Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Oklahoma earthquakes, seismography

  1. May 12, 2015 #1
    Davenn, I live in Oklahoma in the US, and you are probably familiar with the thousands of EQs we've been having in this state, in 2014 it was over 6000 (maybe even 7000+) of all magnitudes. It's hard to tell exactly because the Oklahoma Geological Survey stopped analyzing/reporting all magnitudes some time last year. Now, I think they're only reporting ones that are above 2.0, and still not even all of the of them until you get to about 2.5. However, even raising the specific magnitude, we still get 15-25/day reported. We used to get 15-25/day when they included even the .9s and 1.5s, etc, so things are obviously getting worse (I don't need to look at the numbers to know that though, considering I have felt hundreds over the past 2.5 years).

    Anyways, do you personally own a seismometer? I am considering getting one, even more so now that the OGS doesn't report them all. And unlike most people think, you CAN feel very weak ones when they're close, and so shallow; I have even felt a 1.9! Distance makes a huge difference, the 1.9 was only 1.5 miles away. A ~3.3 about three miles away feels "approximately" the same as a ~4.0 six miles away. Does the energy fall off in an inverse squared in a predictable relationship? I am guessing it's not that simple because I have read that EQs that occur on the east side of the Rocky Mountains can be felt further away than ones that happen west of them...

    Also unlike most people think, there ARE damages to homes from weak ones as well, primarily because of the "swarm" nature of them (e.g. I had about 1,000 of them less than 10 miles of me just in 2014, about 100 of those were 3.0+). Also because of the swarm nature of them, they damage your mind too - they are a very big nuisance, especially in the middle of the night!

    p.s. I don't mean any insult to your intelligence by saying "US" after Oklahoma above, but I also don't want to be arrogant and assume everyone all over the world knows each of the states in the US... but since you seem to follow EQs a lot, you probably know that state in particular!
  2. jcsd
  3. May 12, 2015 #2
  4. May 12, 2015 #3


    User Avatar
    Science Advisor
    Gold Member

    Hi there
    yeah, all that fracking they are doing there and also in Arkansas is causing lots of problems. Tho I'm not sure if fracking is the main cause with the OK events ?
    Definitely is for the AR ones

    yes on the first page of this thread you can see seismograms from my system :) I record data from 6 different sensors
    I have been doing seismology from home since the early 1990's from New Zealand, then set up again after moving to Australia

    For sure, it is rare for swarms to have significant large events. Generally all the events are of a similar magnitude
    One place where that "general rule" was broken was in a town near Reno, Nevada some years back ... there was a good video on youtube I saw/downloaded about that swarm

    No problems LOL ... one thing about people that do geology is they tend to have pretty good geographical knowledge of the world in general :smile:

    For me it is also helped by the fact that I have spent quite a bit of time traversing OK and the rest of the mid-west doing storm chasing :smile:

    Last edited: May 12, 2015
  5. May 12, 2015 #4


    User Avatar
    Science Advisor
    Gold Member

    cool .... you will see from there all the gear I am running

    Have a look on Larry Cochrane's site .... http://psn.quake.net/#Equipment

    he has geophones, datalogger hardware and software and analysis software for sale
    Larry in also an amateur seismologist living in Redwood City ( south side of San Francisco) and is a brilliant electronics
    designer and programmer and his gear is very affordable
    Dozens of us around the world are using his gear and software ... its good because of a compatibility of recorded data etc

  6. May 12, 2015 #5


    User Avatar

    Staff: Mentor

    I believe the USGS is still monitoring the earthquakes in Oklahoma, as well as the other 49 states.

    http://earthquake.usgs.gov/earthquakes/states/oklahoma/seismicity.php [Broken]

    Reawakened Oklahoma Faults Could Produce Larger Future Events
    Released: 3/6/2015 12:31:58 PM

    It seems that it is well established that injection wells correlate with an increase in earthquakes in the vicinity of the injection wells.

    New Insight on Ground Shaking from Man-Made Earthquakes
    Released: 4/23/2015 9:00:00 AM

    In Pennsylvania, it was recently reported that fracking chemicals are now being found in various wells for drinking water, despite assurances that that would not occur. That's another matter though.
    Last edited by a moderator: May 7, 2017
  7. May 13, 2015 #6
    Sorry in advance for the long post! :)

    I have been following the studies and articles quite carefully for the last year or two, so I am pretty caught up on the increasing consensus among geologists that the wastewater injection wells are likely to blame for nearly all of the ones we've been getting... I came across an agenda for a workshop held by the society of petroleum engineers with participants from the industry, academia, US and state geological surveys, etc... The listing of presentation titles is indicative that the industry is well aware they can induce seismicity. This is extremely frustrating because publicly, the industry down-plays the link, saying things like, the "majority" of them are natural, there is no science that can prove it, etc... However, it's not JUST the industry denying or downplaying the link; there are still geologists with the USGS that are not convinced...

    I download the OK catalogs from the OGS website and analyze the numbers on my own, one attribute I added shows approximate distance from my house... I can filter by any quantity, and have plotted the numbers in various ways, such as time of day relation to magnitude and frequency of occurence (frequency as in EQ quantity, not Hz of seismic waves), and a bunch of other ways to look at the data.

    While yes, the USGS is still tracking them, they do not record nearly as many as the OGS, but I have also seen an article that said the OGS wasn't reporting some that the USGS reported. Also, there was a recent change in the OGS's reporting, they barely record any that are less than 2, and only some of the ones between 2 and 2.5... Above that, they get most of them. For example, looking now, USGS 1-day of all magnitudes shows OK having three (3.1 on 5/12@15:59UTC, 2.5@19:11, 3.2@21:30, and nothing through today at about 14:30). The OGS, going back to yesterday at 14:30UTC, shows we had 13, but that might not be complete yet b/c everything is done manually and they don't work outside normal business hours, which is why you can't use the OGS website to see them over the weekend until late monday or tuesday. I have read the USGS doesn't show less than 2.5 for OK, but I don't think this is 100% accurate, as there have been ones higher than 2.5 that USGS didn't report but OGS did, and there have been ones lower than 2.5 that the USGS has reported. CA shows tons of 1.X's, and I assure you, we are having tons of 1.X's too, but nobody is reporting them.

    My issue/concern with OGS raising the specified magnitude for reporting is this will make the data severely skewed whenever you look at all magnitudes. I checked the OGS catalog for a few days after I felt one about 3:45am and never saw one reported. I emailed and they went back and looked at the seismographs and saw there was a 1.9 only about 1.5 miles from me, but they didn't record it in the catalog! That's when they told me they are no longer reporting them all. After I found this out, I went and looked at the data and you can see they made this change gradually in 2014. If you look at EQs that were <2.0 in each quarter... In 2012 and in 1Q2013, there were about 120-180 <2.0 each quarter. Then, in 2Q2013, there were 541. This number stayed in that range for three more quarters, until 2014Q2 when the number of <2.0s began decreasing, but the number of >2.5 was increasing over 2013 numbers. I will show the numbers below and you can see the decrease of <2.0 reported, and the increase of 2.5+. Also, when I just looked at my data, it turns out there was one day (April 16th, 2013), we had 133 EQs of all magnitudes just that day, including a 4.4 and a 4.2, but that was back when they were recording all of them... If that many happened today, instead of 133, I'd guess only about 20-30 would be reported. It seems like they have a finite amount of time/resources devoted to analyzing them, so they pick the strongest ones that will be over 2.5, and then if they have time left in the day, maybe they do some of the lower ones, but it seems once it's the end of the day, if they've picked up all of them over 2.5, they're done for the day and they may "eventually" come back to it.

    Here is the decrease in reporting of <2.0 compared to the increase of 2.5+
    QuarterYear: (<2.0)/(2.5+)
    1Q2014: 549/294
    2Q2014: 305/460
    3Q2014: 208/468
    4Q2014: 132/753
    1Q2015: 40/826

    So, while it "appears" as if there is a a 14-fold (549 to 40) decrease in <2.0, there was nearly a 3x (294 to 826) increase of 2.5+. As I said, this skews the data and masks part of the issue! It appears as if we're having a similar number as we had in early 2014, but really we're having a lot more, especially if they are following the Gutenberg-Richter "law" that predicts occurrence. It's a problem, IMO, that people are feeling EQs that are enough to be a nuisance and wake you up, but yet they aren't getting reported by the OGS nor the USGS. here is more info http://www.okgeosurvey1.gov/pages/earthquakes/faq.php [Broken]

    @davenn Thank you for your reply! I will take a look at Larry's site! I was even wondering if just a geophone hooked up to an oscilloscope set to do a single trigger may pick them up... You mentioned with swarms, they are generally of similar magnitude, i beleive over the past year, they have been increasing in frequency and magnitude... reading about the Gutenberg-Richter "law" concerns me since they're increasing in frequency, and magnitude. I should take my data and look at it to see if that "law" has held with these (likely) induced EQs, or if the law holds mostly just for natural ones. Since I have all the data in excel, I should be able to use a pivot table or something to pull out the numbers... This is from the OGS site:

    "The increase of magnitude of 3.0 and greater earthquakes indicates a greater possibility of having a magnitude 4.0 or greater event in the future. This arises from the Gutenberg-Richter law which expresses the relationship between magnitude and frequency of earthquakes in a given region and time period. Simply stated, for every 10 magnitude 2.0 earthquakes we have we can expect a magnitude 3.0 earthquake. So, for every 100 magnitude 2.0 earthquakes we have we can expect 10 magnitude 3.0 earthquakes and one magnitude 4.0 earthquake. This relationship holds worldwide and does not vary significantly from region to region or over time."

    Also... unrelated to the science and getting a little political, I don't know if you guys saw the latest "controversy" regarding the EQs in OK, but in short - it's like this: The state geological survey and the state geologist are based out of The University of Oklahoma, OU. Supposedly, the president of OU asked the state geologist to meet with him, and also present was the CEO of an OK-based energy company, Continental Resources. This led to speculation that the state geologist was pressured to refrain from blaming it on the industry. What has fueld this speculation are a few things... most notably, he was suppsoedly quoted as saying, something to the effect of, "You don't understand, Mr. Hamm [CEO of cont. resources] won't let me talk about certain things." There were other things that fueled speculation, such as the CEO had recently given $20M to the university, AND the president of OU serves on the board at Continental resources and receives cash and stock compensation from the company (I believe it's common for presidents of universities to have relationships with companies such as this). http://www.eenews.net/stories/1060014342

    Sorry for the long-winded post - I had a lot of coffee and ate lunch at my desk today! When you feel (and hear) as many as I have, and you like math and science like we all do, it's hard for me not to get interested (and maybe slightly obsessed) with the EQs. If I didn't work 30 miles away and was at my house most the day, I'm sure I'd probably hear/feel maybe three or four times as many as I do. That's reminds me, another thing that is supposedly different about a lot of the OK EQs... They are all fairly audible. It's sort of a cross between a sonic boom and thunder... it generally starts soft and gets loud then dies off (over a matter of just a second or two), like distant thunder sometimes, but other times, the "rise time" is much shorter and it's more like an impulse, like a truck hit the house.

    Anyways, I'll keep you updated if I buy/make some equipment!
    Last edited by a moderator: May 7, 2017
  8. May 13, 2015 #7


    User Avatar
    Science Advisor
    Gold Member

    yes but don't forget that that study was done when referring to natural earthquakes
    there was no such thing as fracking back in those days

  9. May 13, 2015 #8
    Right... That's why i said this... :)

    "I should take my data and look at it to see if that "law" has held with these (likely) induced EQs, or if the law holds mostly just for natural ones. Since I have all the data in excel, I should be able to use a pivot table or something to pull out the numbers..."

    I'm sure it was easy to miss, there were a lot of words in that post! :smile: The problem that may come in though now, since they stopped recording many of the ones less than 2.5 and what appears to be nearly everything below 2, i woudlnt have as many data points. However, going back to 2011, I think there are about 12,000 lines in my spreadsheet.
  10. May 13, 2015 #9


    User Avatar
    Science Advisor
    Gold Member

    yes, but unless you have a sample and hold on the scope and or a scope with some form of data logging, you will miss seeing/recording them

    you will find that seismic recording is quite captivating and is easy to get involved with and you would find before long that not only do you want to record the local events but the also the big distant ones.
    Geophones of around 1Hz to 4.5 Hz are great for local and regional events out to several 100 km's. my 4.5Hz 'phones easily show the P and S arrivals of the large New Zealand events 2500-3000 km away. But for better recording of those more distant large events longer period sensors are a must. The home brew "Lehman" seismometer ( pioneered by Jim Lehman --- an awesome guy) did wonders for private amateur seismic stations and their recording of distant events. A large number of us guys around the world have build varying versions of these over the last 30 - 40 yrs since his design came out in the Scientific American article of the 1970's. Its easy to get down to around 10 - 15 second period on these sensors.

  11. May 14, 2015 #10
    Well... I don't care about recording distant events as much as I do very weak local events... I mean like the 1.9 that I felt and heard, but wasn't published anywhere that I saw, only after contacting the OGS did I get the data... That's what kind of bugs me! I am assuming you wouldn't see anything where you are from a 1.9...
  12. May 15, 2015 #11


    User Avatar
    Science Advisor
    Gold Member

    my geophones will record a M 1.9 out to about 60 - 70 km and still be able to easily identify the P and S arrivals

    got a M 1.7 a week or two ago ....

    wow just doing a google earth plot ... much better than I quoted above !! :smile: M 1.7 @ 93 km

    distance to a M 1.7 from home.JPG

  13. May 15, 2015 #12


    User Avatar
    Science Advisor
    Gold Member

    Ohhh you probably will :wink: it becomes very addictive watching the trace start doing its wobbles
  14. May 15, 2015 #13
    Uh oh... that's all I need, to become MORE obsessed with them ;)

    My o'scope does have single trigger and hold, but it bases the sample rate on the time scale... In other words, if the time scale is 10ms/div, a 60hz sine wave (line voltage) is pretty much a perfect sine wave. however, if I do 1s/div, which will give me a 10s wide sample (which would be long enough for most of the EQs here, I think), then if I "expand" it to 10ms/div, the sine wave appears as a sawtooth. In other words, it appears to do a fixed number of samples per division when it is sampling, so as your seconds/division changes, your samples/second changes. Although, considering the low frequency of an EQ, sampling at 1s/div would probably give plenty of fidelity if I wanted to zoom in to a 100ms/division or something, and still allow a 10s long sample... I don't mean to imply 1s/div is one sample/second, it's much, MUCH faster than that... I'm guessing an O'scope, even at 1s/div, is probably plenty fast enough for EQ measurement. I'd have to get my signal generator out and see what it looks like in the 5-10hz range when sampled for 10 seconds and then expanded to something like 100ms/division or something... Is my thinking correct? If I bought just one geophone, I would install it in the Z direction I assume? Sounds like my scope may be able to serve as my A/D and data logger! :)
  15. May 15, 2015 #14


    User Avatar
    Science Advisor
    Gold Member

    To be getting clean traces for local events, you need to be sampling at approx. 2.5 times the highest signal freq. That being around 10Hz, you should be sampling at around 25 - 30 times a sec. Google Nyquist rate relative to sampling, if you would like to know more :smile:
    A lot of us tend to over sample ... mainly just because we can haha and it gives us plenty of data points to play with when filtering etc
    generally I'm sampling at 100 sps .... I really don't need to, it just takes up extra hard drive space, maybe one day I will drop it to 50 sps and see how it looks

    There are horizontal and vertical seismometers and one cannot be used instead of the other.
    It would be preferable to use a vertical one as then you would be able to record events with equal sensitivity from any direction

    does you scope have the ability to output the data to a PC ? ... that would be a great advantage for your situation

    Last edited: May 15, 2015
  16. May 16, 2015 #15


    User Avatar

    Staff: Mentor

    Oil CEO Hamm sought ouster of scientists looking at quakes: Bloomberg

    There are five earthquakes (or seismic events) in Oklahoma, so far today.
  17. May 18, 2015 #16
    Perhaps the fracking is easing tension and might lower the occurrence of potentially more dangerous eqs .
  18. May 18, 2015 #17
    @davenn: I am familiar with nyquist (I'm an EE), but I am not familiar with the highest frequency that I need to reproduce on the scope, but see you said 10 Hz, so the scope is plenty fast enough... It will do 1GS/s down to 50S/s. It will record up to 2500 data points, so at a 5s/div, one "screen" is 10 divisions, which would be 50 seconds, and at a rate of 50S/s, that agrees with the 2500 data points in the spec... page 47. http://mmrc.caltech.edu/Oscilliscope/TDS210 Digital Os User Manual.pdf

    @Astronuc: I saw that too, I was coming here to share the story... This is what I was talking about in the second-to-last paragraph in post number 6 above! What the Bloomberg article left out was that not only does the president of The University of Oklahoma serve on the board of the oil company, it's not a "volunteer" position... he is paid something like $250k/year in stock and other compensation if I recall.

    @psycho rich: At least with natural EQs, that is not true... little ones don't alleviate large ones. It's actually the opposite... The more small ones you have, the more large ones you'll have. See above where davenn and I discuss the Gutenberg-Richter "law" (not really a law, it's similar to moore's law, it's a rule of thumb from observations over many years). If I get around to it, I will take a look at the Oklahoma events to see if it looks like the law holds for the ones here, since the "law" came as a result of natural earthquakes, it may not agree. Also, I am not confident I have all of the smaller ones beginning mid 2014.
  19. May 28, 2015 #18
    It's pretty sad when I can't say how close a reported location of a specific EQ is from my house without practically giving away my address...
  20. May 28, 2015 #19


    User Avatar
    Science Advisor
    Gold Member

    that's OK ... I wont tell anyone else :wink:
  21. Jun 15, 2015 #20
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Oklahoma earthquakes, seismography
  1. An earthquake (Replies: 4)

  2. Oklahoma earthquakes (Replies: 3)

  3. Earthquake prediction (Replies: 27)

  4. NZ Earthquakes (Replies: 32)