Now that the USGS has preliminary come up with a way to quantify the short term hazard of potentially induced earthquakes it’s important to know what these maps are and mean.
Let’s start with what they are not.
They are NOT A PREDICTION. Earthquakes cannot be predicted at present and chances are this will continue to be true, forever. Anyone who says they can predict earthquakes is:
1. A liar
2. A kook
3. Both a liar and a kook
4. Mistaken, and stuck in a post hoc ergo propter hoc loop. No non-kook/liar method has been shown to be remotely reliable.
At best, we can forecast. There are quite a few forecasting techniques that have been developed that each have their pluses and minuses, their possible successes and many failures.
They are also NOT FAULT MAPS. Instead, maps of known active faults and known active seismic sources are used to build the maps.
Instead, they are a an estimate of the probability of exceeding a certain amount of ground motion in 50 years.
2014 USGS Seismic Hazard Map, 2% probability
The map displays where there is a probability of damaging shaking in the US over a certain time period. The maps are used to design infrastructure along with homes and businesses, along with critical facilities that can withstand high levels of shaking caused by earthquakes. They’re used to build building codes around the US, and to plan for hazards such as earthquakes. In the West, especially in California, maps are time-dependent, while elsewhere they’re time-independent. Ground motion is shown as a unit of gravity. Spectral Acceleration
is how buildings respond to ground shaking.
The USGS has an article here.
So, how are these maps created? This is done on a six year cycle. The updated maps were released earlier this year.
First and foremost, it uses written historical records. Generally speaking if a large earthquake has happened in a location in the past, it will happen again at some point in the future. Smaller historical earthquakes can illuminate seismic zones and they’re added in, along with historical maximum seismic intensities. Finally geology, paleoseismology, and other data are added in to build the maps. You can read the technical documentation and download maps here.
A big take away is The Past Informs the Future.
Charleston SC Logic Tree--or how the earthquake hazard for Charleston, South Carolina, is quantified. From Petersen, M.D., Moschetti, M.P., Powers, P.M., Mueller, C.S., Haller, K.M., Frankel, A.D., Zeng, Yuehua, Rezaeian, Sanaz, Harmsen, S.C., Boyd, O.S., Field, Ned, Chen, Rui, Rukstales, K.S., Luco, Nico, Wheeler, R.L., Williams, R.A., and Olsen, A.H., 2014, Documentation for the 2014 update of the United States national seismic hazard maps: U.S. Geological Survey Open-File Report 2014–1091, 243 p., http://dx.doi.org/...
Let’s look at a real world example. I’m going to use New York City.
New York City and its surrounding tri-state area is in a region of elevated risk due to two known historical M5+ quakes beneath the city—one in 1737 and one in 1884. The 1884 quake likely had an epicenter under or just off shore of Coney Island. In addition, the region is densely populated and has a great deal of vulnerable and critical infrastructure.
Faults do exist in and around the city, including one that slashes across Manhattan in thevicinity of 125th Street (PDF), but it is not conclusively known if they are capable of hosting large earthquakes (a 2001 quake—in January, may have slipped along the 125th St. Fault). Epicenters in the eastern US and for the most part the Central US can’t be linked conclusively to the known faults in the area.
2% probability/50 years. From 2015, Seismic-hazard maps for the conterminous United States, 2014: U.S. Geological Survey Scientific Investigations Map 3325, 6 sheets, scale 1: 7,000,000, http://dx.doi.org/...
What this map is saying is the 2% probability of exceedence in 50 years for peak ground acceleration is between 14 and 20% of gravity. What does this mean? That over a 50 year time span there is a 2% probability that an earthquake capable of creating strong to very strong shaking and light to moderate damage (or up to about M5.9) could occur. Because the region is packed with unreinforced masonry and the region’s seismic building codes are new this could be a very nasty potential disaster. The risk is lower at 10% probability, but still not non-existent. Thus, the city plans for earthquakes (PDF)
What does this mean for induced earthquakes? Well, a lot.
The USGS (and basically, the whole world) until this year had no way to quantify the hazard from induced earthquakes, so they’ve begun to develop one
Let’s look at it.
First, the USGS identified a number of regions where potentially induced seismicity has occurred since the 1960s. All but four of them (three in Colorado, one in northeast Ohio) are oil and gas related.
Next the same process that built long-range hazard maps were used except a far historical period was used. Using different years (2011 versus 2012, and 2012 versus 2013, and 2013 versus 2014) yielded different results as different induced swarms became active and subsided as their likely cause (injection wells) were either plugged or injection greatly cut back.
Oklahoma clearly sticks out as the most active and hazardous area.
Close up of Oklahoma. For 3A and 3B Red = 2014, Green = 2013 Blue = 2012 and previous.
From Incorporating Induced Seismicity in the 2014 United States National Seismic Hazard Model—Results of 2014 Workshop and Sensitivity Studies http://pubs.usgs.gov/...
Seismicity in the rest of the eastern US is confined to small geographic areas and has subsided. But Oklahoma’s covers a vast part of the central section of the state, and it extends into southern Kansas.
These are the preliminary maps.
from "Incorporating Induced Seismicity in the 2014 United States National Seismic Hazard Model—Results of 2014 Workshop and Sensitivity Studies", combined with 2014 Hazard Map. Spectral Acceleration is how buildings perform in an earthquake.
What we see here is a short term hazard map. These maps are saying that over the next year, based on the last few of years of seismicity, a large area of Oklahoma has a 1% (and an even larger area has a 0.04%) probability of experiencing damaging shaking. This is a pretty high hazard even if in a given year the probability is low.
Why 1 year and not 50? It is because the patterns and rates of induced earthquakes depend on both the economy and public policy decisions. A 1 year hazard map may be the best possible, but keep in mind this is very much still a work in progress. These maps are used to develop building codes so I’m not sure what the use of a 1 year hazard map would be for engineers, but I can see a clear societal benefit. I highly recommend digging through the documentation, if anything, for the works cited page.
I’d also be remiss if I didn’t point out that this type of hazard quantification is not without its own uncertainties and significant problems, but you can go here to read about them and explore at your own leisure. I'll tackle it in another diary soon.