Science Help with Science Connections Snapshots Solid Earth Cryosphere Environmental & Hydrogeodesy Ocean Atmosphere Human Dimensions Technology

Integrating Geodetic Data For Earthquake Warnings in the Western U.S.

Researchers: J. R. Murray, J. O. Langbein, and S. E. Minson, U.S. Geological Survey, B. W. Crowell and D. A. Schmidt, University of Washington, R. Grapenthin, New Mexico Tech, K. Hodgkinson, UNAVCO, T. Melbourne, Central Washington University, and D. Melgar, University of Oregon

Written by Linda Rowan
3 January 2019


Real-time Global Navigation Satellite System (GNSS) ground-based sites can measure displacements on the Earth’s surface and these geodetic data have the potential to improve earthquake warnings. Here, the benefits of including GNSS in a warning system are outlined and three different algorithms for integrating geodetic data into the USGS’s ShakeAlert system are identified. Independent development of the algorithms has shown they can improve warnings, a systematic and rigorous testing of the algorithms is now required to quantify that improvement.


The United States is developing an earthquake early warning system called ShakeAlert along the West Coast. The U.S. Geological Survey is developing the system and is working with many universities as well as UNAVCO to build, operate and maintain the system. A prototype system uses the first few seconds of seismic waveforms recorded by seismic instrumentation to determine the earthquake’s location, magnitude and origin time. Recent large earthquakes, e.g., the 2011 Tohoku earthquake and the 2010 El Mayor Cucapah earthquake, have shown that relying on the first seconds of the waveform can lead to an underestimation of the earthquake magnitude. The growth of continuous, high-rate GNSS networks around the world in the past decade has provided data that shows GNSS can measure large displacements and be used to rapidly determine an accurate magnitude and other earthquake characteristics, even before the fault has finished rupturing.

On the West Coast of the U.S., large magnitude earthquakes (>M7) are likely and a Tohoku-like earthquake and tsunami is possible along the Cascadia Subduction Zone that slices beneath Northern California, the Pacific Northwest and parts of Canada. Adding geodetic data for large events to any warning system would help to mitigate damage as the information it provides would yield more accurate estimates of impending ground shaking. UNAVCO operates and maintains ~640 real-time GNSS sites on the West Coast of the U.S. The U.S. Geological Survey is using these data to develop methods for integrating the geodetic data into ShakeAlert.


Three algorithms have been identified for testing to integrate geodetic data into ShakeAlert, BEFORES, G-FAST and G-larmS. All three will be tested with synthetic and with real historical geodetic data. The tests will involve the integration of the geodetic algorithms into the Earthworm software and use the ActiveMQ messaging system that has already been developed for the seismic-based warning system. The tests will be undertaken at the University of Washington, Seattle, University of California, Berkeley, the USGS at Menlo Park and the California Institute of Technology. The results will identify strengths and weaknesses of each algorithm and will attempt to quantity the improvement in warnings if a geodetic component is included in an early warning system. In the future, ShakeAlert can be used in Alaska and anywhere else that seismic risk is significant. In addition, the geodetic data could help to provide tsunami warnings for events related to earthquakes, landslides or volcanic eruptions through testing and integration of the appropriate algorithms.

Related Links

Related Data Links


Murray et al., Development of a Geodetic Component for the U.S. West Coast Earthquake Early Warning System, Seismological Research Letters, 2018, doi:doi:10.1785/0220180162.


earthquake early warning, GNSS

Map Center
Sacramento, California

Send questions or comments about this page to


Last modified: 2020-02-08  01:11:07  America/Denver