Scientist Michael Mann on ‘Low-Probability But Catastrophic’ Climate Scenarios

Photo: George Rose/Getty Images

This week, to accompany our cover story on worst-case climate scenarios, we’re publishing a series of extended interviews with climatologists on the subject — most of them from the “godfather generation” of scientists who first raised the alarm about global warming several decades ago.

Michael Mann is a climatologist at Pennsylvania State University, known primarily for his 1999 “hockey stick graph” of global mean temperatures (which shoots up quite dramatically in the 20th century). He has also been steadily involved in the United Nations’ periodic IPCC state-of-the-planet reports, and with Lee Kump has even adapted those reports into a popular, accessible book summarizing the findings. The book, which was updated to reflect the most recent reports, is called Dire Predictions.

Shortly after this week’s cover story was published, Mann took to Facebook to voice some criticism of it — primarily about its framing, which he described as counterproductively “doomist.” Personally, I don’t think we’re doomed, just facing down a very big challenge. But I own up to the alarmism in the story, which I describe as an effort to survey the worst-case-scenario climate landscape. We have suffered from a terrible failure of imagination when it comes to climate change, I argue, and that is in part because most of us do not understand the real risks and horrors that warming can bring, especially with unabated carbon emissions. For the sake of clarity: I do not believe that the planet will become uninhabitable in 2100. As I write in the story, our complacency will surely be shaken before we get there. But I do believe that it is important to contemplate the possibility that parts of the tropics and equator will become cripplingly hot, for instance, or that our agriculture will suffer huge losses, so that we may be motivated to take action before we get to those eventualities. And I do believe that, absent a significant change in human behavior across the globe, they are plausible eventualities.

Mann also took issue with a few particular points of science. He stressed that the danger of the carbon frozen in the arctic permafrost was not a “game-changing arctic methane time bomb” and, separately, he suggested that the recent upward revision to a particular satellite data set on warming was less significant than I made it out to be. My purpose in raising the permafrost issue was to illustrate how uncertain much of our current modeling can be, not to suggest a sudden methane release would be the major cause of devastating warming: I based none of the warming scenarios described in the piece on a dramatic methane release effect but rather on the high end of the IPCC’s business-as-usual estimate, which gave a roughly 5 percent chance of our hitting eight degrees of warming by 2100. Regarding the data set, I grant that the upward revision may have been less meaningful to the scientists close to the data, who understood it as a revision toward expectations, than it was to journalists covering the development from afar, who focused on the fact of the revision itself.

I have an enormous respect for Mann, and for his perspective on climate change — he has been an invaluable force both as a scientist and as an advocate. That is one reason why I called him up, during my research, to talk to him about what he thought about the low-probability, high-horror possibilities of climate change. Given his criticisms of my story, we’ve decided to run this transcript unedited.

Maybe the way to start is for me to just tell you about what I’m trying to do and what I’m after. My basic perspective as a kind of close but amateur follower of all this stuff is that the business-as-usual forecasts even in the IPCC get us to some pretty scary places. And then there are some reasons to think that those projections may be a little bit conservative, even over the single-century timescale.

It seems to me that there hasn’t been much out there about worst-case scenarios. Maybe that’s in part because scientists have been so anxious that the world — or at least the American public — not impugn their work as speculative or dangerous. And so scientists have felt a need to be a bit restrained in talking about what is possible. But to me it seems like it neglects a lot of really terrifying possibilities and that those possibilities are important to consider because they spur action.

So I’m just hoping to do a piece that walks through what would happen in a world that is four, five, even six degrees warmer: What kinds of threats and challenges we’d face; how likely they are; and what are the things that are quite likely that we should be worrying about quite a lot; and the things that are less likely but so scary that we should be watching out for their arriving and being sure to forestall them when we can. So I’m sort of doing a listening tour, trying to talk to as many people as I can about these things to see what they think in that context. What are things that they think, what are the threats that they think have been underemphasized, what do they think the world would look like if it was four, five, six degrees warmer, and how to talk about that world in a way that’s both sensible and, you know, sort of fair to the risks.
Yeah, I mean, there has been quite a bit made about the so-called scientific reticence — the tendency for scientists in general to actually be conservative in what they state and the conclusions that they state, particularly when it involves the public sphere. Naomi Oreskes, I don’t know if she’s on your list of folks to talk to.

So she’s written on this, specifically with regard to climate change and how there has been a tendency for scientists to often understate, sort of, the potential risks and the timeframe on which they may unfold because of this sort of combination of innate conservatism among scientists and also sort of this assault on science by climate-change deniers.

I think the intent of that assault has been to sort of cow scientists into retreating from the public discourse and frankly intimidating scientists into being very guarded and very conservative about their public statements. I think that’s been the intent, or one of the intents, of the fossil-fuel-industry-funded attack on climate science. This is something I’ve written about in a book about my own experiences, The Hockey Stick and the Climate Wars, and more recently this book with Tom Toles, The Madhouse Effect. Which is sort of more up to date. And we talk quite a bit about that. So I think your observation is correct. There has been a tendency to understate risk.

And the other point that you make is a very important one, which is that when it comes to societal decision-making, it’s critical to not just consider the most likely impacts, but those sort of low-probability but catastrophic sort of cost scenarios. What we call the so-called tails of the probability distribution. The things that may not be, we can’t conclude that they’re likely, but we can’t rule them out, and if they were to happen, they would have such a catastrophic impact that it makes sense to take them into account. And sort of do any cost-benefit analysis of the importance of acting to avert ongoing warming and climate change. So one scenario, and something I’ve done a fair amount of work on: you have a combination of sea-level rise and potentially stronger tropical storms and hurricanes. The flood risk for New York City may be such that we literally have to abandon New York City and many of our largest coastal cities and many of our largest naval bases like the one in Norfolk, Virginia. All would be very vulnerable to even modest amounts of additional sea-level rise. If you look at the last IPCC report, which is sort of the state of scientific consensus, the report concluded that the sort of upper end of the range of likely sea-level rise was somewhere in the neighborhood of a meter. Over three feet of sea-level rise by the end of the century. But the science that has been done since then, just over the last couple of years, now suggests that we probably have to double that. The most likely sea-level rise by the end of this century is probably somewhere in the neighborhood of six feet, two meters, and that has to do with science that’s been done since the last IPCC report, work by climate scientists to put into the models some of the processes that were not really resolved by the models previously, and that’s a recurring theme.

Uncertainty, of course, can cut both ways, but in many respects, the progress in the science over the last decade or so has been such that we tend to see more risk. We tend to see the probability distribution shifting in the direction of greater impact, larger magnitude impacts. And that’s true with sea-level rise, it’s true with tropical storms and hurricanes, it is true with drought, where we are seeing a pretty dramatic increase in both the extent of drought in middle latitudes. In California, of course, the worst drought on record. The paleo-climate scientists tell us it’s probably the worst drought in at least 1,200 years. There is science that’s emerging, some of that is science that we’ve done, that has isolated sort of subtle mechanisms by which climate change can influence extreme events like droughts and like floods and like heat waves, and again these are things that, there are processes involved that are pretty subtle, and not well resolved by the global models that have typically been used for climate-change projections, and, again, by the nature of the science, is that these sorts of impacts are more likely than the projections would have had us believe just five years ago or ten years ago. So there is this recurring theme of the science moving in the direction of the impacts being larger than we expected and part of that is a function of the reticence of scientists and the tendency to sort of be very conservative.

A good example is the last IPCC report — actually, sorry, well, two IPCC reports ago, the fourth assessment report, and this got quite a bit of attention at the time, this was back in 2006 where the report gave an upper sort of end range of sea-level rise that was just on the order of a foot, a little over a foot over the next century and there was sort of an almost like an asterisk next to that conclusion, it was almost buried in the footnotes if you will, was the fact that they didn’t include the contribution from melting ice sheets, from the Greenland and Antarctic ice sheets, because they couldn’t estimate that contribution. And therein lies the rub, because it’s the major driver and it can only weigh in in one direction, so there’s sort of an asymmetric neglect of processes that almost invariably imply worse changes than what are being projected. So, yeah, your thesis, I would agree with your basic thesis here. A worst-case scenario — and when you asked about business as usual, and I assume by business as usual you mean if we don’t really, if the world adopts a Trump-like approach to climate change over the next century where we literally don’t act at all, we continue to escalate our burning of fossil fuels, we don’t move away from fossil fuels towards renewable energy. In that sort of scenario, then, by the end of the century, you’re probably talking about a four to five degree Celsius, seven to nine degree Fahrenheit warmer globe. The warming is even greater than that where people live because we live on continents rather than in the oceans, and the land warms up faster than the oceans so when people quote a global average temperature that’s actually misleading. Most of us will experience more warming than the global average because the global average is held down by the oceans, that don’t warm as much. Doesn’t help us, those of us that live on land. So more than the seven to nine degree Fahrenheit warming of the globe …

And in certain areas, even more pronounced than that.
Twice that much in the Arctic, where we’re already seeing some of the worst impacts. So the loss of sea ice and the ramifications and all this stuff sort of circles back, again getting at this issue of processes that haven’t really been fully incorporated or represented in the main climate model projections, because when you lose Arctic sea ice, when you warm up the Arctic that fast, you lose Arctic sea ice faster than you expected. Greenland melts faster than you expected. The Greenland ice is of course going, Greenland is losing ice quite a bit faster. The climate models projected that Greenland and the West Antarctic ice sheet probably wouldn’t lose ice mass until the middle of this century. Just ten years ago, that would’ve been sort of the state-of-the-art projection, and what we’ve learned is that first of all, the satellite measurements tell us it’s already happening, so clearly that’s not correct, they’re already losing ice; and now we’re beginning to understand the processes by which that happens and we’re beginning to incorporate those into models and the sea-level-rise projections are increasing, the projection of melting is increasing, and when you melt all that Greenland ice and that water flows into the North Atlantic, not only does it contribute to sea-level rise, it freshens the North Atlantic and it potentially shuts down the so-called conveyor-belt ocean circulation. The Day After Tomorrow scenario — of course the movie is a caricature of the science but there is an underlying grain of truth to that, and we published an article just a couple of years ago, my colleague [Stefan Rahmstorf] and others showing that this already appears to be happening.

Shutdown of conveyor-belt ocean-circulation pattern — that could, you know, it’s not going to lead to about anything that’s portrayed in the movie The Day After Tomorrow, but it would mean potential decrease in the productivity in the North Atlantic, which we rely upon for fish, seafood and fishing, worst drought, far more extreme heat waves, longer duration, more intense heat waves.

A good fraction of the tropics are now too warm essentially for human habitation and there are studies that show that once you warm temperatures by even a few degrees by what they are currently in the tropics, basically you can’t work outside. Productivity plummets, agriculture plummets, basically these regions become unlivable so you’ve got a smaller amount of surface area on the Earth to support a growing global population. You see a tendency for worse drought and decreased freshwater availability over much of the continents of the world. Again, as I mentioned, food, productivity …

That seems like to me a really scary and also quite underappreciated by the broader public, the effect on food.
No, absolutely. Food, water, land, you know? The basic resources that we rely upon. All of them are adversely impacted by climate change and with a growing global population. So you’ve got more competition over fewer resources among a growing global population. It’s a recipe for a conflict nightmare. And this is why when you talk to national-security experts, many of them will tell you that climate change may be the greatest security threat we face in the years ahead, it’s what they call a threat multiplier. It heightens existing tensions, it heightens conflict, especially when you’re talking about more competition for fewer resources. And interestingly enough, and this is not actually that widely appreciated, this is sort of a dystopian scenario that Hollywood imagined decades ago. The movie Soylent Green, with Charlton Heston — if you watch the very beginning of the movie, it’s briefly mentioned, but the underlying cause of that dystopian future is global warming, is climate change. So there was — it was sort of an oddly prescient, some of these early-’70s dystopian novels and films and Soylent Green foresaw exactly the sort of future that we’re talking about where climate change leads to decreased resources. In that case, it was about food for a growing global population, and it’s a dystopian future. A worst-case scenario — a worst-case future does not look that different from the dystopian visions that Hollywood has already provided us.

It’s really interesting to me to think that in the ’70s we were starting to get an understanding that the climate was warming, but it’s also the case that most of the emissions that we’ve seen globally have come since then. I think we have a little bit of a bias in the developed West to think of this all as like a debt from the industrial revolution that we’re repaying, but it’s really striking to me to think about how much of the damage that’s been done just in the last couple of decades, and therefore how much damage we’re doing every day and how consequential it is.
You make a good point. Most of the carbon that has been emitted up through now is still from the industrial world, but as you note, moving forward, it’s all about the developing world. China, India, South America, and so there is this legacy, of course we’re responsible in a sense for much of the climate change we’ve already witnessed, but what will push us over the threshold of dangerous interference with the climate now depends very much on what the developing world does, and that’s where you get into these very tricky global negotiations where you start talking about, Well, what about reparations from the West to— yeah, so it gets into some thorny issues of international diplomacy, but that’s the bottom line. Moving forward what’s going to determine our fate is much more the path taken by the developing world.

Obviously they’re in a position much less prepared to deal with it, they’ve got much less money to build technology, to move people, it’s all, sort of, that’s a threat multiplier too, poverty.
That’s right, and that’s why it’s important to, within the Paris agreement, quite an effort to, for the world to provide resources to help — and reparations, but really resources to help them transition directly to a renewable-energy economy rather than going through the same fossil-fuel energy economy stage that we went through. We can’t afford for them to do that.

Right. Can we talk in a little more detail about sort of each of the scenarios? I don’t know if you have time, we were talking about food, water, and land. I was hoping maybe you could just walk me through in a little more detail how each of those areas, maybe even into particular regions, where you think things will be hardest hit, where you think pain points will show up first, that kind of thing.
Yeah, so in terms of drought, what happens is the sort of dry desertlike subtropical regions expand forward, because the region of subsiding air is what leads to dry conditions. When you have a tendency for rising air, like in the deep tropics, that’s where you get the rainfall, or again in the middle latitudes in the form of mid-latitude storms that are associated with a tendency for rising air, but in the subtropics there’s this tendency in the large-scale circulation of the atmosphere for descending air. So that’s where you get the dry desert belt. And that descending air, the sort of down-welling part of the circulation to the global atmosphere that expands polewards, and so it’s not coincidental that California has seen such bad drought. That Syria, the Middle East, the Mediterranean, these regions are seeing worse drought. The subtropics are both drying, and they’re expanding poleward and so the southern half of the U.S. is increasingly going to experience more and more prolonged and pronounced droughts.

The Middle East, you know, which is of course again— when you talk about all the basic stresses, food and water and land, that’s what’s driven conflict in the Middle East. Some would argue it’s very much been about, ultimately, conflict in the Middle East grew out of competition for precious water resources and so you see in the case of the Syrian uprising, it was driven by another drought that is the worst on record, the Syrian drought. Again, the paleo data suggests it’s the worst drought in at least 900 years — that’s as far back as they could go in fact — and that drought is ultimately what displaced rural farmers. They moved into the cities and there was increased conflict that arose from that and ultimately ISIS was able to flourish in that environment of instability and conflict and now we’re seeing the global repercussions of that from a national-security standpoint. So, drought. That’s a big one. It’s behind a lot of the conflict that we see, it’s increasingly impacting the key agricultural regions in North America for certain, that now, so that’s— I forget, how did you want to break this down in terms of, uh …

I was just going off of what you said. You said land, water, and …Yeah, yeah yeah yeah. Yeah. So land, global sea-level rise, I think it’s something like 25 percent of the global population lives within something like 30 feet of mean sea level. So even three to six feet of sea-level rise will displace many millions of people, and so that’s where you start getting into issues of, hundreds of millions. We’re not yet in the billions department; I think you have to get close to 30 feet until you’re displacing a sizable fraction of the global population, but yeah, in the hundreds of millions of people at six feet of sea-level rise, I think is fair to say. That of course leads to, that’s the source of environmental refugeeism— you know if you’re, you’re literally displaced by flooding, you have to go somewhere else. The worsened drought and worsened heat with basically some studies suggesting that the entire tropical band becomes unlivable at four to five degrees Celsius, seven to nine Fahrenheit warming of the planet. So you basically now have to take all of the population that currently inhabits the entire tropics, which is a good chunk of the global population, and now they are forced to higher latitudes up into Eurasia, up into North America. That’s driven by drought, driven by sea-level rise, driven by excessive heat, so you’ve got environmental refugees and the competition for food that, in the tropics, most cereal crops are growing, it has to do with the productivity of cereal crops, is a function of temperature and in the tropics, you’re very close to sort of the optimal point in that curve. Crops are about as productive as they can possibly be, so you’re right at the peak. And if you’re right at the peak, then even a little bit of warming causes you to go down the other side of the peak, so you see very substantial decreases in agricultural productivity. Rice, sorghum, maize, all of the major crops, very sharp decreases in productivity with even a couple of degrees of warming, and we’re talking about well beyond that. It used to be, though, that warming, a little bit of warming, was actually good in the mid-latitudes, you get longer growing season, shorter winter, but what we’re also finding, and we saw that with the Siberian wildfires two years ago, with some of the devastating floods that we’ve seen in recent years, that weather instability, extreme weather events, are also on the rise. There is an underlying connection with climate change, and the destruction by extreme weather will likely completely offset any marginal gain we might’ve expected from nominally longer growing seasons.

It’s another example of how a sort of naïve view of this issue, even five years ago or so, has now been replaced with a more nuanced understanding of how climate change will impact productivity of crops and livestock, and as we develop a better understanding, we realize that the impacts are likely greater and worse than we thought.

What about freshwater — not water in the sense of drought but the availability of freshwater aquifers and reservoirs, is that something that concerns you?
Yeah, I mean what’s happened, there’s some good examples like Lake Powell in the West. We’re seeing these lakes drop as they’re depleted by human usage, but also as the climate dries, rainfall decreases, and evaporation increases, so you’re getting it from both ends, you’re seeing these reservoirs disappear. And yeah, we can tap aquifers, but that’s sort of a, um, it’s a Faustian bargain in a sense because you know ultimately, that is millions of year old water and once you deplete it, it’s not getting replaced for millions of years. So we go into this sort of water debt. And we can do that for a while, but we’re tapping out a lot of the major aquifers and there are other problems here, like in parts of Texas where they’re experiencing, where they’ve experienced really bad drought in recent years and where the water that they do have is being used for fracking for example in West Texas and central Texas, so they’re actually trucking in drinking water and using the local water that’s available for energy production. That’s sort of a Faustian bargain, you’re sort of going into debt when you’re relying on aquifers when your reservoirs aren’t being replaced, when your loss through evaporation outpaces what you’re accumulating from rainfall and snow melt, and with California, the real problem and what contributed to this record drought was both sort of a decrease in rainfall and snowfall with storms sort of tending to migrate north of the state, but record warm temperatures, soil temperatures, warmest summers on record, meaning increased evaporative loss, and very little snowpack, and when you don’t have snowpack, you don’t have freshwater runoff in the spring that you rely upon for agriculture and other purposes. It’s sort of a perfect storm of more evaporation, decreased snowpack, and decreased rainfall all coming together to yield the worst drought in a millennium or more.

What about disease? Is that something that you’ve thought a lot about, worried about?
Most of these things aren’t things I’ve literally studied, but I’ve tried to review the existing literature so that I’m somewhat knowledgeable about the array of impacts. As it happens, I actually have published some work on infectious disease and climate. Infectious disease, there’s a very strong group here at Penn State and the Huck Institutes of the Life Sciences where they study infectious disease, including malaria, and I’ve done some work with those folks taking climate model projections and looking at how that could impact infectious diseases like malaria. And the science again, a lot of it, a lot of what we’re finding is nonintuitive and much more nuanced than if you had asked scientists a decade ago. For example, with malaria it turns out that there are a number of factors obviously that come together. Whether there’s a favorable habitat for the mosquito that carries the malaria parasite, but the other piece there is the malaria parasite itself and how temperature influences the sort of life cycle of the parasite. And it turns out that that life cycle is a very strong function of temperature and once you get above about 18 degrees Celsius, so once you get into the 60s, basically, you start to see, or once you get into the 70s rather, you start to see a very sharp increase in the rate of reproduction of the malaria parasite and you have cities like Nairobi that were built at elevation to be basically above the malaria line and now with global-warming temperatures, that line is moving upward and so, over time, you’re going to have extremely large populations that were once sort of protected from malaria are going to become vulnerable to malaria, like Nairobi, these high-elevation cities in equatorial and tropical Africa. Now, on the other hand, it turns out that some of the warmest parts of Africa at low elevation, interestingly, if you warm temperatures enough, you go, so, at 18 degrees Celsius you’re sort of climbing up that peak, but once you get to about 25 or 26 degrees Celsius you’re coming down the other side of that peak and so some of the warmest parts of Africa could actually see a decrease in at least this factor, the rate of reproduction of the parasite. Now there are other factors that come into it, whether there’s a habitat for the mosquitos, which depends on rainfall, and there are many other factors as well.

You’re also moving the sort of area that’s most affected from regions that have adapted some degree …
That’s exactly right, that’s exactly right. And the faster those changes, the harder it is, the less adaptive capacity you have, and that’s true across the board. The faster climate changes, the harder it is for areas that could in theory see a potential benefit to take advantage of that benefit, and those regions that see negative impacts, the impacts occur faster than they can adapt to — so that’s exactly right.

*At Mann’s request, we removed a brief discussion of an embargoed paper.

Scientist Michael Mann on Climate Scenarios