Locating earthquakes in the Yellowstone region

Release Date:

Ever wonder how seismologists determine the location of an earthquake in Yellowstone?  It’s an intricate process, but thanks to experienced scientists, up to thousands of earthquakes are located in the Yellowstone region every year!

Yellowstone Caldera Chronicles is a weekly column written by scientists and collaborators of the Yellowstone Volcano Observatory. This week's contribution is from Jamie Farrell, assistant research professor with the University of Utah Seismograph Stations and Chief Seismologist of the Yellowstone Volcano Observatory.

Seismograms showing M4.8 earthquake near Norris Geyser Basin in Yellowstone on March 30, 201

Record section showing horizontal component seismograms from stations in the Yellowstone region from the M4.8 earthquake that occurred near Norris Geyser Basin on March 30, 2014.  The vertical blue dashed line represents the origin time of the earthquake at 12:34:39.16 UTC.  The red line represents the P-wave arrival with a velocity of ~5.7 km/s.  The green line represents the S-wave arrival with a velocity of ~3.2 km/s.

(Figure by Jamie Farrell, University of Utah.)

Over the past decade, The University of Utah Seismograph Stations (UUSS) has located an average of ~1,570 earthquakes in the Yellowstone region every year—that’s a rate of several earthquakes per day!  The most in any given year during that time was 3,427 in 2017 and the least was 708 in 2011.  Seismicity rates naturally fluctuate, but most of the variation in Yellowstone depends on the number and size of earthquake swarms that occur during the year.  For example, in 2017, over 2,400 of the earthquakes located that year were part of a single swarm—the Maple Creek earthquake swarm!

When an earthquake happens and you see wiggles on a seismogram, the first question to answer is “where is the earthquake located?”  This is important for a number of reasons.  First, the location can tell what fault caused the earthquake.  Second, and potentially most important, in the case of a large and possibly damaging earthquake, its location can be used to help direct first responders.

So how are earthquakes located in the Yellowstone region?  Before we talk about that, let’s get some terminology straight.  Sometimes people talk about the epicenter or the hypocenter of an earthquake, and these two terms may be confusing.  The hypocenter is the point in the earth where the earthquake occurs and includes the longitude, latitude, and the depth (X,Y,Z).  The epicenter is the point on the surface of the earth directly above the hypocenter and only includes the longitude and latitude (X,Y).

Seismic record of Yellowstone station YHB for the M4.8 earthquake of March 30, 2014

3-component seismograms from station YHB for the M4.8 earthquake that occurred near Norris Geyser Basin on March 30, 2014, and showing the P-wave arrival pick (red) and the S-wave arrival pick (green) as determined by UUSS analysts.  The vertical blue dashed line represents the origin time of the earthquake at 12:34:39.16 UTC.

(Figure by Jamie Farrell, University of Utah.)

When an earthquake ruptures a fault at depth, it generates seismic waves that radiate out in all directions from the hypocenter.  Depending on the type of wave and the characteristics of the rock they are travelling through, the waves travel at different speeds.  These waves are then recorded on the network of seismometers in the Yellowstone region.  The two waves that are most important for locating an earthquake are the primary wave (P wave) and the shear wave (S wave).  On average, the P-wave velocity is ~5-8 km/s and the S-wave velocity is ~3-4.5 km/s through solid rock.  Due to the different speeds, the P wave always arrives first, but because of the character of the waves, the later-arriving S wave is stronger.

The seismic data are all radioed back to the UUSS in real-time and are recorded and analyzed on sophisticated earthquake computer processing software that automatically picks the arrival times of the P waves at as many seismic stations as possible, and then calculates a location and magnitude for the earthquake.  Depending on how large the earthquake is and its location, the system then sends out alarms to various UUSS staff, including the “duty seismologist”—a rotating position occupied by a seismologist who responds to earthquake alarms 24 hours a day, 7 days a week.  The duty seismologist reviews the earthquake to make sure it is a real local earthquake and not a mislocated event from outside the region, or a sonic boom, or some other event, and then assesses the computer-determined location and magnitude for accuracy.  The computer calculations of location and magnitude are usually close, but always require some refinement by a trained analyst (such as picking S waves to get better depth estimates)—this is why the magnitude and exact location of the earthquake sometimes change after the initial report.  With more data that are carefully analyzed by an experienced seismologist, more accurate determinations of magnitude and location are possible!

Seismic stations used to located the March 30, 2014, M4.8 Norris earthquake in Yellowstone

Station map showing seismograph stations used in the location of the M4.8 earthquake that occurred near Norris Geyser Basin on March 30, 2014.  The yellow star shows the earthquake epicenter.  Red triangles represent seismograph stations with a P-wave arrival pick.  Green triangles represent seismograph stations with both a P-wave and a S-wave arrival pick.  Cyan triangles represent seismograph stations that were not used in the location.

(Figure by Jamie Farrell, University of Utah.)

Determining the hypocenter location is more complicated than just picking the arrival times of the P- and S-waves on multiple stations.  Once those arrival times are picked, UUSS uses a seismic velocity model of the Yellowstone region to simulate the propagation of seismic waves given an initial hypocenter.  The velocity model describes the speed that seismic waves travel through the earth at a given depth.  A computer algorithm then compares the arrival times at each station from the simulated run versus the actual arrival times picked by the analysts.  The differences between the two are called the “misfit.”  The computer program then picks a different hypocenter and does the simulation again and calculates a new misfit.  It does this over and over until it finds the smallest misfit between the simulated and actual data.  This becomes the final location of the earthquake.  It can take up to several hours to zero in on the earthquake source, depending on the size of the earthquake and the number of stations that record the shaking. 

In general, to solve for the hypocenter, you need to observe the earthquake on a minimum of 3 seismic stations.  If you want to also calculate the time that the earthquake occurred, you need a minimum of 4 observations.  However, to get the best possible hypocenter solution, you would ideally have many stations that are situated all around the earthquake and at least one station close to the event.  The more stations that record the earthquake, the more accurate the location!  Conversely, if an earthquake is only recorded on one or two stations, it cannot be located.  This is why seismic reports tend to say the “number of earthquakes located” and not the “number of earthquakes that occurred” when giving earthquake statistics in Yellowstone for a given day, month, or year.

Of course, this procedure only covers determining the earthquake location.  But what about the magnitude?  For that story, check out a previous Caldera Chronicles article on the topic at https://www.usgs.gov/center-news/how-big-was-earthquake.

Determining the location and size of an earthquake requires a state-of-the-art seismic network and a team of trained analysts who are available to examine data 24/7.  Yellowstone is fortunate to have both, thanks to the seismologists at the University of Utah Seismograph Stations!  For the latest information on earthquakes in the Yellowstone region, visit the UUSS website at https://quake.utah.edu.

Related Content

Filter Total Items: 5
Date published: June 21, 2021

What causes earthquake swarms at Yellowstone?

Earthquake swarms are common at Yellowstone, but why do they occur?  Are they driven by magma migration?  Water?  Steady creep along faults?  All three are possibilities, and tracking the style of the earthquakes can reveal the causes.

Date published: November 16, 2020

How Big Was That Earthquake?

Earthquake magnitudes may seem straightforward, but a lot goes into their calculation, and multiple methods can be used.  These methods must account for many complications, like the impacts of local geology on the amplitude of shaking.  In Yellowstone, a modern seismic network and experienced analysts make it possible to determine the magnitudes of even the smallest earthquakes!

Date published: July 13, 2020

What Goes into Operating the Yellowstone Seismic Network?

Have you ever wondered just what went in to going from ground shaking detected by a seismometer to a located earthquake available for viewing online?  In this week’s Yellowstone Caldera Chronicles, we trace the path from the ground in Yellowstone to your web browser!

Date published: January 20, 2020

Earthquakes in and around Yellowstone: How often do they occur?

The U.S. Geological Survey produces seismic hazard maps for the United States and the 2018 nationwide long-term assessment shows that the Yellowstone region has some of the highest seismic hazard values in the Intermountain West.

Date published: December 10, 2018

A closer look at the 2017 Maple Creek earthquake swarm

In June of 2017, an earthquake swarm began beneath the western edge of Yellowstone National Park, just east of Hebgen Lake. This swarm proved to be one of the more persistent swarms observed in Yellowstone, with the main episode lasting more than 3 months and producing thousands of recorded earthquakes.