2018•04•19     Japan     Yomiuri Shimbun

Prof. Jakob Rhyner, Director of UNU-EHS. Photo: UNU-EHS

In January of this year, Japan’s early warning system alerted citizens of an imminent strong earthquake, but it turned out to be a false alarm. The incident drew criticism from a Japanese public still fresh with the memories of Tōhoku, an earthquake and tsunami in 2011 that resulted in nearly 16,000 deaths and was described by the former prime minister as “the toughest and the most difficult crisis for Japan” since the Second World War.

On the occasion of the 7th anniversary of this earthquake and tsunami, UNU-EHS Director Jakob Rhyner wrote an op ed that was published today by Yomiuri Shimbun, Japan’s most read newspaper. In it, Rhyner explains why false alarms in early warning systems are not necessarily only a failure, and how they can be used as an opportunity for improvement.

To read the article on the Yomiuri Shimbun website, please click here. Below you can find an English translation of the op ed:

“In January, Japan’s sophisticated early warning system detected an imminent earthquake. ‘Prepare for strong jolts,’ an alert message sent to millions of citizens read, predicting a 6.5 magnitude tremor.

But it was a false alarm. According to newspaper reports, after investigating, officials suggested the system had misinterpreted two smaller quakes.

But a false alarm should not be interpreted only as a failure. In fact, occasional false alarms are unavoidable in early warning system, and are important ways to learn and further develop the systems. Failure is a necessary part of this process. But to be instructive, they must also be accompanied by transparency and thorough analysis.

Before I joined the United Nations University’s Institute of Environment and Human Security I worked at the Institute for Snow and Avalanche and Research in Switzerland. We were responsible for avalanche safety, including the daily forecasts of the avalanche danger for Switzerland, which were used by local authorities to instigate safety measures, but also by the public.

In the context of avalanches, we had a ‘hit rate’ of correct forecasts of around 80%. While on the surface this may appear successful, and indeed the system has a high public reputation, we actually spent a lot of energy reminding the public and the safety services about imperfections of our system – that our warnings were wrong or at least inaccurate around once per week. As scientists and researchers working in risk management and risk reduction, we were weighing two critical parameters against each other: that people shouldn’t be hit by an unexpected event – in our case, an avalanche; but also that false alarms shouldn’t be frequent enough for people to stop caring, or paying attention.

It is critical that this delicate balance is part of the message offered to the public. When a mistake is made, like the one in Japan in January, the safety services and the government can do two very different things. They can say: ‘This was a mistake, it will never happen again’. Or, they can say clearly to the public: ‘We deeply regret this. However this is a part of what early warning systems are. Occasional failures are unavoidable. But they are helping us to improve.’ Clear communication is key, and this is also a crucial way to maintain trust in the system itself.

Informed by the devastating events in Tōhoku, the international community came together in 2015 to sign the Sendai framework. One of its seven key targets relates to early warning systems, pledging to ‘substantially increase the availability of and access to multi-hazard early warning systems and disaster risk information and assessments to the people by 2030.’

Since 2015, there has been a global leap forward thanks to a better general awareness of the issues, technological advancements and effective policymaking, and Japan has played a leading role. But if we are to meet Sendai’s aim we must be transparent about the failures – as well as the successes – of our early warning systems. We must properly analyse false alarms, as well as undetected events, not as a way to apportion blame but to better refine and improve the sophistication of these crucial and life-saving systems. Warning systems are not perfect, and they will never be perfect. Transparency is key.”