2014 South Napa earthquake GPS strain analysis – Instructor notes
Phil Resor (Wesleyan University) with editing by Vince Cronin (Baylor University and Beth Pratt-Sitaula (UNAVCO)
Summary
The 2014 South Napa earthquake provides an excellent example for studying crustal strain
associated with the earthquake cycle of a strike-slip fault with clear societal relevance. The
largest earthquake in the California Bay Area in 25 years, the South Napa earthquake caused
hundreds of injuries and over $400 million in damages. It was the first large earthquake (Mag 6)
to occur within the Plate Boundary Observatory GPS network since installation. This activity
uses a single triangle of GPS stations (P198, P200, SVIN), located to the west of the earthquake
epicenter, to estimate both the interseismic strain rate and coseismic strain. By the end of the
exercise, the students also have direct evidence that considering the recurrence interval on a
single fault, which is part of a larger system, is not reasonable. The activity is designed as an
extension of the longer Infinitesimal strain analysis using GPS data module, although with some
introduction this activity can be done as a stand-alone.
We envision this exercise being used in one of two ways: 1) As a student-centered activity
associated with the GPS-strain analysis module or 2) As a largely qualitative instructor-led
discussion of the earthquake cycle using GPS time-series and velocity/offset data in map form.
In both cases the suggested outline of the material would be 1) an introduction to the earthquake,
2) a review of the earthquake cycle as measured by GPS geodesy, 3) estimation of interseismic
strain rate and coseismic strain, 4) comparison of these values and a discussion of their
significance for rates of seismicity. The GPS network data are quite rich so that a variety of other
topics could also be addressed. For example, the triangle defined by stations P202-P200-P264
could be used to explore along strike strain loading due to the event.
Context
Audience
This activity was designed for and tested in a structural geology course. However it can also be
successfully used in geophysics or tectonics courses or possibly even a physics course seeking
practical applications.
Skills and concepts that students must have mastered
Ideally students will already have completed at least some component of the GPS-strain analysis
module. Students should be familiar with using Excel or Matlab and should already have had an
introduction to the concept of strain.
How the activity is situated in the course
This activity is designed as single extended discussion or lab exercise with associated homework
and write up.
Goal
Students are able to access and analyze GPS data in order to calculate and interpret interseismic
and coseismic strain in the region between three neighboring GPS stations.
Questions or comments please contact presor AT wesleyan.edu or education AT unavco.org
Version Dec 4, 2014
Page 1
2014 South Napa earthquake GPS strain analysis – Instructor notes
Activity Description
The class period before the lab, make a few introductory comments about the 2014 South Napa
earthquake (the initial couple slides in the pptx can be used for this). Tell the students you are
going to assign them a little web research and reading about the 2014 South Napa earthquake in
preparation for the lab. They will each be able to choose the topic of their brief research from
three options: seismology, faulting and damage. Give a very brief account of the types of damage
one might expect from such an earthquake, followed by a listing of some of the seismic
information that would be available during the lab (epicenter location, depth, focal mechanism,
some sort of fault-slip characterization, etc.), and finishing with a few comments about coseismic
surface faulting. Then pass around a signup sheet with the three topic headings (seismology,
faulting, damage). Tell them that, at the beginning of lab, they would be put into a group
consisting of "experts" on the other two topics, and that their job would be to convey the results
of their research to their (ideally) 3-person group.
At the beginning of the lab give the presentation and ask those with relevant expertise to feel free
to speak-up as they feel that they knew something about a given slide (which they do). At the
end of that presentation if you have the means, spend a few minutes working with a physical
model that uses dry sand and powder, and generates a pull-apart because of a releasing bend in a
right-lateral fault. (The map of ground-surface rupture from the Napa earthquake includes an
apparent right step-over that would constitute a releasing bend.) Then work for a few more
minutes with physical model of a strike-slip fault that includes elastic deformation of the
basement. This leads into the activity itself.
Have the students work in teams but complete the Student Exercise individually. They will need
to refer to the GPS Data Tables and during Step 1, plot the vectors on the Vector Maps
worksheet. Students can input the data from Step 1 into the Strain Calculator individually or in
teams, but it is faster to do it in front for the whole class. These results are used in Steps 2 & 3.
Step 4 gives the students a chance to evaluate whether the graphical method produces reasonable
estimates of translation-vector magnitude and azimuth and sense of rotation (it should).
Step 5 in has the data interpretation, which provides and excellent opportunity for the students to
see that initial model assumptions can be wrong and they need to reevaluate the simple fault
model presented in Figure 1 of the Student Exercise (see Teaching Tips section).
Teaching Notes and Tips
During Step 1 be sure to discuss how to determine the translation vector using the mapped
information. After that, work on visualizing whether the triangles rotated in a clockwise or
anticlockwise direction. Finally, ask the students whether the rotations indicate a left- or rightlateral fault. It can take a bit of drawing on the board to get them to understand how a rotation
might relate to a sense of slip on a planar fault.
The interseismic deformation involves a clockwise rotation in the crust west of the northnortheast striking right-lateral strike-slip fault. The coseismic deformation caused an anticlockwise rotation. This is an interesting feature of the data from this event, and one that should
not be overlooked.
Questions or comments please presor AT wesleyan.edu or education AT unavco.org
Version Dec 4, 2014
Page 2
2014 South Napa earthquake GPS strain analysis – Instructor notes
The results from the Strain Calculator in Steps 2 & 3 yield an improbably recurrence earthquake
interval which is a great time to discuss how simple fault models may sometimes need to be
reevaluated. Using the interseismic velocity data, the Strain Calculator yields a maximum shear
strain rate of 300.0810 nano-strain/year. Using the coseismic displacement data, the Strain
Calculator yields a maximum shear strain of 1050.8180 nano-strain. Following the suggestion in
the worksheet, students divide the coseismic by the interseismic maximum shear strains to get
3.5 years as a model-based recurrence interval. Not feasible!
To resolve this contradiction, help the students to recognize that the West Napa fault was
between two active faults with higher average slip rates and note that the GPS site triangle
crossed the Rogers Creek fault. With this help, students understand that the presence of these
faults introduced conditions that were different than the original model (Student Exercise Figure
1), which features continuous elastic half-spaces up to the fault surface on both sides. However,
within the confines of a 2-hour lab, it is not clear how many understood exactly how that
difference was reflected in a calculated recurrence interval of 3.5 years on a fault that had not
had a similar-sized earthquake in more than a century.
Resources
Related files:




Exercise: Finding location and velocity data for PBO GPS stations – optional precursor
exercise for learning how to download PBO data
Strain calculator – two versions of the calculator are available: Excel and Matlab. The
Excel version is more “black box” whereas the Matlab version would be more
appropriate if instructor wants students more involved in the mathematics and
programing end.
Document: Explanation of calculator output – runs through the different output
parameters with explanations and images to aid in real world understanding of results.
Other related files can be found on the UNAVCO module website.
Selected Reference:






California Geological Survey, 2013, California Historical Earthquake Online Database,
accessed 10/14/2014,
http://redirect.conservation.ca.gov/cgs/rghm/quakes/historical/Index.htm
Nevada Geodetic Laboratory, 2014, Preliminary Coeseismic Offsets for American
Canyon Earthquake, Northern California Bay Area, accessed 10/14/2014,
http://geodesy.unr.edu/billhammond/earthquakes/nc72282711/nc72282711.html
Plate Boundary Observatory GPS data http://pbo.unavco.org/
San Francisco Bay Network GPS data
http://earthquake.usgs.gov/monitoring/gps/SFBayArea/
UNAVCO, 2014, Data Event Response to the 24 August 2014 Mw 6.0 South Napa
Earthquake - 6km NW of American Canyon, California, accessed 10/14/2014
http://www.unavco.org/highlights/2014/south-napa.html
U.S. Geological Survey and California Geological Survey, 2014, Quaternary fault and
fold database for the United States, accessed 10/14/2014, from USGS web site:
http://earthquake.usgs.gov/hazards/qfaults/
Questions or comments please presor AT wesleyan.edu or education AT unavco.org
Version Dec 4, 2014
Page 3
2014 South Napa earthquake GPS strain analysis – Instructor notes
Questions or comments please presor AT wesleyan.edu or education AT unavco.org
Version Dec 4, 2014
Page 4