Allen CV
Seismo Lab
Earth & Planetary
UC Berkeley

Earthquake Early Warning: ShakeAlert

Basics | ShakeAlert | Timeline | Japan | Mexico | Key Publications

E-larmS | G-larmS | T-larmS | MyShake

Potential ShakeAlert warning times. From Allen “Seconds count”, Nature 2013.

What is ShakeAlert?

ShakeAlert in the name of the public earthquake early warning system now being implemented in California, Oregon and Washington. ShakeAlert has been running as a demonstration system since 2012 issuing alerts to test users. In 2013, California passed legislation to implement a public warning system in the state. In 2014, the US Congress started to fund the transition of the demonstration system to a full public system that could providing warnings for everyone.

ShakeAlert is a collaboration between the US Geological Survey, UC Berkeley, Caltech and the University of Washington. The system uses the seismic and geodetic networks operated along the US west coast and feeds the data into several algorithms that then detect earthquakes and generate alert messages. Alert messages are aggregated and a single alert feed is provided to users. ElarmS is one of the core ShakeAlert algorithms in California, Oregon and Washington and is typically the source of the first alert message. The necessary funding to make this a full public system has not yet been provided so only authorized test users have access.

USGS Factsheet: ShakeAlert - An Earthquake Early Warning System for the United States West Coast, 2014.

What are E-larmS, G-larmS and T-larmS?

The real-time seismology group at UC Berkeley are researching the science behind earthquake and tsunami alerting systems. Using data from past earthquakes around the world we can better understand the physics of the earthquake process, and design real-time algorithms to automatically detect and characterize earthquakes underway. We can then generate shaking alerts and local tsunami alerts. Our research started in 2001, and our methodologies are now in use around the world: Chile, Israel, Korea, Turkey and the US west coast.

ElarmS uses the first few seconds of data from an earthquake to locate the epicenter and estimate the magnitude, allowing the shaking intensity to be predicted and an alert issued. It can be applied in earthquake prone regions that have regional seismic networks. It is possible to provide alerts before any strong shaking; the speed of the alerts depends on how dense and fast the seismic network is. In California alerts are generated in as little as 3 seconds after the earthquake initiates and feed into the ShakeAlert system. More about ElarmS

G-larmS is designed to provide additional hazard information for the largest earthquakes. It uses real-time geodetic data from regional networks close to the earthquake to better-characterize the shaking distribution for earthquakes greater than magnitude 7. It does this by estimating the area and location of the fault as it breaks. It can be used as a complement to ElarmS in subduction and strike-slip (San Andreas-type) regions, and can generate alerts in about 12 seconds.

T-larmS in the natural extension of earthquake early warning: Tsunami warning. The rapid characterization of an earthquake in a subduction zone can be used to calculate the tsunami wave generation, and then propagate the wave to the nearby coastline. This information could be available in about 3 minutes, meaning it could be available before the tsunami reaches the nearby coast.


Support for the earthquake early warning efforts at UC Berkeley is provided by: