Visitors to the Whakaari/White Island during its violent eruption in December 2019 had no warning that it was going to erupt and spew acidic steam at them. The eruption caused the deaths of 21 people, with survivors experiencing serious injuries and burns. The eruption was the catalyst for the development of an early warning system designed to pick up patterns in seismic activity to help alert people when an eruption is imminent. If the system had been in place during the earlier eruption it would have given people sixteen hours to clear the area before the Volcano erupted.
There are a number of other Volcanos in New Zealand that also pose a similar eruption threat and early warning systems could help to save lives around the country. Early warning systems can elp to provide visitors and park manager with real time information about potential hazards and safety issues.
Whilst New Zealand has monitoring instruments and measures in place, they have not previously delivered real time information or made estimates on the likelihood of previous eruptions. New technologies continuously monitor data and use machine learning algorithms to read signals that show an imminent threat of eruption. Machine learning algorithms are also capable of learning from past experience so that they can better forecast future events.
The new early alert system for Whit Island will generate an alert if an eruption is more than 8.5% likely to occur which means there’s about a 1 in 12 chance of eruption – the new system had it been implemented for previous eruptions would have raised an alert for four of the last five big eruptions.
Similar early warning systems could help to ensure that visitors are protected but there may be a trade off in that people are kept away at times when an eruption does not occur, the technology will surely improve over time however.