1. Education
Send to a Friend via Email

Discuss in my forum

Andrew Alden

USGS on Human-Triggered Earthquakes

By July 12, 2013

Follow me on:

Yesterday the journal Science published a pair of papers related to earthquakes that occur near sites where various waste fluids are pumped deep underground. The mainstream news sites tended to make a muddle of these papers, but their upshot is that some patterns are emerging in the relationship between industrial activity and earthquakes.

One paper noted that some injection-well sites responded with little quakes to the seismic waves of very large distant earthquakes (dynamic triggering) in the months before they had earthquake swarms of their own. The idea was that the faults that failed in the swarms were signaling their highly stressed state through their earlier sensitivity to dynamic triggering. It's possible that this pattern could help in managing injection-well fields.

The other paper compared the local earthquake record with the record of steam production in a set of geothermal power plants in southern California and found a link—not with straight steam production but with net fluid production, hot water pumped out minus condensate pumped back. That's a small bit of progress on the larger problem of managing underground processes in highly stressed crust.

More important for public purposes was a third paper, by U.S. Geological Survey seismologist Bill Ellsworth, reviewing the whole topic of injection-induced earthquakes—what many people persist in thinking of as "fracking quakes." Fracking does not cause this kind of earthquake except in a one-in-a-million sense. These are more like one-in-ten-thousand events that occur in areas where deep injection wells dispose of fluids related to petroleum production in areas that happen to have unmapped faults. Science operates behind a paywall, but the USGS has put up its own condensed version of the paper under the title "Man-Made Earthquakes," and it's good reading.


July 15, 2013 at 5:15 am
(1) Ken in San Jose says:

If scientists can figure out how these injection wells cause small earthquakes, could we also figure a way, by using these injection wells to cause small earthquakes to release stress on large earthquake faults. Thus reducing the chance of a major earthquake? Or would there be too much chance of triggering a major earthquake in trying to prevent one?

July 15, 2013 at 1:32 pm
(2) Geology Guide says:

Ken, no one will ever propose such a thing. Think of what could go wrong.

July 15, 2013 at 11:32 am
(3) Barry McMullan says:

Not much mention was made of mine collapses, in Wyoming’s Green River Basin two were on the order of 5.0 and a number of others over 3,0.

July 15, 2013 at 1:31 pm
(4) Geology Guide says:

Barry, you’re right, the USGS page focuses on injection-well quakes. I mention some of the other human causes of earthquakes on this page.

July 15, 2013 at 6:50 pm
(5) Christie says:

To Ken in SJ:
The experiment has been done – google “Rangely Colorado Earthquakes” and some people think it would be worth doing again, under controlled circumstances far from a population center. However, the number of small earthquakes it would take to reduce the stress on a fault equivalent to a large earthquake is impossibly large. Using some very rough scaling relationships, you can compare the size of the fault that slips and the amount it slips for small and large earthquakes. For example, say you wanted to release the stress on the San Andreas equivalent to the M8 1906 earthquake. The fault rupture was 500 km long and 30 km deep = 15,000 km2. A magnitude 4 earthquake (maybe the max you would want to trigger in your stress-releasing projects) ruptures a fault area of about 1 km2. So to cover the same geographical area you would need 15,000 earthquakes. Also, the M8 slipped about 8 m while a M4 would slip < 1 mm. So, you'd have to repeat those 15,000 slips 8000 times each, for a total of 120 million M4 earthquakes. If the recurrence interval is ~300 years, that means at least 1000 earthquakes a day, every day. And, as Andrew pointed out, keeping earthquakes to the target size is not something we know how to do.

Log log log.

- Kanamori + Brodsky 2004; Wells + Coppersmith 1994.

July 15, 2013 at 10:08 pm
(6) Geology Guide says:

Thanks, Christie. “Log log log” says it all.

July 16, 2013 at 2:27 pm
(7) Craig Booker says:

We have experienced a huge increase in earthquakes in Oklahoma. I have lived here 30+ years and have never seen this level of activity.

July 16, 2013 at 2:51 pm
(8) Geology Guide says:

Craig, that’s true. In an area like yours where natural quakes are rare, anything counts as a large increase. And to a population that doesn’t experience them often, it’s an unsettling change. The oil people would like to learn how to avoid earthquakes as much as possible so that the jobs and income can continue.

July 21, 2013 at 2:52 am
(9) J.C.Chang says:

I wouldn’t say never… There have been proposal to trigger “controlled” earthquakes. I was in a little get together at AGU a couple years back, where we discussed different mechanisms for the triggering.

Leave a Comment

Line and paragraph breaks are automatic. Some HTML allowed: <a href="" title="">, <b>, <i>, <strike>
  1. About.com
  2. Education
  3. Geology

©2014 About.com. All rights reserved.