November 22, 2016
Here’s what earthquake early warning looks like
Posted by Austin Elliott
When a large earthquake struck the coast of Japan near Fukushima Tuesday morning, people all across the country were alerted nearly immediately–most in advance of significant shaking at their location–by the nation’s sophisticated early warning system. Early warning doesn’t predict the onset of an earthquake, but it does predict the shaking level and time of arrival at locations all around the epicenter once one has begun but before its seismic waves have propagated across the land.
As more and more quake-prone countries acquire the technology and instrumentation needed to implement earthquake early warning, the challenge becomes how to effectively deliver warnings to the public. In the video below (and this whole YouTube channel) you can see an array of early warning applications all monitoring the national network and delivering seismic alerts in slightly different ways. First, a moderate earthquake emanates from northwest of Tokyo. Then, at a little after 3 minutes into the video, the large M6.9 strikes, naturally triggering a major alarm from all the programs. Many of them display the intensity of shaking currently being recorded by each seismometer. Some display a predicted maximum seismic intensity distribution based on the size and location of the quake (which is subject to change during this time window, as you see). Others give a warning of the intensity of shaking at the user’s location along with a countdown clock anticipating arrival of the damaging S waves.
Watching seismic waves spread in real time across the country’s dense seismic network is incredibly gratifying as a little glimmer of control over a chaotic and unpredictable force of nature. Other countries such as the U.S. have this kind of preparatory warning to look forward to, but most outside of Japan face significant hurdles in setting the system up.
Three things make Japan’s earthquake early warning system the global leader:
- The abundance of earthquakes: the sheer number of earthquakes that strike Japan on a monthly basis lead to a well-informed and seismically aware populace, allow thorough testing of the algorithms that drive early warning, permit checks of system performance, and increase the number of positive early warning experiences relative to false alarms, increasing public confidence in the system.
- The dense network of seismic monitoring instruments: because of the island nation’s high seismic hazard, Japan has an enormous number of seismometers and GPS stations constantly tracking the movement of the land at an enviable spatial scale. This pretty much guarantees that there are instruments around to formally record ground motion before most people sense it. Seismic networks detecting quakes before population centers are shaken is critical to successful early warning. The sparcity of seismic instruments relative to people is the primary obstacle to most early warning systems, but with Japan’s astounding instrumental network the country has largely surmounted this.
- The nature of subduction zone earthquakes: most of Japan’s earthquakes (though by no means all) occur offshore to the east, where the Pacific ocean crustal plate is diving down beneath the island in subduction. When an earthquake begins out there, it takes a finite amount of time for its ripples to reach land, so most people are spared the most violent shaking, and shaking of most population centers occurs a finite time after the initiation of the earthquake. Thus the nature of subduction zone earthquakes partially mitigates deficiencies in network density, as quakes inherently begin far from any population, beneath the sea. Mexico has had a successful earthquake early warning system for decades despite a relatively sparse seismic network thanks to the nature of its principal seismic hazard being offshore.
Scientists are working hard to find efficient and successful algorithms to implement early warnings when faced with more challenging circumstances than Japan, i.e., fewer monitoring instruments (they’re expensive!) and earthquakes that may be more commonly centered beneath cities, permitting little to no warning time.
Do you think with the advancement in technology, earthquake prediction will soon become a reality? Though there have been a few such prediction models discussed in the last few years, do you think its the next big thing in our field?
Definitely not. What we need to accurately and reliably predict the location and timing of future earthquakes are fundamental advances in our understanding of the active tectonic system, not just advances in the technology used for detection and computation. We’ve learned a HUGE amount about earthquakes in the past century, and so far it largely points to a rather chaotic system that may be inherently unpredictable.
Definitivamente el conocimiento de a la tectónica de placas nos permitirá comprender mejor la causa de estos terremotos. No podemos predecir los eventos del naturaleza, si bien los avances tecnológicos nos permitirán ver como se desencadenan los procesos como licuación de suelos, deslizamientos etc.,