Monday, September 12, 2016

Volcanoes, Tsunami, and Earthquakes – GeoNet kicks into high gear

Okay, I’m going to be honest. The last couple of weeks rocked GeoNet (see what I did there?)....

It all started on the 31 August, a Wednesday, with the Tier 4 Exercise Tangaroa. Tier 4 means a national exercise and that means everyone, from local, regional and national authorities, participate. It was a large tsunami scenario; GeoNet played a pretty big role in the exercise. Not 48 hours later, it was like Ground Hog’s Day: a large earthquake striking off the Northeast corner of New Zealand. Unlike the exercise, this earthquake barely met our criteria for activating the Tsunami Experts Panel, but was scarily close to how the Tangaroa scenario started for us! More on that later.

We’ve had a pretty busy series of aftershocks since then, and our scientists worked tirelessly to get the Operational Earthquake Forecast out as quickly as we could. Then…Volcanoes! Ruapehu, and White Island. That’s right, never to be left out of the action, Mt. Ruapehu decided that it had had enough of a cold crater lake and that it was time heat things up. And there was a small eruption at White Island. To top it all off, we had a M6.2 off Macquarie Island, 1,000 km south of the South Island (and technically part of Australia), which caused a M5.3 ghost quake in our automatic system (proving again we have more work to do!).

GNS Science staff at the science response meeting in the GeoNet Media Room following the M7.1 East Cape earthquake

After all this, it would be easy to just take the weekend off, put up our feet and drink some kind of adult beverage (single malt and/or red wine for me thanks). But, we’re GeoNet. We don’t get to rest or stop. We can’t turn off our phones that wake us up at night. Our project manager is New Zealand itself; we are on her time.  And, we find that after a busy time, it’s good to sit back and reflect on what we’ve learned.

Here’s a brief rundown:

  • Shaking is still the best alert: We love technology at GeoNet - I personally love technology as I have stated many times. LOVE. IT. But sometimes technology is not the answer. This is why I want to thank everyone living on the coast who did it right - felt the earthquake and evacuated. It’s not an easy decision to make but had the tsunami been larger…well, let’s not go there. Anyway, well done.
  • Decision-making with only a little bit of information is tough: remember I mentioned the Tsunami Expert’s Panel? The Tsunami Experts Panel is a group of senior scientists from across New Zealand who advise whether a tsunami will be generated or not and on the potential impacts. In this case, the source of the earthquake was very strange and with our instruments confined to the coast, our data was pretty limited.  And making decisions with little information does not come naturally to most trained scientists; we like enough information to make calm, informed decisions. This is why we partner so well with our friends at the Ministry of Civil Defence and Emergency Management. 
Which is why…
  •  Partnerships are critical – people were critical of our friends over at the Ministry of Civil Defence and Emergency Management and others for the slow or lack of alerting.  And we understand, with all the technology, it seems natural that we should be quicker alerting people. But behind the scenes, it isn’t as straightforward as pushing a button (that mythical big red button).  Ironically, a large event is simpler because it is well over the threshold for action. For the M7.1 it was well below the threshold with high uncertainty - higher than any I’ve seen for a while.
  • And returning to the Exercise Tanagora – that giant pretend tsunami generated by a pretend M9.1 earthquake 1000 km north of Tauranga. In the exercise universe the Experts Panel had problems with forecast model calibration because none of the deep sea buoys (DARTs) in the area were operational. The news is that this is the real world case – all three DART buoys north of New Zealand have currently failed. These buoys, which are operated and maintained by the United States, are the most expensive to repair because of sea conditions and distance from normal repair facilities. We understand it will be some time before they are repaired.
  • People still love the long Felt Reports – we had so many complaints about our classic Felt Report system that we thought going to a Rapid system would be universally embraced.  Well, we were wrong. So we’ve been putting together Felt Detailed. We’ve trailed this long form Felt Report, it still needs some tweaks to be ready for all earthquakes, but we are working on it. Personally, I’m amazed at how many people are willing to give us feedback on a survey that takes longer than 20 minutes. But, there you go (personally I have great problems fill in forms!).
  • Probabilities are still hard to communicate – we’ve got all kinds of social science research on how to communicate probabilities. But even with our table, the scenarios and charts, it is still hard to discuss probabilities without causing some kind of confusion.  So, why bother? Because we think that it is better to tell people what we know (the probabilities) than not. We’ll keep trying to get better at it but we know it’s hard.
At GNS Science, we’ve got over 30 staff (and many more from across the organisation) dedicated  to keeping  GeoNet running.  I just want to give a massive thank you to my hard working teams of duty officers, app and website developers, public information specialists, seismologists, and the technicians who spend hours in the cold fixing and updating our 600+ instruments throughout the country. You are individually and collectively awesome!