At fault

So, who remembers that earthquake back in December? You know, the one in Ohio? The fact that it happened in Ohio is what makes this earthquake so memorable. A magnitude 4.0 on the Richter Scale, while definitely noticeable, is not the thing that widespread news coverage is commonly made of. A 4.0 on the Richter Scale in a place where one does not expect there to be an earthquake, however, does tend to be newsworthy. And Ohio is really one of those places you don’t expect earthquakes. In one of his many books on geology, writer John McPhee once quipped that in the United States, “politics and plate tectonics tend to be at their most stable in the Mid-West”. He does have a point. Ohio is in the middle of the North American craton, a solid chunk of tectonic plate that has been cohesive for a little over 1 billion years. 

    For those of you with a limited geology background, that solidity is an important part of the story, and underlies the surprising element of an earthquake in Ohio. Technically, an earthquake is a vibration of the Earth’s surface following an event that releases energy into the crust. So, the shockwave that follows a bomb going off at ground level, or the shaking surrounding a volcanic eruption are, by definition, earthquakes. Usually, however, an earthquake is caused by movement of the tectonic plates at fault lines. Plates can move in several ways, with each form of movement causing earthquakes in its own special way. In each case, plate movement somehow puts strain on the crust, strain that is released in the form of an earthquake.

Areas where new crust is forming and pushing two plates apart (spreading centers) are prone to small, shallow quakes - the newly formed crust is so weak that it can tolerate very little strain. Areas where one plate is being dragged under the other are prone to earthquakes at all depths in the crust. The process of one plate being dragged beneath another is neither smooth nor linear - periodically the plate going down with get jammed, and eventually break loose to sink further, accompanied by an earthquake. Areas where mountains are forming are also earthquake prone, unsurprisingly. When two plates are colliding and deforming to push huge amounts of crust into the air, there are going to be times when the collision process jams, strain builds up, and eventually the jam breaks, the mountains shoot up (metaphorically), and a huge amount of energy is expended in earthquakes. The catastrophic earthquakes of Pakistan, Iran and Turkey all owed their origins to this mountain building process. Then there’s the quake-generator most familiar to Americans, two plates sliding past each other. Again, “sliding” isn’t quite the right word. More like “two plates trying to move past each other, getting jammed, breaking free with an earthquake energy release, and getting jammed again”. The famous (infamous?) San Andreas fault system of California is such a place. Ok, so that’s earthquakes 101. Ohio, however, is quite clearly none of those things. It’s not even near any of those things.

Compare this map of earthquake locations worldwide to the following map of the tectonic plates. Now, try to find Ohio.

 

This is where we get into the exciting and ill-understood world of intraplate earthquakes. These are rare events, comprising maybe 0.5 percent of all earthquakes in a given year. Still, they can be dangerous. An intraplate earthquake in Gujarat, India in 2001 killed nearly 20,000 people, while a series of intraplate earthquakes in New Madrid, MO rank as some of the strongest earthquakes in the recorded history of the U.S. Intraplate quakes tend to be more destructive than the more common plate-boundary quakes, precisely because of their position in the midst of a plate. Cracked, broken rock as you would expect to find in an area prone to earthquakes does not transmit seismic waves very well - an earthquake occurring along a fault line, while potentially strong and dangerous, is unlikely to shake more than the immediate area. The solid, unbroken rock characteristic of the middle of a plate, however, is an excellent conductor of seismic energy. An 1886 earthquake epicentered in Charleston, SC rang church bells 900 miles away in Boston- at 7.6 on the Richter scale, the quake was strong, but a comparable earthquake in a fault zone in Mexico was felt only weakly 600 miles away. The recent magnitude 5.8 earthquake in the Washington, DC area was weak enough to merit laughs from more earthquake-prone regions of the United States, but was felt over 500 miles away in Maine. That would be like the recent magnitude 5.6 earthquake in northern California being felt as far away as Los Angeles. As it happened, the quake was barely felt 200 miles away in Sacramento. Intraplate quakes are hard to miss, but also hard to understand.

At this point, there are several theories out there to explain intraplate quakes. All three have one principal in common - while the centers of plates are certainly stable, they are neither uniform, nor one hundred percent solid. Pressures at the edges of plates can cause warping and deformation hundreds of miles away from the active faults, while a geologic phenomena called hot spots can cause volcanism and even rifting deep into a plate. We might get into hot spots in the future, as they’re very cool. And responsible for some really wacky geology out West. Anyway, crustal plates can get all bent out of shape far from a plate boundary. In fact, plates periodically rift apart, or begin to rift apart, but stop. The resulting areas of weakness from a so-called failed rift, hot spot volcanism, or past deformation from an episode of mountain building can all react to stress emanating from a plate boundary. While the North American craton is generally stable, it is crisscrossed with zones of weakness, and actively colliding with/sliding past other plates. The stress from those plate interactions is periodically relieved by earthquakes in the lower Mississippi Valley (site of a failed rift), the South Carolina coast (deformed by rifting 100+ million years ago), or a host of other places (including, worryingly, a rift running right across Manhattan Island).

So that’s what happened in Ohio, right? Not quite. Remember back when I said that the rare human activity can cause an earthquake? Well, it looks like that’s what happened in Ohio, but it wasn’t a bomb this time. No, for this we can credit disposal of waste fluid from a natural gas hydraulic fracturing (aka fracking) operation. The timing, location and depth of the quake was closely correlated with fluid disposal via injection. Additionally, research out of Oklahoma indicates that injecting fluid into a faulted or otherwise weakened area of rock that is under stress (i.e. any potential intraplate earthquake zone) can cause an earthquake by lubricating the rock, and thus reducing the friction keeping the rock stable. Earthquakes happen when the strain on an existing zone of rock weakness becomes stronger than the forces keeping the zone immobile (typically friction). By reducing the stabilizing force, introduction of fluid to an earthquake zone can cause an earthquake. Fluid injection doesn’t increase strain, but it does diminish friction, which is kind of the same thing.

The good news is that hydrofracking induced quakes tend to be shallow and weak, but they are still earthquakes. Intraplate quakes are messy and hard enough to understand or predict, without introducing further complications. As in all things in science, more research is required, but we might want to take a closer look at the geologic maps for a given area before doing, well, anything with it.