A warning before the shake: New system gives communities crucial seconds

In a groundbreaking step toward earthquake preparedness, a team of researchers from Te Kunenga ki Pūrehuroa Massey University has successfully tested a decentralised, low-cost Earthquake Early Warning System (EEWS) in Wellington. Led by Dr Chanthujan Chandrakumar during his doctoral studies at the CRISiSLab within the Joint Centre for Disaster Research, this pioneering project showcases the potential of community engagement and affordable technology in disaster risk reduction.

A first for New Zealand and the world

Aotearoa New Zealand is known as one of the most seismically active countries on the planet. However, to Dr Chandrakumar’s surprise, there is no official early warning system currently in place.

Dr Chandrakumar explains that his initiative represents a significant milestone: the first real-time trial of a community-based EEWS in New Zealand and the first global demonstration of a decentralised, community-engaged EEWS.

“Unlike traditional national systems that rely on expensive equipment and centralised data processing, this innovative system utilises low-cost Micro-Electro-Mechanical Systems (MEMS) sensors hosted by local volunteers.

“Each sensor independently detects seismic activity and issues alerts, eliminating the delay of central processing and allowing for faster local responses, which could ultimately reduce devastation and potentially save lives,” he explains.

How the system works

At the heart of the system are two specialised algorithms developed for the New Zealand context:

1) NZ-PLUM: An adaptation of Japan’s PLUM algorithm tailored to New Zealand's Modified Mercalli Intensity (MMI) scale. This focuses on detecting damaging Secondary (S) waves - slower, high-amplitude seismic waves that cause the most shaking and damage during an earthquake.

2) NZ-PLUM-P: A complementary method that uses data from Primary (P) waves - the seismic waves that arrive before the more damaging S-waves and typically cause minimal shaking. To predict the intensity of the incoming S-waves, enabling alerts would be issued earlier, increasing lead time for protective action.

By combining these algorithms, the system enhances its responsiveness and accuracy, offering a dual-layered approach to early warning.

Promising results from real-world testing

The system was tested in Wellington over a three-month trial period, producing some very promising outcomes. It accurately detected all significant earthquakes during the study, with NZ-PLUM-P consistently providing faster alerts than NZ-PLUM. While a few false alerts were recorded, mainly due to the early-detection nature of NZ-PLUM-P, they remained within acceptable limits, reinforcing the system’s potential for operational use.

Dr Chandrakumar says that the study also highlighted the value of a decentralised alerting model, demonstrating how locally-generated alerts can match the performance of centralised systems, while reducing infrastructure and operational costs.

Community at the core

Dr Chandrakumar notes that one of the most distinctive aspects of this research is its community-centric design and the strong community interest it has gained.

“Local volunteers not only hosted the sensors, but also played a vital role in maintaining the system. This participatory model fosters trust, awareness and preparedness at the community level, key elements for resilience in disaster-prone regions,” he says.

Why it matters

For many regions around the world, especially those with limited resources, implementing a national EEWS is prohibitively expensive. This study offers a compelling alternative: a low-cost, scalable and community-driven model that delivers timely and accurate earthquake alerts.

“It’s a solution tailored not just to New Zealand but to vulnerable communities globally,” Dr Chandrakumar says.

As climate change and urbanisation increase disaster risk, the findings of this research could help shape the next generation of EEWS. Ones that are inclusive, accessible and designed to empower people where they live.

Dr Chandrakumar says with proper maintenance and sustained community involvement, the number and quality of sensors can be scaled up.

“This kind of network could significantly enhance coverage and act as a valuable support to a future national EEW system implemented by the government. Similar low-cost EEW networks are already in place in other countries to complement their national systems. There’s no reason New Zealand couldn’t do the same. This research proves it’s possible,” he says.

The paper has just been published in The Seismological Research Letters and can be read in full here. The project was supervised by Associate Professor Raj Prasanna, Dr Caroline Holden, Dr Max Stephens and Dr Marion Tan.