Tuesday, April 2, 2013

Damaging earthquake caused by drilling?

In November of 2011, a series of earthquakes struck just east of Oklahoma City. There were a total of 5 earthquakes of magnitude 5.0+, including a magnitude 5.7 quake on November 6, 2011 (overnight). That quake was the largest ever recorded in the state of Oklahoma and was felt across the Midwestern U.S.

New research published this week by the journal Geology makes an explosive claim about this earthquake; that it was directly caused by water injections related to oil drilling.

A bit of background on Earthquakes is in order to discuss this topic. An earthquake happens when rocks have been stressed, pressed against each other to the point that they break and move.

Just about every rock in the world is under some stress, even far away from plate boundaries. Plates transmit stress through them as they move. So, even in a place like Oklahoma, the rocks are under some stress, possibly driven by forces thousands of kilometers away.

Normally, a place like Oklahoma will just transmit this stress through it. The rocks will behave “elastically”, they’ll transmit stress and even bend, but it won’t be enough to break the rock or cause an earthquake. But, if a fracture or fault already exists, that fracture can reach its breaking point. Whether a fault breaks or transmits the stress it is a complicated question, but can be boiled down to a simple example.

Put a book on the desk next to you and push it on the side. It should slide fairly easily. Now, do the same thing, but first, push down on top of the book as hard as you can. The book should be much harder to move; when you apply pressure to the book, even if there’s a boundary between them, it’s much harder to make it slip. A fault works just like this; when rocks are squeezed tightly together, they do not break easily and they’re more likely to just transmit stress through them.

Now we come to the Oklahoma case. In this region, oil drilling has been going on for decades. One thing many may not know is that while extracting oil, most companies also pump water into reservoirs. This pumping happens for 2 reasons: first, it prevents contaminated water from being dumped into rivers, and second, most oils float on water, so pumping water into an oil well causes the oil to rise up and float towards the top of the reservoir, where it is more easily extracted.

Now, just to be clear, this is different than hydraulic fracturing (fracking) which has been in the news lately and covered previously at TES. Fracking is done with high pressure fluids that break rocks deliberately; this is just pumping water into the ground at low pressures.

However, water plays an important role in faulting. In a fault, water acts like a lubricant. Remember my book example? That extra force you applied to the top of the book…when rocks are saturated with water it counteracts that force, it pushes back up at you. Pumping water into a fault zone thus makes it easier for the rocks to slide; it’s like taking away the extra pressure you put on the top of the book.

Geologists have long known that pumping water into the ground can lead to minor earthquakes for exactly this reason. They are directly associated with pumping in some areas; however, most of the earthquakes are small they’re hardly noticed, if at all.

This region of Oklahoma was not strongly active seismically for most of recorded history. However, in ~2007, that started to change. The number of earthquakes in the state jumped from ~5 to 50 a year, with most of them happened in this region east of Oklahoma City. These events built until the large, damaging earthquakes in late 2011.

This region has undergone oil drilling for nearly 20 years, including pumping of water into the ground, so the quake surge wasn’t associated with the start of pumping. But, researchers from the University of Oklahoma and Lamont-Doherty Earth Observatory investigated the sites near the pumping and found that water levels have increased substantially due to this long-duration pumping. Rather than setting off quakes immediately, it took over a decade to build up enough water pressure to affect the nearby faults.

Perhaps most convincingly, the 2011 earthquake happened only 200 meters from the site of an injection well, and occurred within the same rock units having water pumped into them.

The Oklahoma Geological Survey and the EPA stated in 2011 that they believed the earthquake surge was naturally occurring, and they have not yet altered their original statements. But, if this new research is correct, then this quake may be the most powerful earthquake ever caused by the type of pumping that regularly happens during oil exploration. Furthermore, it suggests new hazards, including large earthquakes occurring years or decades after the initial start of pumping need to be taken into account as threats associated with exploration. Finally, this work also suggests that only the geology sets upper limits on earthquake size, not anything associated with the pumping; even larger earthquakes could certainly be possible, if the geology was right.

The estimated damage from the 2011 earthquakes likely totaled tens of millions of dollars. The earthquake wasn’t large enough for federal disaster assistance, and many in Oklahoma were without earthquake insurance, since after all, it’s the middle of the continent. If nothing else, I imagine that the authors of this paper have their phones ringing off the hook today from people trying to recoup some of what was damaged in that event.

USGS Shake map image from 2011 event:

USGS Summary of the Oklahoma Earthquake Swarm:

Link to original research paper in Geology (Subscription required):

Summary of study from Bloomberg:

Summary of study from Livescience:

Damage cost estimate:

Previous TES stories on Hydrofracking:


No comments: