Researchers are pretty sure at this point that the increase in earthquakes in the middle of the US in recent years have been caused by fracking.
The USGS put out a news release in February that starts with this:
Large areas of the United States that used to experience few or no earthquakes have, in recent years, experienced a remarkable increase in earthquake activity that has caused considerable public concern as well as damage to structures. This rise in seismic activity, especially in the central United States, is not the result of natural processes.
You can see in the chart at right that California, the light blue line, has a noisy but relatively stable number of earthquakes every year. Nothing about the dark blue line, earthquakes detected in Oklahoma, looks stable.
The question now is what to do about it.
A new paper in Science suggests that we need to detect much smaller earthquakes than we have done in the past, in order to better predict when a big one is going to hit.
Related story
Right now, according to the paper, the "detection threshold within most of the contiguous United States is M3 [magnitude 3]." The authors suggest that the threshold detected and recorded by the US Geological Survey needs to be much lower — magnitude 2 or lower — in order to detect faults that might be deep underground near oil and gas extraction wells.
Earthquakes caused by fluid injection often happen not during drilling or extraction but in the disposal process. Waste fluid is injected into the ground in a place where there are deep underground faults, which causes the earth's plates to move. And it's been happening a lot, as shale gas production has grown exponentially in the last couple of years.
Scientists know that it happens, but they aren't exactly sure about the mechanisms involved. There is some debate as to whether injecting more fluid into the ground causes more or bigger earthquakes. And most of the time the earthquakes caused are really small, unable to be felt by humans.
But, the authors of this paper think that by detecting more of the super small earthquakes, the USGS might be able to better identify and predict when an injection site has hit a fault line.
"For instance, a seismic network capable of precise locations of small earthquakes could reveal the presence of a large, possibly dangerous, fault being reactivated to to fluid injection," the paper's authors write.
The USGS seems to be taking this seriously.
In fact, according to a news release after the Science article came out (the same one mentioned above), they're already working on it:
The USGS is currently collaborating with interested stakeholders to develop a hazard model for induced earthquakes in the U.S. that can be updated frequently in response to changing trends in energy production.
[Update: A previous version of this post had a diagram of the fracking process without a wastewater injection site. Just to reiterate: the wastewater injection process is what has been determined to cause earthquakes, not the fracking itself. However, there's no wastewater to inject without the hydraulic fracturing. Mother Jones has a really cool gif of the process here.]
