Earthquake early warning system: Is this what it could look like in NZ?

Scientists now have a better idea of what a national earthquake early warning system could look like in New Zealand after using clever technology that’s already been tested on some of our latest big shakes. Photo / NZTA Waka Kotahi

It could give Kiwis a few life-saving seconds to prepare before a major earthquake hits.

Scientists have been trialling clever technology that could pave the way to a national earthquake early warning system (EEW) - so far testing it on some of our biggest recent shakes.

These systems work by detecting an earthquake’s fastest-travelling “P-waves”, before the more damaging S-waves arrive - moments later.

New Zealand does currently have some early warning systems: a Google-pioneered system that pings messages to Android phones before shaking starts – and local company Jenlogix operates a network used by several councils, ports and power companies.

Yet unlike Japan and many other countries where earthquakes are common, New Zealand doesn’t have an earthquake early-warning network run and funded by the Government.

Such networks typically take tens of millions dollars to set up and operate.

In what’s something of a first step toward one, New Zealand scientists working under the Rapid Characterisation of Earthquake and Tsunami Programme have begun operating a “finite fault rupture tool”, or FinDer – designed to rapidly estimate the extent and location of an earthquake

It draws on real-time ground motion data gathered by seismic sensor networks - and already operates in the US and Central America.

In New Zealand, where GeoNet is trialling more than 500 sensors, FinDer continuously scans data for patterns that could indicate earthquakes.

If an event is detected, the pattern is turned into a 2D map image and then compared against a catalogue of pre-computed scenarios.

“This is done continuously as the rupture evolves, giving us a series of snapshots of the rupture as it grows over seconds to minutes after its nucleation,” GNS Science seismologist Dr Jen Andrews explained.

“This means we get information on the earthquake’s extent, and also sometimes the direction a fault ruptured, both of which are critical for rapidly understanding the risks and impacts of the largest events.”

While FinDer hasn’t yet been tested in real-time by a major New Zealand quake, scientists have found it performed well in capturing smaller ones, including recent magnitude 6.0 events near Paraparaumu, Porangahau and Geraldine.

When a magnitude 5.2 quake hit Porangahau last month, FinDer yielded reliable magnitude and location information within 14 seconds.

They’ve also simulated it on some of our biggest shakes in recent times, including the 2010-11 Canterbury sequence and 2016′s magnitude 7.8 Kaikōura event.

New research suggests for all the clever science behind earthquake early warning (EEW) systems, Kiwis know little about how to use them.

“Through real-time testing, we’ve been exploring the behaviour of the method in terms of reliability, speed and accuracy,” Andrews said.

“We now better understand how well the method works for New Zealand’s seismic network and our unique and complex tectonic environment.”

The tech had some drawbacks: it wasn’t designed to deliver high-accuracy solutions, couldn’t determine robust details about which exact faults were involved and didn’t perform well for offshore events – limiting its usefulness for estimating tsunami hazard levels.

Still, Andrews said having it running in New Zealand would have benefits for getting rapid information during an earthquake response – while enabling scientists to explore EEW methods.

“We can now assess how close or far our systems are from being able to deliver reliable, accurate and timely warnings, and begin to understand the cost-benefit of implementing a national New Zealand system.”

The tech, described in a study just published in the Bulletin of the Seismological Society of America, comes after researchers recently found Kiwis know little about EEW systems.

Focusing on Google’s Android-based system, a team from Massey University’s Joint Centre for Disaster Research surveyed people to assess how much they knew about it and how they responded to it.

While most found it useful, there was also low awareness about it – and most people who received alerts didn’t immediately take the right action of dropping, covering and holding.

Only a quarter of the respondents also knew the alert came from Google, with others believing it’d been sent by the National Emergency Management Agency or GeoNet.

Whether earthquakes could be forecast, meanwhile, was something yet to be realised.

One major recent study identified unique seismic signals eight months ahead of Turkey’s devastating 2023 earthquake – but its authors pointed out that forecasting in the mid to short-term nonetheless remained challenging, if not impossible.

GNS hazard and risk scientist Dr Annemarie Christophersen said scientists here had a good understanding of where and how often quakes occurred in the long term, with a 20-year forecast model built into the recently-updated New Zealand National Seismic Hazard model.

“The best way to prepare for future earthquakes is to build well with the understanding of the latest science.”

Jamie Morton is a specialist in science and environmental reporting. He joined the Herald in 2011 and writes about everything from conservation and climate change to natural hazards and new technology.