When the Hunga Tonga-Hunga Ha’apai volcano erupted in Tonga on January 15, the result was devastation. The eruption literally blew up an island, caused mass flooding in the surrounding areas, coated whole communities in a thick layer of ash, and took out telecommunications for weeks. Yet in that eruption, we got lucky, according to a new commentary article published in the scientific journal Nature. Michael Cassidy, a volcanologist at the University of Birmingham, and Lara Mani, a volcanic risk researcher at the University of Cambridge, say things could have been much worse if the eruption had gone on longer, spewing more gas and ash. Next time, they explain, a similar disaster could have dire global consequences—and we need to be more prepared.
Four people died as a result of the eruption, but if the volcano had been in a more densely populated area, many more could have been killed. A few slight tweaks, and the eruption could’ve triggered mass infrastructure and even global supply chain disruption. A larger event could’ve easily caused worldwide climate destabilization (much more rapidly than what we’re currently experiencing with human-caused climate change). Which has happened before: The 1815 eruption of Mount Tambora, the largest ever recorded in human history, triggered a cold period commonly known as the “year without a summer.” Crops failed worldwide.
This year’s Tonga eruption hasn’t been formally ranked on the magnitude scale yet, but Cassidy and Mani have estimated that there’s a one in six risk of an eruption 10 to 100 times larger occurring in the next century. One in six—a roll of the dice. “The world is woefully unprepared for such an event,” wrote the authors. “The Tongan eruption should be a wake-up call.”
So just how concerned should we be? And how can we get ready for the big one? I spoke with Cassidy to find out. Below is our conversation, lightly edited for clarity and length.
Lauren Leffer, Gizmodo: Why publish this commentary now?
Michael Cassidy: I think it really came out of covid. We were all locked down, in this pandemic that we all sort of knew could have been on the horizon but never really expected. I think the probability, since now people have looked into it, was something like one in three, this century, of a pandemic arising. And, you know, lots of countries weren’t very prepared for it. But the biomedical community responded amazingly, and produced vaccines and other treatments. And I was left thinking, ‘How would my field, volcanology, cope with something on a similar scale, some sort of large catastrophic event?’ And my initial thought was, ‘probably not very well.’
I think that comes down to the size of the community, which is quite small, and also the level of investment. So I contacted Laura, who works for the Center for the Study of Existential Risk. And she’d been thinking about this as well.
Gizmodo: What gives volcanic eruptions such potential for chaos and damage?
Cassidy: Large eruptions can cross borders. They can produce lots of different hazards. So, for instance, extensive ash flow over thousands of kilometers, which can interrupt things like electricity and potentially collapse roofs of important buildings, buildings that could be critical to our sewage or other infrastructure. You know, there’s also things like tsunamis that can come off from volcanic flows. And we’ve seen the devastation that the Tonga tsunami and the Krakatoa tsunami in 2018 produced, and there’s also volcanic flows. Pyroclastic flows are very, very dangerous, normally fairly confined regionally, but they have a fatality rate of nine in 10. And then there’s things like lahars--big mud flows that can occur due to rainfall and loose, unconsolidated ash, even many kilometers away and just produce these really devastating floods.
Gizmodo: And really these are just the more immediate impacts, right? You didn’t even get into the atmospheric or the longer term climate issues.
Gizmodo: But knowing all of these potential impacts, in summary, how prepared would you say we are for this possibility of disruptive eruptions?
Cassidy: I think it depends on area. In some regions, we have a lot of resources set up to monitor volcanoes so that they can evacuate regions, you know, places around Europe, in particular, the U.S., Japan, Indonesia, actually, even with the resources it has, is also very good at this. And there’ve been lots of successful operations in the Caribbean. But there are some eruptions that catch us by surprise. And that can be a problem. So location is one point.
And I think we were far more resilient than we were, say 100 years ago, or 80 years ago, or even 50 years ago, to respond to smaller eruptions–which is great. And that’s due to innovations with health care and mitigation and increasing monitoring. But I think this doesn’t really include these kinds of larger eruptions that are more unprecedented. We haven’t experienced a larger eruption like this since 1815 in Tambora. And, yeah, I think we’re less set up for those larger events.
Gizmodo: You start with the example of the recent Tonga eruption in the commentary article, and describe all the ways we got lucky with it: The duration was relatively short, and it wasn’t in a super densely populated area. What would the worst-case scenario have looked like for that eruption?
Cassidy: That eruption we think was probably a magnitude 5 or a magnitude 6 eruption. So you know, about 10 times or even 100 times less powerful than what we’re talking about in this article. If that eruption happened in an area with lots of critical infrastructure and lots of people exposed, it could’ve been particularly hazardous. And this is where we’re less prepared.
Sometimes we can tell if an eruption’s going to happen. But we’re really quite bad, at the moment, at telling what style it will be. So whether it’d be really explosive or effusive, or just a lava flow, for instance. Or how big it will be. And I think that’s really key: that we don’t know how to predict or forecast that.
If that Tonga volcano event occurred instead in high density areas such as southeast Asia, or near the coast of the Philippines, it could have caused huge problems, affecting global trade–in places like the Straits of Malacca.
Gizmodo: You keep referencing the magnitude of an eruption. I think people are generally familiar with hurricane or earthquake magnitudes, but how are eruptions graded?
Cassidy: They’re generally graded on how much rock or ash they’re able to emit. So, something like the Tambora eruption, which is magnitude 7, is essentially something like the equivalent of covering the UK in 40 centimeters of ash.
A magnitude 8 eruption would be something like covering the whole of the U.S. in about 10 centimeters of ash. So these are very large eruptions. And a magnitude 8 eruption is rare. There’s a chance of about one in 170, in the next 100 years.
Gizmodo: To dig into this figure that you referenced in the article a little bit more, this one in six figure for a magnitude 7 eruption or greater—can you just walk me through how you got to that? I know it’s based on a 2021 scientific study you referenced.
Cassidy: Yeah, volcano researchers, they can look at ice cores. And from that they can pick out large eruptions from the amount of sulfur that they emit. So essentially, if you see a sulfur peak in the ice cores, that can tell you that there’s been a large eruption. They found about 97 of these through a 60,000-year period. They found about 97 eruptions the size of Tambora and above.
And so when they did this in 2021, they found that magnitude 7 eruptions or anything above occurred every 625 years, which equates to about one in seven. Then a magnitude 8 eruption—full disclosure, there were only three or four of those they found, so this is not a big data set—was about every 14,000 or 15,000 years. So that’s about one in 150. I used the geological record, which has a lot more data. And that provides a recurrence rate of super options at about 17,000 years. So it’s quite similar. So I used one in 150 and the one in seven to combine those risks to give us one in six this century.
Gizmodo: Understanding that statistic very well as a volcanologist, do you walk around feeling as if, at any moment, the world could just be thrown into utter chaos by a volcanic eruption?
Cassidy: I guess ‘no’ is the answer, although it is a relatively high risk. It’s more likely than not that we won’t have an eruption of that size in the next 100 years. But another part of me understands that, if there was a one in six chance that my house was going to fall down in my lifetime, then I think I would do something about it. I think I would insure it. I think I would invest money to try and improve that, and improve my family’s safety. And it’s kind of curious to me that we’re not really doing that for volcanoes.
Gizmodo: To get into what we could be doing differently, like what insurance for volcanoes might look like. I mean, you make a comparison to asteroids, and you talk about the planetary defense systems that are in development. I can imagine what throwing an asteroid off course or trying to prevent that collision would look like. But how do you stop a volcano from erupting?
Cassidy: At the moment, we can’t. It’s slightly in the realm of science fiction. But I think actually, it’s a mistake to just suggest that that’s where we stop.
I think there’s a good argument that we could start looking into this and start researching it. It comes with some really obvious and worrying risks. So, any sort of research needs to be done very carefully. But if we start researching these methods now, then in the decades to come we might be in a position to mitigate eruptions.
I should also say, there’s things we can do now, which would be really, really helpful, and would help mitigate a lot of the risk. As we point out in the article, pinpointing where the volcanoes are is important, because we don’t really know where some of these really large spikes in the ice core came from, and perhaps they’re close to populations, perhaps they’re in these really critical pinch points. So we really need to find those. And if we find those, and we intensify the monitoring around those, then I think we’ll be doing a lot of good.
And then we can prepare communities around those volcanoes. And we can also develop things like nowcasting. When there’s a larger eruption going on, we can, perhaps, send information to people’s devices to suggest what they can do. So, tell people ‘remove ash from your roof,’ if it’s going to collapse, if there’s a probability of more than five centimeters or so expected in their location.
Gizmodo: Literally the house falling down metaphor.
Cassidy: The roof collapsing, yeah. True, good point.
Gizmodo: How would you characterize the state of our current monitoring? Where are the big holes? How extensive is it?
Cassidy: For the investment, and the resources that countries have, I think they do an amazing job. I’ve seen this firsthand in Indonesia, and the Caribbean, and in Mexico and Chile.
But I think there is a problem that they’re run from national institutions. And that means that countries that do have the investment to put into volcanic research like the U.S., for instance, like New Zealand, like Italy—they’re a lot better prepared. They’re a lot better prepared for these volcanic eruptions, than less well-developed countries like Indonesia, Papua New Guinea, Philippines, even places like Turkey, for instance. In that respect, I think some areas are not very well monitored. I think it’s something like, just half of the volcanoes in Indonesia are monitored properly.
Gizmodo: How often do we know that a volcano eruption is going to happen before it happens? And how much warning do we get when we’re monitoring it?
Cassidy: We’re actually pretty good at telling, you know, if an eruption is coming. In terms of what style of eruption will occur, that’s where it does become a lot more tricky. And also predicting the size of eruption, that becomes really tricky.
Gizmodo: Is it that we need more research to iron that out and figure out the specifics? Or is it that we just need to always be prepared for the absolute worst case for every volcano that we’re monitoring?
Cassidy: I think we definitely need more research and things will improve. But we also might never get to the stage of being able to tell exactly when a volcano will erupt and in what style. It’s kind of like going to a doctor and them trying to predict when you’re next gonna sneeze. We might never get to that stage of precision. But I think we do need to set up our communities so that they’re prepared for a worst-case scenario. And that our critical systems, like electricity and trade networks and water security and food security, are set up so that we can be more resilient to shocks like this.
Gizmodo: You reference a few places that already are relatively well prepared. You referenced St. Vincent specifically and their Volcano Ready Communities project in the article. What does a well-prepared place look like for volcano eruptions?
Cassidy: It can mean a lot of community engagement, involvement, and education of both children and adults, so they can know what kinds of hazards to expect and where they can evacuate to. They did that really well in St. Vincent, and there were programs going for several years.
Gizmodo: Right, so there’s the local community level, and there’s the monitoring level, where we could improve research. But then there’s also the big global picture, right? Because, as you mentioned, these eruptions don’t have borders, and they can have really large scale global effects. How do we prepare for that side of things?
Cassidy: This is where I think we need more involvement from the UN, or maybe not the UN, but global bodies. Because currently, at the moment, every volcano observatory is run separately, and we really need to have this global, coordinated community. We need policies in the instance that one of these large catastrophic eruptions happens. You know, let’s stop countries from having export bans. If there’s a food crisis because there’s a global volcanic winter, which would kill crops in many places, then that would help alleviate it.
With monitoring networks, we normally can tell that eruptions are going to happen. If we know it’s from one of these big volcanoes that’s capable of a large eruption, then in the weeks or months prior to that, we can start to come up with plans to change trade, to make sure that trade goes through a different way. Or that in that country, they have access to extra water, masks, excess food, backup electricity, and things like that. I think we need global coordination and proper investment, basically.
Gizmodo: A lot of this sounds like what we would need globally for any sort of major disaster. Right?
Cassidy: Yeah, yeah, totally. And I think there’s lots of parallels to nuclear winters or even asteroid winters and other types of large, global catastrophic risks. So I think there’s certainly things that can be learned from past volcanic eruptions.
Gizmodo: Getting back to the more speculative, or sci-fi, ways we could potentially mitigate volcanic eruptions. What do you think it would look like to create a planetary defense system for volcanoes?
Cassidy: There’s a few things. First off, the atmospheric after-effects. We could find almost no studies into this, but it feels fairly intuitive that one of the biggest risks globally for volcanic eruptions is this volcanic winter effect, where large amounts of sulfur are ejected into the stratosphere, and they spread across the world, and they deflect solar radiation, cooling the Earth. That can have impacts on drought, it can push around monsoon areas and things like that, and also destroy crops. This kind of food insecurity problem, it’s caused kind of historic famines in 1815, 1257, after some of these large eruptions. So we know that it’s a really big deal. One of the most intuitive things is really to kind of try and counteract that. So that could look like speeding up the removal of those sulfur aerosols in the stratosphere.
And generally, big eruptions have up to about a five-year impact on the atmosphere, although sometimes it can be longer. What that means is that you can inject short-lived global warming agents to counteract that cooling, just for a few years. So that’s after the eruption has taken place already.
Then there’s the other way, which is kind of being looked at, though I think this is a long, long way off. But it’s to start drilling into the crust. There’ve been about three or five instances where we’ve actually, by accident, drilled into pockets of magma. What it tells us is that we’re able to interact with magma bodies. Granted, these are really, really small—nothing compared to what we’re seeing with these large magnitude eruptions. But by understanding how we can potentially manipulate those magma bodies under the crust, maybe we can increase the fracture networks around those magma bodies so they de-gas.
Gizmodo: Like a pressure valve?
Cassidy: Yes, exactly. This research is just starting to be looked at now. One of the areas in Iceland where they accidentally drilled into this magma pocket to look for geothermal resources, they’ve now decided to make that an international magma observatory. This could help us with loads of different things. It could improve monitoring, like all the ways we were talking about earlier. We could put sensors down there. But also we can potentially learn how to manipulate eruptions directly.
Gizmodo: And you mentioned in the paper that human-caused climate change itself has the potential to exacerbate some of the consequences of volcanic eruption. Can you explain that a little bit?
Cassidy: It’s almost three effects. So firstly, you know, if you have multiple extreme weather events, and droughts, and fires and stuff like that, that really stretches your humanitarian organizations and the countries that try and manage these sorts of disasters. It makes us less resilient to deal with another big catastrophe.
The second reason is that, as you start to melt glaciers, as you start to increase sea level rise, as you start to increase rainfall, like we are seeing with climate change—these can all have subtle but real effects on volcanic systems that have to do with stresses inside the crust. When we de-glaciated from our last ice age about 10,000 years ago, we saw an increase in volcanic eruptions. It’s probable that we will see an increase in volcanic eruptions. Though, we’re not sure under what timescales these will occur, perhaps it’s the next 100 years, perhaps it’s the next thousands of years.
The third reason is that, if we do enter a very hot climate—and I’m talking next century, 2100 sort of territory—then we might be seeing increased atmospheric circulation. And you might also see differences in how our oceans are layered. And that means that any sort of eruption from the tropics might have a much more amplified effect.
Gizmodo: You mention a dedicated satellite monitoring system in the article. What would having a system like that change about our monitoring ability and knowledge?
Cassidy: So this could either look like a dedicated infrared satellite before eruptions occur. It could also look like a pseudo satellite, or high-altitude drone, sent out once we have indications that an eruption is coming.
It wouldn’t be able to replace current monitoring systems, but it would give us extra tools before, during, and after an eruption. Before, in prediction. And during, it can tell us how fast a plume is spreading from above, which is something we didn’t really have in Tonga. If we could tell how big an eruption is in real time, then we can get better information, faster to the communities who need it.
Then thirdly, post eruption. You know, in Tonga, for instance, we had to rely on a plane to take off, and it couldn’t land because ash was on the runway for several days. The satellite could help us work out what areas are damaged, it could tell us who’s in need, what the humanitarian situation is, and how we can better save lives in the area close to the volcanoes.
Gizmodo: Given that this idea to publicize this issue came out of observing the covid response and given where we are with the covid response now, how optimistic are you that these big investments and changes you’re outlining are on the horizon?
Cassidy: It’s a good question. So this came out of covid but then also obviously the Tonga event. And, you know, I think we can only improve incrementally. We’re doing a lot of work now in the background. This kind of article is hopefully gonna give us impetus to have more conversations with humanitarian organizations, global policymakers, and funders. We’re hopeful that we’ll see some incremental changes in this sphere.
We also hope that it might change discussions within volcanology. There is a focus on these smaller and more frequent eruptions, which are really hazardous. But by shifting to look at some of these really big eruptions and following some of the steps that we suggest, there’s actually a trickle-down effect. So we’ll see.
And, you know, I hope that pandemics also get more investment. $10 billion going into pandemic preparedness, I think that’s even too little. But, volcano research gets next to nothing.
Gizmodo: Is there any last thing that you want to get across? Is there anything important you want the public to know, that we didn’t touch on?
Cassidy: I think people might say, ‘well, that’s great and all Mike, but these eruptions are very infrequent. And they don’t happen a lot. Is it worth spending the money?’ But magnitude 6 eruptions can go up to about a $7 trillion amount of damage. If we had a magnitude 7 eruption, or even higher, that can have global consequences. So, we’re looking at multi-trillion amounts of loss from that. When you spread that over the 600 years or so of the recurrence rates, that equates to about a billion dollars each year. I think, even an investment of several tens of millions, which is 1% of that, to reduce that risk is worth it financially and economically.