Is 7 seconds long enough to react to an earthquake?

If you had only a few seconds to prepare for an earthquake, what would you do? Following Japan, Mexico and India, the US Geological Survey has recently announced its plan to implement an $80million earthquake early warning system for California. The system is designed to detect earthquake signals seconds or even minutes before the earthquake hits, with the aim of mitigating any damage before it occurs. But is a few seconds really enough time to diminish the effects of an earthquake, and can we justify the costs of such a system?

The scheme and how it works

It is important to understand that an ‘early warning system’ is NOT able to predict when an earthquake is going to occur. Instead it’s designed around a series of sensors, which instantly notify users once an earthquake has been detected, giving them several seconds to minutes to prepare before it arrives at their location. These systems work based on two principals: a large network of sensors to intercept earthquake signals between the epicentre and human populations or infrastructure, and the fact that earthquakes transmit their energy in the form of two different types of seismic wave: P-waves (pressure waves) and S-waves (shear waves).

How an early warning system can alert people before any shaking happens; faster P-waves and a large network of sensors. Public domain image by USGS.

How an early warning system can alert people before any shaking happens; faster P-waves and a large network of sensors.
Public domain image by USGS.

As luck would have it, P-waves move faster and with lower amplitude, so they tend to do very little damage compared to the highly destructive S-waves, and can be used as an early indication of the magnitude and origin of the impending damaging waves. P-waves travel at around 5000ms-1 through the Earth’s crust, while S-waves are up to 40% slower; meaning that every 10km away from the epicentre of an earthquake adds roughly 0.8 seconds to the warning. To put that into context, the Nepal Earthquake on the 25th of April 2015 occurred at a depth of 15km, 67km away from the nearest major city, Bharatpur. At this distance, a sensor in the city could have given almost 7 seconds of warning.

Distance is a major factor determining how useful an early detection system can be, but the size and breadth of the sensor array detecting them is arguably more important. Having a large array of sensors spread around the region, especially focused near potential hazard epicentres (in the case of earthquakes these are active faults) can make all the difference. A sensor near the source minimises detection time, and since information travels almost instantly (i.e. the speed of light in most modern communication tools), warnings can be transmitted to more distant locales before even the fastest P-waves have arrived. Taking this into account would extend Bharatpur’s 7 second warning to around 21 seconds.

Information travels much faster than seismic waves, Image used under a creative commons licence from xkcd comics.

Information travels much faster than seismic waves.
Image used under a creative commons licence from xkcd comics.

What can we realistically do in a few seconds?

Although it might not seem like much, a few seconds of warning can have considerable implications for mitigating the effects an earthquake, saving lives and reducing damage. This can be split into two main functions:

  1. Automated response
  2. Warning systems can be linked into many private and public processes, automatically implementing failsafe actions. Medical staff can be alerted to stop delicate procedures, factory lines, industrial systems, power stations, and gas/water mains can be shut down, and emergency response units can be prepared. Such systems are already in place; in Japan and California, trains will automatically slow down or stop to prevent derailing – California’s BART claim that with a 10 second warning, trains can slow from 70mph to 40mph, or from 30mph to stationary.

  3. Notifying the public
  4. Current systems in Japan and India use media outlets to notify the public, and new app-based systems like ShakeAlert automatically sound an alarm on the user’s phone. A few seconds of warning directly to the public can allow them to move to safety, take cover, turn off stoves, and stop vehicles safely.

Some of the automatic response and public notification functions made possible with only a few seconds of warning. Image courtesy of the Japan Meterological Agency

Some of the automatic response and public notification functions made possible with only a few seconds of warning.
Image courtesy of the Japan Meterological Agency.

So the question is, does it work in practice? The simple answer is yes; the Japanese and Beta ShakeAlert systems have reported major successes already; but the reality can be a little more complicated. With earthquake detection, it all depends on location. If, for example, the epicentre is deep in the earth, tens of km from key locations, the system has sufficient time to detect and react to mitigate the damage. On the other hand, if a city is built directly over a fault and the earthquake is shallow (<15km), the system is severely limited and is unlikely to make any significant difference. The limiting factor is the reaction time available after detection.

Other limitations include the need for a dense sensor network, potential false alarms caused by industrial activity, and poor estimates of earthquake magnitude and seismic attenuation (amount of shaking) at different locations.

Even with these limitations, it is clear that with the right strategies and systems in place, the use of early warning systems could considerably reduce the devastation caused by many earthquakes. A few seconds can, it seems, make all the difference.

Further Reading

The Economic Times India working on early warning system on quakes.
Kqed Science California’s Earthquake Early Warning System Is Ready to Get Started.
USGS Earthquake Early Warning
ShakeAlert

Tagged with:

Close
Evolution and significance of fault damage structures on plate boundary faults: the southern San And...


This user has not published any posts

Stay informed

Subscribe to our RSS newsletter by email.


Find Us

University College London is the administrative lead.

Pearson Building, UCL, Gower Street, London, WC1E 6BT

Follow us on Twitter