes between 2008-2013 (10). No other state contributed more than 11 % … These earthquakes are part of a 40-fold increase in seismicity within Oklahoma during 2008-2013 as compared to 1976-2007.

Considering how obvious the causal relationship has been, for some time now, I have to wonder why the fossil fuel industry has spent so much time and effort trying to deny it. Considering how undeniable the relationship is now, I have to wonder whether or not the fossil fuel industry will continue to deny it.

43 responses to “Fracklahoma

  1. Of course the answer to your question has nothing to do with the evidence or the facts– it turns exclusively on what they think they can sell to the people who count.

  2. Bern from Aus

    Of course they will continue to deny it. Just like the tobacco companies fought on long, long after the link between smoking & cancer was abundantly clear, by trotting out their tame ‘scientists’ and unleashing the hordes of lobbyists upon government, along with bucketloads of cash.

    It will probably buy them a decade or so, if they’re lucky (and we’re not).

  3. I started a small environmental company in the early 1990s featuring a very low cost (and at the time) novel airborne multispectral digital mapping service that could quickly and cheaply identify oil spills, leaks, etc., among other things. The business model based on regulatory compliance, loss of product and imputed cleanup costs made sense, at least on paper. We quickly found that the fossil fuel industry (and any related industry) was not at all interested in this service and was very keen on making sure to subtly dissuade anyone else from using it. The reason was plausible deniability – something not accounted for in our business model. The same is happening here. It has little to do with what we (normal people) perceive as common sense based on physical facts, but does make (odd) sense based in terms of legal exposure, minimizing fines and delaying costs (i.e., its all about the money). By not finding the cause themselves a company can avoid a certain level of fines and legal responsibility (What? No one ever showed this to be a problem until now! Gosh! We never knew!). From their “isolated world view” position denying makes perfect sense. The rest of us are left scratching our heads. Fortunately/unfortunately the environmental disasters of the last 10 years or so may be making this approach less cost-effective, hence less practical. We shall see.

    • On the other hand, my experience is that some companies do run helicopter borne sniffers. The surveys are run over pipelines, facilities and well sites. The provide a baseline which can be used to identify problems and avoid lawsuits. I´ve also seen sniffer surveys which located natural seeps. When the ground is soaked with hydrocarbons and there are natural leak pathways, they hydrocarbons can end up leaking all over in very tiny amounts.

      Regarding the earthquake issue, simple common sense tells us that high pressure injection into naturally pressured (undepleted) reservoirs can lead to seismic events. The question is whether these events are harmful, and whether eventually they will lead to the “big one”. This means a simple seismic event frequency count isn´t the whole answer. What you need to do is tabulate the cumulative energy released by these seismic events.

      From a regulatory standpoint, if I were the regulators in Oklahoma I would look over very carefully at the injected volumes and have the operators report back the static pressures and dynamic pressures observed in each injection well, as well as submit a report by a qualified (certified engineering) consultant regarding the effect of the overpressures they create on the local earth stress field.

      [Response: The total power of the very recent swarms of earthquakes in Oklahoma exceeds the power in pre-2009 earthquakes. By a lot.

      I wouldn’t trust a report by a “certified engineering” consultant. I might trust a report by a *qualified seismologist* consultant.]

  4. Wouldn’t that be Frackenstein? or Frackula?

  5. No fracking near the San Andreas fault then?

  6. JGradie The solution for that problem would be to double the fines for companies that don’t take reasonable precautions to detect potential problems. Ignorance is only an excuse when it is unavoidable, when there was no way you could possibly have known.

  7. It is not actually stated, but I assume fracking in Oklahoma started in 2009 (or late 2008).

  8. I have a nitpick with this – surely if you are comparing to regions for seismicity, you have to give an indication of total seismic energy release, given how this release is usually dominated by infrequent, high magnitude events.

    I’m not a great fan of fracking, but still.

    (John B above – I believe it has been suggested that we do pump water into potentially dangerous faults to try and get them to move via creep rather than fracture. The insurance premiums for such an undertaking may be slightly high, though).

    • Indeed. It’s quite clear that fracking injection wells increase quake frequency. But that’s not necessarily a bad thing. If what is actually happenning is the release of the stored energy in lots of small events sooner, rather than one big one later, is that actually a problem?

      Fracking is a terrible idea because we already have too much accessible carbon, but probbaly not because of the earthquakes.

  9. The already-standard denial line is “fracking doesn’t cause earthquakes, injection wells do!”

    Often followed by “You’re a liar for saying fracking causes earthquakes!”

    • Horatio Algeranon

      One of my favorite songs by Ellis Paul is Autobiography of a DissemblerPistol in which he says
      “Guns don’t kill people, it’s the bullets that do”

      Applying similar logic “Power plants and cars don’t cause global warming. CO2 does” So, humans are not responsible. QED.

  10. I live in Edmond. I felt my first quake at age 31 in 2011. My son felt his first quake at 10 months. Seems natural ; )

  11. metzomagic

    Now if we could only get figures of similar veracity concerning ground water/aquifer contamination, which I personally consider to be even more of a bad side effect of fracking with long term consequences for the local residents, then we’d have a good chance of shutting these greedy so-and-so’s down once and for all.

    • Horatio Algeranon

      “Fracking Trade Secrets”
      — by Horatio Algeranon

      Those seismograms are classified
      A secret of our trade
      Like water that’s been gasified
      And toxic waste we’ve made

      Tamino will be notified
      To cease and then desist
      From posting fracking/earthquake slides
      On such a site as this

  12. Horatio Algeranon


    — Horatio Algeranon’s rending of Rogers and Hammerstein

    Fraaaacklahoma, where the quake comes sweepin’ down the crack
    And the fracking oil can sure smell foul
    When the quake comes right behind the frack.
    Fraaacklahoma, Ev’ry night my honey lamb and I
    Sit alone and shake, from the fracking quake
    Thinking “Now we’re really gonna die”

    [Response: (Applause)]

    • And now we belong to the land,
      ’cause it’s shaking to damn hard to stand
      We’re singing Fraaacklahoma’s shaking
      Fraaacklahoma’s shaking
      You’ll lookin’ shaky Fraaacklahoma
      Fraaacklohoma, strike-slip away!

  13. You know the drill.

    1) It’s not happening.
    2) It might be happening, but it’s not us.
    3) Even if it is happening, it’s not so bad.
    4) Adapt.

  14. Aaron Lewis

    RCRA contains a powerful exemption stating that a “waste” is not a waste it it has a beneficial use. One driver of fracking is that they can pump refinery waste into the ground by pretending that it is a beneficial use. This avoids all the costs and permitting associated with hazardous waste disposal by injection well.

    People may not like fracking, but “fracking” never sounds as as bad as hazardous waste injection well.

    Fracking is just another name for a hazardous waste injection well disposal of refinery waste.

  15. thoroughly grim news for Oklahomans.

    out of interest (since you appear to have the numbers to hand), approximately how much of the decrease in your final graph can be attributed to the increase in the number of wells? on the basis that the more wells there are, the less far any given point will be from a well. or am i misunderstanding what you’re plotting there?

    [Response: I avoided that issue by computing the distance of each quake from the nearest well, even if the quake happened long before the well existed. That way, I have identified a certain change in the geographical distribution of Oklahoma earthquakes.]

    • One could also compute the distribution curve for an equal number of randomly chosen points within the state boundaries, and show that the result is significantly different from the quake epicenters.

    • thanks, that makes more sense!

  16. “By this time, it’s abundantly obvious that yes, wastewater injection associated with fracking is the cause of the dramatic increase in earthquake activity in Oklahoma.”

    Not so fast.

    From the second page of the study:

    “The large disposal wells are part of dewatering plays (Fig. S4). Dewatering production wells produce substantial wastewater volumes with initially up to 200 times greater water per barrel of oil than conventional production wells, and up to 1000:1 water to oil ratio (16-17).”

    Dewatering plays have nothing to do with hydraulic fracturing. They have everything to do with draining vast amounts of water out of a reservoir to get the last dregs of oil out of it. It might take a year or more before any appreciable amount of oil starts flowing out of these wells and these wells are not hydraulically fractured. This water, of course, has to be disposed of, so it gets injected into disposal wells.

    It’s one type of unconventional development versus another.

    I won’t argue that hydraulic fracturing waste water can’t cause earthquakes during disposal (of course it can, just like any fluid could). But, there are hundreds of thousands of legacy oil and gas wells in the United States that aren’t part of post-2008 oil and gas development yet are still producing vast amounts of wastewater that needs to be disposed of.

    So, yes, it’s still very misleading to call it a hydraulic fracturing problem, at least for these Oklahoma wells. You can make an argument for this in states that weren’t injecting significant amounts of fracking-related wastewater before (like Ohio), but it’s not applicable everywhere, like Oklahoma and probably even Texas

    I think the overall lesson from this paper is that wastewater injection, regardless of the source, not only induces seismicity (this has been known for a long time and isn’t really a surprise) but can also induce seismicity a surprising distance from the disposal well in the case of large-volume injections. 35 km distance? Jebus. Regulators ought to be looking at this study carefully.

  17. “Look! There goes a squirrel!” – FF industry spokesman

  18. The drastic increase in earthquakes in Oklahoma is due, according to a study I linked several posts back, to them having fracking operations on fault lines there. Stop fracking on faults, and you’ll stop the earthquakes. Seems like it should be a pretty easy fix and a rather commonsense thing to do.

  19. I made a big post, only to see that I misread you. So everything in the bin, and done once again.
    I will only quote two facts that may enlighten the people here :
    – I happen to know a well placed guy in a seismological monitoring company which is now monitoring fracturation operations for oil companies. He told me the usual monitoring setup is several thousand geophones forming 5 or 6 clusters in a small area around the fracturation target (some kilometers at most). They are not that strong in geomechanics and risk analysis, so they have the following rule of thumb : anything above M=0 triggers an alert, and they advise (advise only, they can’t decide since they are external contractors) to stop the operation. Fracturation in itself is not the most dangerous thing *for usual oil operations* (I have seen Mw>2 when a geothermal exploitation tried to reopen the fracturation network in a granitic formation 4 km below the surface)
    – Wastewater disposal is more worrying and *known* to spell trouble when it is injected into faults ; once again, geothermy experts know that since the Basel earthquake (magnitude >4 right under a big city, no deaths fortunately but several damaged houses and lots of angry owners of said houses). Hence the very good clustering you obtain when you look at the distance well-earthquake.

    there are methods to draw the graph parent-children for earthquakes : a “father” earthquake eases the stress where it occurs, but transfers some stress to other points thus triggering earthquakes and one more round. Finding “father” earthquakes and looking at its location could be fun. Another thing to do : earthquake relocation for clusters, to see where they occur and have a better location – and see if some might happen by chance right under the well.
    Lots of interesting things to do. Of course, some oil companies may not find it very “interesting”. And we may see some people claim “it’s a natural cycle” (and our host knows well what I think about such claims and what kind of incoherent bluster I may spout out of rage :] )

    I will try the relocation thing, if I am able to get a not too bad speed profile for Oklahoma (I may switch to PREM, but PREM is for the whole earth …).
    Did you get your earthquake location datas from USGS ?

    • well, disregard my questions – of course the authors did the relocation job, I just saw that in the supplementary materials. And with a local speed model inverted jointly with locations, because they are doing a good job.

  20. last post, I promise, but I ventured into Wattsland.
    Of course, I found a post by the Master himself. Of course, he uses the old tricks. And of course, the comments are … well.
    Second comment ; “why did they use km instead of miles ?” WHY ??? (velest, hypoDD and 100% of seismology software use meter/kilometer disregarding ‘murica feelings for that matter).

    Oh dear. And I will go on reading. Not commenting, of course. Utter waste of time.

  21. It is pretty clear that liquid injection into faults will cause them to slip, leading to earthquakes. However, relieving stress is not necessarily a bad thing remembering New Madrid, e.g. a large number of smaller slips, may be a lot better than one big one, and the injection appears to be causing smaller slips.

    [Response: Not in the Oklahoma data. There’s no sign of reduced earthquake magnitude, and the largest in the data set occured in late 2012.]

    • Careful with that reasoning: it takes about 1000 magnitude 4 quakes to release the same energy as a magnitude 6. You’d have to trigger many, many little quakes to substantially reduce the risk of a bigger one.

    • Are there papers evaluating whether the relieve-small-amounts-of-stress hypothesis is actually true? I hadn’t seen any last I checked — nor any good arguments to disbelieve it.

    • The size of the quakes from liquid injection would not necessarily be smaller, the point is that this might reduce the chances of a really big one like New Madrid. On the geoengineering side there have been proposals to induce smaller quakes in CA, but a) the risk is too high there because of the accumulated stress, and b) it is not clear exactly what would happen. Willy nilly with fracking we are in the test tube.

    • three points :
      – as noted above, you need lots of small earthquakes. I don’t think the number of earthquakes is enough now – something to compute, but 1000 mag 2+ is not enough to diminish a Mw 6 to a Mw 4. Maybe to a Mw 5, which can be viewed as a bonus – but a shallow Mw 5 in a urban area can be delicate to handle. But, yes, given *many* small earthquakes, you may ease a big one. But …
      – I’m not sure at all that this reasoning holds if you are in a locked situation. Small earthquakes “ease” down the stress where they happen, but build up stress at the extremities of the ruptured fault plane (if you wish, I could find some articles about that), because you compress/extend the rock at these points. If the stress buildup strikes where the fault is not locked, fine, you trigger another small earthquake at this point, and the ruptured fault plane extends a bit – and won’t be sollicited for a bigger one. But if it happens where the fault is locked, it will add stress without rupturing anything. Your small earthquake didn’t do anything, and the big one awaits the fault unlocking. This situation happened with Haiti some years ago.

      Since I’m not familiar with the New Madrid fault system, I would be cautious about the efficiency of stress release through small earthquakes.

    • horatio Algeranon

      “relieving stress is not necessarily a bad thing remembering New Madrid”

      …or necessarily a good thing, remembering New Madrid:

      “Even if a fault zone has recently experienced an earthquake, however, there is no guarantee that all the stress has been relieved. Another earthquake could still occur. In New Madrid, a great earthquake was followed by a large aftershock within 6 hours on December 6, 1811. Furthermore, relieving stress along one part of the fault may increase stress in another part; the New Madrid earthquakes in January and February 1812 may have resulted from this phenomenon.

      How Earthquakes Happen (by USGS)

  22. “You can prevent large earthquakes by making lots of small ones, or by “lubricating” the fault with water
    “… there are never enough small ones to eliminate the occasional large event. It would take 32 magnitude 5’s, 1000 magnitude 4’s, OR 32,000 magnitude 3’s to equal the energy of one magnitude 6 event. … As for “lubricating” faults with water or some other substance, if anything, this would have the opposite effect. Injecting high- pressure fluids deep into the ground is known to be able to trigger earthquakes—to cause them to occur sooner than would have been the case without the injection. This would be a dangerous pursuit in any populated area ….”

    Seems to me the crust likely is cracked everywhere and what we call ‘faults’ is what our mayfly lifetime notices happening, but I’d guess that as strain moves it’s more a matter of cracks starting to move, rather than of vast expanses of uncracked crust developing a new fracture.

  23. PS, evidence that Tesla may have been right?

    I wonder what the electrical potential can be, after deeply burying a metal pipe full of highly saline water, compared to undisturbed ground — under a thunderstorm?

  24. The fossil fuel industry continues to say that the latest earthquakes near the waste water injection sites are no big deal, according to this article:

    “Energy in Depth, a research group started by the Independent Petroleum Association of America, says on its Web site that “hydraulic fracturing has been catching some unmerited headlines lately” as a result of questions about seismic activity.

    “Despite what you may have heard, hydraulic fracturing has been a safe and proven technology for decades and does not pose a major risk of inducing felt seismic events,” the statement says.”

    • The comment from the Petroleum Associate sounds familiar –
      “…has been a safe and proven technology for decades and does not pose a major risk …”

      We may have heard that applied to other products:
      Red Dye #4
      the Corvair car
      chicken and beef processing – recalls at various times.

      We really cannot let industry spokesmen pronounce on risk