This is a special edition of Scientific American’s Science Talk, posted on June 18, 2020. I’m Steve Mirsky.
Mark Fischetti: Welcome everyone to Science on the Hill.
Mirsky: And that is Scientific American senior editor Mark Fischetti.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Fischetti:
Today we're holding a special panel discussion about the future of our climate, hosted by Scientific American and Nature Research, which are both part of Springer Nature, with support from the National Academies of Sciences, Engineering and Medicine, and especially support from California Congressman Jerry McNerney, who has been our partner for all four of our Science on the Hill presentations.
Mirsky:
We usually go to Capitol Hill once or twice a year for panel discussions about scientific issues facing the country. Because of COVID, we did this latest Science on the Hill panel with all the participants reporting in from their homes. Before we get to that panel discussion, here’s a brief conversation between Fischetti and our congressional Jerry McNerney.
Fischetti:
Climate change affects the economy, energy, agriculture, emergency planning, public health. There are certainly some disagreements about the extent of climate change and the pace of climate change. But even with that is, is it possible to have a unified federal effort to do what we can to protect the country and the planet?
Jerry McNerney:
I think it is. And in the past, there has, that has been the case. And I mean, the country has always had a, a political system with sides that disagree with each other. And yet we've been able to step up significantly say in in the great Wars and we've been able to step up in the depression and I think that's needed. Now we need a, a federal effort that is a matter of consensus that people can get behind that the federal government is willing to fund and puts leadership and confidence into the American people that we're going to be able to overcome this. And right now I don't think we have that. And unfortunately it's setting us back, especially with regard to the pandemic, but also it's prevented any sort of realistic, important action to prevent the climate change from, from continuing its path.
Fischetti:
So given that context what one or two clear messages can representatives and senators provide to their constituents about how the future of climate change will affect them?
McNerney:
Well, I think the one message that's so important is how serious the situation is. I mean, just because the climate change is acting slowly doesn't mean that its not gaining momentum. It doesn't mean that it's not opposing a serious threat to our wellbeing. And so that's important. We need to take action to slow and ultimately hot climate change. And I think it can be, the next message would be that it would be a significant opportunity to transform the, because enemy and create employment and create a sustainable environment. And, and and other very beneficial side effects if we continue and think this thing through take reasonable steps. You know, in creating efficiency and creating a new energy technologies and taking advantage of technology telecommute more. I mean, these, these things make sense. They're, they're doable. And I think if we work together, we can make him happen. Another thing that's really important is that the younger generation of Americans are very concerned about climate change and they're, they're not going to be ignored. I mean, they, they are demanding to be heard on this issue. So the longer we ignore the younger generation, the bigger price there's going to be to pay on this. So I think there's a lot of reasons that we need to move forward and there's some good messages that will help help us do that.
Fischetti:
Is there anything else you wanted to raise or address?
McNerney:
Sure. You know, the thing is climate change and the pandemic are, climate change is not necessarily causing the pandemic. I mean, they seem to be independent, but there's an underlying issue that, that is fueling both of these. And that is the human beings are encroaching further and further into nature, and pushing harder and harder against nature. And of course, nature's going to push back. It's inevitable that it will. So what we need to do is pay attention to what our, our ecosystems are telling us, you know, is this if we add carbon to the atmosphere, if we burn more fossil fuels you know, we're gonna, we're going to see a pushback from nature. If we incurred further and further into wild ecosystems, you know, they're going topush back in their own way too. So let's step back and examine, you know, what effect our species is having on the planet and how we can develop a sustainable future. And that will be something that we can all in our, our descendants can thrive in. So it's a big challenge. We're not, we're not there yet. But I think these kinds of events are going to awaken the public to the need to start thinking in those terms.
Mirsky:
Now on to the panel. Mark Fischetti was joined by three senior scientists who have worked extensively with climate models. Robert Kopp, a climate scientist and director of the Rutgers Institute of Earth, Ocean and Atmospheric Science at Rutgers University; Katherine Calvin, an earth scientist and specialist in human and earth system modeling at the Pacific Northwest Laboratories Joint Global Change Research Institute; and Claudia Tebaldi, also an earth scientist with the Joint Global Change Research Institute. Like Bob and Kate, Claudia is a lead author for the upcoming climate assessment report from the Intergovernmental Panel on Climate Change. Here’s Mark Fischetti:
Fischetti:
Our ability to predict and understand our future climate depends on scientific climate models. We hear a lot of scenarios about how much higher temperature will go, how much higher sea levels will rise, how much worse storms will get, and other serious threats to our lives and our jobs, such as deadly heat waves and crippling droughts. How should you, our audience, interpret the projections you hear? What can scientists say for sure about projections from the climate models and what's less certain? And what can U.S. Representatives and Senators say to their constituents about how the future of climate change will affect them? Bob, to kick us off, can you briefly lay out the range of climate projections that we hear about and introduce us to a few of the landmark reports that are based on those projections?
Robert Kopp:
Sure. When we think about climate projections, there's really three parts to the problem. So first, what are humanity's greenhouse gas emissions going to be? The uncertainty in things like technological change, economic change and policy change is the largest driver in the spread of scientific projections of future climate beyond the next couple of decades. Because we can't predict these changes. All we can do is look at scenarios. And Kate is one of the world experts on the integrated assessment models used to help explore these scenarios, so you'll hear more about this from her.
The second problem is how the climate system, both sort of at a global average but also locally in specific places, is going to respond to these greenhouse gas emissions. And the key tools here are called global climate models or sometimes earth system models or sometimes people will just call these climate models and mean these particular types of models. And these models represent our understanding of the physics and the chemistry of the climate system and have roots going back to the 19th century.
When people talk about climate models, this is usually what they mean, the models of the physical system. And the third problem is how changes in the climate, how changes in heat, changes in humidity, changes in rainfall, changes in sea level affect people and the things we care about like public health, the economy, ecosystems, infrastructure. And there's a variety of tools here that go under the label of impact models. And I think we're going to talk more about those later. The big assessment reports, things like the intergovernmental panel on climate change, which all three of us are lead authors on and has its next big report in 2021 or the U.S. national climate assessment too, which I think we've all contributed as well and has the next one due in 2022, these draw upon the peer reviewed literature, which includes all of these modeling tools and many other lines of informations, whether it's geology telling us about past warm climate, instrumental observations telling us about how things have changed over the last couple of centuries, the basic physics.
And they use all these to characterize the current scientific understanding of how the climate's changed so far and how likely it is to change under different scenarios. And so you asked about the range of climate predictions and, you know, it's a broad range because of particularly the uncertainty in the emissions. But there's a basic common lesson that has emerged over the last decade and a half. The rise in global average temperature caused by carbon dioxide is roughly proportional to the cumulative emissions of carbon dioxide. So that means every ton of CO2 we emit makes the planet a little warmer for centuries to come, most likely by about one degree Fahrenheit for every trillion tons of carbon dioxide we emit. So right now we're emitting enough carbon dioxide—and we're emitting it fast enough—to raise the planetary temperature by about one degree Fahrenheit in a quarter of a century. And the only way to stabilize the climate is to stop.
Fischetti:
Great, thanks. Claudia, can you distinguish between predictions and projections and briefly explain the role that models play in creating the projections.
Claudia Tebaldi:
Thank you, Mark. And thank you, Bob, for this great introduction. There is a fundamental difference between the way we think of future climate and therefore we talk about projections and the way we think about future weather and in that region we like to use the word prediction. And the way you want to try and describe this is thinking of when we look at weather forecasts. We open an app like weather.com or the national weather service website and we look at what's going to happen tomorrow or the next days. And we see, you know, forecast, that say it's going to be cloudy with a certain chance of rain or it's going to be sunny and the temperature is going to be so and so. And we don't find the disclaimer at the end of these pages that says, by the way, be careful because everything could change radically depending on how much traffic there is going to be in your city tomorrow.
That's the fundamental difference when we look at climate projections, where the external conditions that are driving changes in climate can change dramatically depending on our choices and our policies. And so what we say about climate futures needs to have the disclaimer, needs to have a little asterisk that says, here is what my climate simulation is telling you. But this is based on assumptions about our greenhouse gas emissions, about trees that we plan to or that we cut down. And depending on those assumptions here is our trajectory of climate future. But that can change radically if we decide to curb those emissions, if we decide to curb our pollution, if we decide to plant trees rather than cutting them down. And so that's the fundamental difference between projections that are conditional on assumptions about our own actions and policies and predictions that are unconditional. The second part of the question was about the role of models in creating these projections.
The point here is that our climate system is complex and what we're going to do to change the trajectory of climate can vary widely. And the effect of those forcers is very difficult to just, you know, guess in terms of extrapolating current trends or running simple statistical models. We need very complex and comprehensive models of the earth system to produce the result of our, you know, greenhouse gas emissions trajectories and all the other external forces that can change in the future. And so climate models are our best way to project ahead the changes in our system with our best understanding of the science behind it. But I also want to make the point that, you know, if you're interested in very short term, sometimes you don't need climate models to do some, you know, robust policy study about adaptation or a mitigation. Sometimes if you're interested in just the next few years what's happening right now and what happened in the recent past can be enough to guide your decisions.
Fischetti:
Thanks, Claudia. Kate, can you, can you tell us more about the considerations that go into the models that forecast future climate, and tell us what comes out of the models.
Kate Calvin:
Yeah. Thank you, Mark. And thank you Bob and Claudia for setting that up. So as Bob and Claudia both mentioned, these climate projections, they're based on emission scenarios. And so what goes into these models is an assumption about how emissions might change in the future. And what you get out are estimates of changes in the earth system, like changes in temperature, precipitation, and carbon, and those can be used to quantify the impacts of climate change.
The emissions scenarios that go in are looking at how the world's energy, agriculture and land systems might evolve in the future. And as Bob mentioned, these are coming from models too. And it's for the same reasons Claudia mentioned with climate, that the interactions between energy, agriculture and land, and emissions, they're really complex. And so we use a model to help us understand those relationships. But as you get further into the future, how those changes might evolve depends a lot on uncertainties and things we do.
So the scenarios we use in climate models, they differ in terms of their population, their income, behavior, technology and policy, all of which affect emissions in the future. So, just as an example, there are scenarios out there where population continues to grow, reaching 12 billion in 2100. And there are scenarios where population peaks mid-century and declines after 2050. There's scenarios with rapid declines in the cost of renewables and there are scenarios with more modest declines in those costs and all of those factors play out. But the biggest difference between the high and low emissions scenarios that climate models use is policy. So the incentives and regulations that we include in a scenario have a significant effect on the future emissions.
Fischetti:
Can you give us just one example of a policy measure that's been enacted or has been discussed, and how that might affect the output of a model?
Calvin:
Yeah, so if you're thinking about ones that have been discussed or enacted, ones that have been enacted in the past, there's things like renewable portfolio standards that incentivize renewables and that'll affect the deployment of that. A lot of what ends up being used in the models, particularly when you get towards 2100 are carbon prices. And there are some examples of carbon prices in the world. There are some examples of discussions of carbon prices in the U.S. but that's the tool that's most often used in a model in order to change emissions in the future, and it provides an incentive towards lower carbon fuels and options.
Fischetti:
Great. Claudia, we we're going to get into this more, but can you give us a little bit of sense, you know, of how closely the models agree or don't agree. I mean, there are different models run on different, many different factors such as we’ve already heard. And we'll talk about scenarios in a moment, but can you just give us some sort of general sense about that?
Tebaldi:
Yes. I hate to start my answer by saying it's complicated. It is a little bit complicated in the sense that it really depends on the scale at which you're looking, at, you know, the outcomes, or you know, if you go from global average to very local changes chances are that the models will agree much better on the global average than the local changes. If you're looking at something relatively simple to project ahead like temperature and instead you're looking at something more complex like humidity or soil moisture chances are that models are going to agree better on the temperature then the more complex quantities. But the point is that, you know, models agree on the general behavior of the climate system. So there is no model that says we are going to cool rather than warm. They also agree on the spatial pattern of change.
Tebaldi:
So models have been very robust over the last few decades actually, being able to pinpoint those areas of the world that are going to warm more than others, for example. When it comes to things like precipitation models agree at fairly consistent large scales on the area of the world that are going to see increases in average precipitation and the area of the world that are going to be hotspots for things like drought and the consequences of drought. Like for example, wildfires. So models agree on the general direction of change, agree on where those changes are going to be more or less severe. When it comes to actual numbers, models still disagree or at least give us a range that will have to be taken into account for any kind of adapation policy. But we have a lot of information from this model that is actionable.
Fischetti:
Thanks Claudia. So it sounds like there's a range of scenarios and we do hear about certain scenarios a lot such as business as usual or if we are really aggressive about reducing carbon emissions. Can, can one of you sort of give us the general range of what the model scenarios are?
Calvin:
Sure. I can start. And so, the scenarios that we run in climate models, we've designed them to span a range of plausible outcomes. You want sort of to be able to understand what might be possible or plausible in 2100. So they're designed specifically in that way. And they're, you know, we want a small number. Running an earth system model takes a lot of computer time and a lot of people time. So we have a small number spaced evenly among that. But it's again, we've designed them to span a range. And if you look at sort of the, the high end and the low end, they're, they're very different in terms of the things I mentioned before, population income, behavior, technology and policy, and the big differences tend to be, you know, like the high end one is going to have a lot of fossil fuel emissions and the low end is going to have a transition towards low or no carbon energy systems. I would say that there's, it's important to note there's many paths to the same level of emissions. So while we've given the climate models one particular path to high or low emissions, there are other ways to get there.
Fischetti:
And what do we mean by business as usual? We hear that often.
Calvin:
So I actually don't like the term business as usual. I think it's a bit imprecise, like what do we mean? Do we mean keeping the economy as it is today? Do we mean extrapolating our technologies or policies? So I think it's, it's nonspecific. In the, in the scientific community, we more sort of talk about what's assumed in there. So are we assuming continued trends in population or technology? Are we assuming that there is climate policy or there isn't. And we'll often also talk about scenarios based on something related to their carbon emission or their potential warming. And so that's the way we more talk about it. I do see businesses as usual used a lot. I just don't like the term because I don't think it's very specific.
Kopp:
Yeah. If I could add one point on there. Is it because of this complexity and the challenges defining what is business as usual? I think we've, we've seen a trend over the last few years in the climate science literature to also thinking about looking at things simply as a function of, of global mean temperature more, more commonly. So, we can talk about, well, what does a world look like with two degrees Celsius or 3.6 degrees Fahrenheit or three degrees Celsius or four degrees Celsius of warming. And one of the advantages of that is for many sorts of climate parameters it works quite well. You know, you can look at things like temperature or humidity or rainfall projections in the 21st century as a function of temperature and you get a pretty consistent story from a particular model. And then that allows us to recognize while there are in fact, as Kate was saying, multiple scenarios, multiple pathways of policy and technology that would get us to two degrees Celsius or three degrees Celsius or four degrees Celsius at different points in time in the 21st century. So it's sometimes useful to just think about, okay, well what does a two degree C world look like? What does a three degree C world look like? And not get too lost in the details of what does the particular scenario that got you there.
Fischetti:
Can you tie that back though to, you know, what we typically hear about is okay, if we don't do anything differently than what we're doing now to lower emissions versus if we are really aggressive and not only stop emitting carbon dioxide as much as possible to start drawing, you know, drawing down, taking it out of the atmosphere to get us to lower levels—do the scenarios kind of line up with that at all. And if so, can you give us some sense of that?
Calvin:
Yeah, I would say I think the point Bob is trying to make and is it, you know, for a given level of emissions, there's a relationship between emissions and temperature that he mentioned earlier. And then there's a relationship between temperature and impacts. But when you're going for that first step to get that level of emissions, that's where you've got multiple options. You can think about it. You could either change energy demand or energy efficiency and reduce emissions that way. Or you can change the way you supply energy. So still the same amount of energy supply, but you’re changing the ratio of carbon to non-carbon fuels and you'll get the same level of emissions. And so the idea that there's multiple paths to the same emissions or temperature level doesn't change the fact that emissions are related to temperatures, so if you want to reduce temperature, you have to reduce emissions. There's just multiple ways to do that. And as I alluded to, some of the way that scenarios are designed are around these temperature targets. So we want to limit temperature to two degrees. The climate science says that this is the emissions you can allow for that. What are different ways to reach those levels of emissions?
Kopp:
And one thing I might add here is it's actually sort of helpful thinking in that framework because we, the low bound of scenarios is easier to define than the high bound. The high bound is sort of, well what if economy, economics and technological development and policy sort of continue along their current trajectory? And that's a pretty hard thing to explain, that the low bound is saying, well what if we want to limit warming to one and a half degrees Celsius or two degrees Celsius or three degrees Celsius. Well, then the physics tells you you have to get global greenhouse gas emissions to net zero and they tell you roughly what timeframe you have to do it in. So if we want to limit global warming to less than one and a half degrees Celsius, basically we have to get to net zero emissions by about 2050. If we want to limit it to two degrees Celsius, we have to do it by about the 2070s. And there's some fudging on that based on how sensitive the climate is to warming and how much negative emissions you might want to have afterwards. But basically, you know, that's sort of easier to define in many ways than what business as usual is, because there's a physical constraint there.
Fischetti:
Yeah. I want to get back to this a one and a half degrees C versus two degrees C. Before we get there, do the models account for surprises? I know this gets into sort of the economic theory of extreme events or catastrophes. And I don't know that we need to get into all that, but are they, are the models considered generally overestimating, underestimating? There seems to be more discussion lately that maybe they've underestimated the effects.
Kopp:
I can talk about this. We talked about this. There's a whole chapter in the fourth National Climate Assessment that Katharine Hayhoe and I led looking at this idea of potential surprises. And I would say there's sort of two basic categories of potential surprises we might worry about. So one involves positive feedback, self-reinforcing cycles that can accelerate human-caused climate change and even shift the climate into dramatically different States. For instance, states with drastically smaller ice sheets or different, large scale ocean and atmospheric circulation. To some extent, elements of this are representative in the climate models, but to a significant extent they're not. For example, at a most basic level most earth system models do not have coupled ice sheets. And so all sorts of feedbacks involving how changes in ice sheets affect the ocean and the atmosphere and changes in the ocean and atmosphere affect the ice sheets aren't there.
And the geological record suggests that there are important gaps in climate models. Climate models have a systematic tendency to underestimate how warm sort of the past warm climates revealed by geology are. Were, I should say. And this suggests that climate models might be missing important feedbacks that come into play as you heat up the planet. And so climate models may be more likely to underestimate than to overestimate warming. That said, some in the most current generation of climate models seem to have a pretty high sensitivity to warming. So that might not be true of those models. So perhaps we could refine that to saying is that the models that do the best jobs of explaining the past might be underestimating what's going to happen in the future because they're missing some of the things that seem to become important when we look to the geological record.
There's another sort of surprise that’s worth mentioning that we also talked about in the National Climate Assessment, which is called a compound extreme. So that's like where you have multiple heat waves or droughts happening in different places around the world at the same time or in close succession. And you can imagine like if you have droughts in multiple breadbasket regions of the world, for instance, at the same time, that's going to have an effect on humans that is larger than if it were just one and then another and then another sequentially. Because having all of those disasters happening at once adds up to more to the sum of the parts. And those to a large extent are in the climate models, but systematically studying them is something relatively new. And we're kind of living through one of these compound extremes in the most extreme case right now, right? We're having disasters all around the world right now caused not by climate change, but by COVID. And I think this is sort of driving home like how vulnerable some of our fundamental social systems are to having disasters that you might be okay for it for just happening in Wuhan or just happening in New York. But when it starts happening everywhere at once, the impact is considerably larger than the sum of the parts.
Fischetti:
So we're talking about impact here. I'm like, I'd like to talk a little bit about agriculture and labor in particular. Let's start with agriculture since that seems much more directly tied to climate. What are the models telling us, or what are they causing us to think about as far as agriculture goes?
Calvin:
So I mean, to borrow a phrase from Claudia, it's a bit complicated. So what they do tell you, so they're basically what goes into a crop growing, and there's a lot of different factors that are important. Some of it has to do with the temperature. Some has to do with water availability and precipitation. Some of it has to do with nutrients and fertilization. And so when you're looking at an individual crop, whether climate change is going to make its yield higher or lower, it depends on the crop. It depends on where it's grown. It depends on the magnitude of the temperature rise. It depends on how much, how precipitation might change. So what you see though is, you know, there's a certain level of, of warming where it could be beneficial for crops in some places, particularly if you account for the fact that increased CO2 might actually help with fertilizing those crops. But above a certain threshold, all crops have heat tolerance.
And so once you start to get above certain levels of warming yields will decline. And then if you throw in the fact that changes in precipitation are also tied to changes in climate, you might actually have a decline in yields because of precipitation effects. And so what you tend to see is when you look at risks of climate change to crop yields and to agriculture, every degree of warming increases those risks. Increased risks of extreme heat events, you start to move things outside of their adaptive tolerance, and declines and yields do have effects on agricultural prices. Again, it's complicated. It's not uniform. There will be some places in some crops where the yields might actually go up because of climate change. But there are places where they'll go down as well. And that's what the literature tells us right now.
Fischetti:
I brought up agriculture. Is there another sort of segment of society that, that any of you would like to address in terms of major effects that the impact models show?
Kopp:
Sure. So I mean another major one, and perhaps really that the largest one from, from many perspectives are the human health impacts, right? So when we think about human health impacts of climate change, there, there are sort of two effects, right? So reduced exposure to cold can be beneficial, increased exposure to heat can be harmful. And so we can try to figure out how much, what those relationships are and how strong they are. And I am actually part of a collaboration called the Climate Impact Lab, which is a partnership of number of schools. And our focus is on using big data sets to try to learn from past experience. Things like how does, for human health, how does exposure to an additional day of 95 degrees Fahrenheit affect mortality versus exposure to a day of say 70 degrees Fahrenheit? And how does that differ in regions of different income across the world?
And how does it differ in regions that have different sort of climates, different, different past experience? And how much does it cost for say Seattle, which has a cooler climate, to start to adopt some of the behaviors and say air conditioning and technologies to have it a more flat response like somewhere like Houston does. And so, you know, the, the pattern that emerges tells us that if people sort of are adapting through the same approaches they've adapted to over the last 20 or 30 years that we can, that we can see in the record, then unchecked climate change could be as fatal worldwide as infectious diseases are in a typical year right now. And even moderate emissions reductions can have a huge effect in lowering that risk, by about a factor of five. And also that these risks are, are extremely unevenly distributed, right? So low income populations tend to die as a result of the health effects of say, extreme heat. Whereas wealthier places, people tend to spend a whole bunch of money to avoid dying. So it's, it's costly in both places, but in one way it's costly in human lives and the other, it's costly in cash.
Fischetti:
Claudia?
Tebaldi:
As Bob was talking, first of all, if it's really difficult not to realize the uncanny similarities with what we are experiencing right now in terms of COVID. But what I wanted to, to mention, it's also something that makes me think of the COVID experience, which is, it's one of the things that is probably most difficult to model when we look at the future is the adaptation aspect. How are people going to adapt and, and what is that going to do in terms of, you know, limiting the damage of climate change. And of course, adapting will be easier if the pace of climate change is slower. Same way as, you know, it is important to flatten the curve to, to let hospital, you know, deal with the influx of, of patients. So we mentioned before the idea of looking at what happens at two degrees, three degrees, four degrees of warming. But we also need to remember that the pace of warming is important to keep in mind. And that's where scenarios actually bring, you know, additional information, because scenarios are a consistent trajectory into the future that accounts both for the level of warming and the pace at which it's reached.
Fischetti:
That's a really good analogy in terms of the pace of things. We've brought this up a few times now, 1.5 degrees C, two degrees C. So let's talk about that a little more directly. What I think people have heard and maybe get a little confused about is this idea that society is supposedly already surpassed its ability to limit warming to one and a half degrees C or even two degrees C. First, can we talk about that briefly and then if there's some truth to that, then why should we keep trying? There's a sort of fatalistic side of this, but let's, let's start first with whether we've surpassed this and what models tell us about all that.
Calvin:
Yeah, so what I would say, it depends a bit on how you define 1.5 degrees C. But what the scientists tell us right now is it's still technically possible to get back to 1.5 degrees Celsius. There's generally, we're probably going to go above that for a little bit. But it is technically possible. But every year matters in how hard that is and whether or not we get there.
Fischetti:
Every year matters. Can you tell us what you mean?
Calvin:
Yeah, so every year, so carbon dioxide emissions stay in the atmosphere for a long time. We've mentioned that before. So every year that we emit, we get more carbon in the atmosphere, more warming. At the same time, a lot of the emissions-intensive infrastructure that we build, like power plants, lasts a long time. And so the more years of this infrastructure that we build, the more costly it could be to reduce emissions. So each year of emitting results in more warming and more costly reductions. And that means more impacts and some of the impacts of climate change could actually make it harder to mitigate. And so some of these impacts we talked about, you could change the, how much carbon you store in land. And so the less carbon you store in land, the more you have to reduce emissions elsewhere. And so the change in climate can actually make it harder to mitigate.
Tebaldi:
I'm here thinking, you know, that these warming targets are important and are motivating and are good icons of, you know, our, our aspirations. But it's important also to remember that they are sort of magic numbers that don't mean necessarily at all. And if we get over 1.5 by tenths of a degree it’s not going to be you know, a failure. If you, if we get to 2.1 rather than two, it's not going to be a failure. It's important to aim for these low targets and, and the lower we go and the slower we go, the better for us. But it's, I find it sometimes a little bit detrimental to focus too much on the round numbers, be it temperature or the year by which, you know, we have to get to negative emissions. And I'm sure when Bob mentioned 2050, 2070, it didn't mean that if by 2050 exactly we are not going to be at negative emissions we are all going todie. But yeah, let's, let's remember it's a continuum of impacts and effects. And the more we curb, the more we limit, the better we are off.
Kopp:
Yeah. If I could chime in and sort of emphasize that. I mean, you hear a lot of people say, well, we have a decade left to avoid catastrophic climate change. And I think the people who are saying that, you know, comes from a good place and there's a sense in which it's true, right? If you want to hit those temperature targets, we need to have, you know, get off a trajectory of slow emissions growth and onto a trajectory of substantial emissions reduction in this decade. But as Claudia was saying there's not like a cliff out there, at least not a well identifiable cliff in the climate, at temperature below which everything's fine and above which we’re devastated. And you certainly see confusion about that. I think we've seen more and more confusion about that over the couple of years. Because we've been successful in communicating the idea that climate change is harmful and more climate change is more harmful.
But have done it to some extent with these targets. I mean, the important point is that climate change causes an accumulation of harms and every ton we emit causes a little bit more damage than the last ton. So if we overshoot as in your question, if we overshoot one and a half degrees Celsius or two degrees Celsius, there's still a huge value in getting towards net zero emissions and a stable climate. And just as as one example of that, we had a paper out earlier this year, it was led by a postdoc in my group, Dawei Li. And he looked at the number of people who would be exposed to extremely hot humid days, the sort that the army would call a black flag day, under different levels of global warming. Assuming that people were distributed around the world the way they are now.
So not looking primarily at population changes, just taking that as fixed. So right now we have a global average temperature of a little over one degree Celsius or about two degrees Fahrenheit above the late 19th century. And there are about 275 million people who experience a black flag day around the world in a typical year. If we get to one and a half degrees Celsius that grows to almost twice that to five over 500 million. If we get to two degrees Celsius, that grows to almost 800 million. If we grow to three degrees Celsius, it continues growing to about 1.2 billions. Right? So wherever you are on that trajectory, you get a benefit from stopping continuing along that trajectory and getting to net zero in a stable climate.
Calvin:
Hi, I was going to say something similar to what Claudia and Bob said that, you know, we talk about precise numbers like 1.5 because we're using them in a computer model where you have to program in a precise number. But there is a lot of, you know, there's a whole range in between the numbers we choose. And so if you think back on the net zero emissions target that Bob mentioned for 1.5 C, it’s around 2050. If we don't get to net zero by 2050 then there's a lot of options beyond that, right? So one of which is, you know for two degrees C it's net zero by 2075 so that's 25 years later, it’s a half a degree more warming. There's also solutions and options that you can reverse some of that, those additional emissions later. They're not proven at scale for a lot of them, but they do exist in there and so it's possible to not meet these precise targets and still get back to them or to go above them. And I, and I don't know that 1.6 degrees Celsius might not look a lot different than 1.5, but we do know that 2.5 would look different than 1.5. And so there are levels at which you would really notice and others where we've chosen a precise number because we're using a precise computer program.
Kopp:
One more point. It's really helpful I think for, for thinking about what are the impacts of climate change, to think on these different temperature targets. For policy I’m not sure it's that helpful, right? So Paris Agreement had two degrees, well below two degrees Celsius, as close as possible to one and a half degree Celsius. But I think the more important goal in there was really getting the, the for the first time in the international discussion, talking about getting to net zero emissions in the second half of the century. Because emissions are something that are a direct output of our activities. That's something we have as a society, as policy measures, that can pretty directly control, right. To go from emissions to temperature there's a bunch of steps that we don't have control over. And so, you know, thinking about when we get to net zero and, and what other complementary tools we might deploy thereafter, that that sort of puts things I think closer to the things that we actually are able to shape.
Fischetti:
Just a quick question about that Bob. So the things we can control, as you said, emissions. What are one or two of the major things we can't control, and how do you take account of that in predictions from the model?
Kopp:
Yeah. So, well, we put some carbon dioxide into the atmosphere, right? That's the thing we control. And then it causes a bunch of feedback. So those feedbacks may vary how strong the warming is. For instance, how do clouds respond to a change in carbon dioxide concentrations and the associated warming? That's one of the big things that causes a spread in what the model different models say, what the response of clouds are. How do forest respond? How does the Arctic respond, right? These, these, which can be additional sources of carbon dioxide and methane, right? Those affect how much warming you get. And we don't get to control exactly how the forests respond to warming or how the clouds respond to warming. But we do get to control, you know, those emissions that initially caused that warming.
Fischetti:
So every, every half a degree rise matters, every year matters. We talked a little bit about net zero, Kate, you brought that up. Could, could you just explain real briefly what we mean by that and how, you raised the idea that, okay, even if we overshoot, we can, you know, basically draw down carbon dioxide levels to get us back. Could you just sort of address how to think about all those things?
Calvin:
Yeah, so net zero, so if you think about it's net first not gross. So some of the activities we do produce emissions, others take emissions out of the atmosphere. And some of the things that take emissions out of the atmosphere, things like trees growing. So as you grow a tree it absorbs carbon and it removes it from the atmosphere. And so when you're thinking about something like net zero, it basically means that we're balancing all the positive sources of carbon with negative sources of carbon. And I think the easiest one for people to understand is a plant growing. But there are other options out there. So there are some technological solutions where you would just take the carbon out of the air. It's called direct air capture and sequestration. And it's not proven at scale, but there are some demonstration and prototype sorts of options out there that could remove carbon. And so when you talk about net zero, it's balancing positive and negative. And what I said by, if you don't reach net zero by a particular year, then there's potentially options to draw down warming later that would be employing more and more of those sorts of solutions that absorb carbon in the future.
Kopp:
And I can just chime in because there's a confusion I've run across a few times. And it's actually sometimes ambiguous in some of the language used around here. So when we talk about net zero, we mean net zero human-caused emissions. Right. So, so Kate was talking about trees. And so trees in this case would be like trees we cause to grow so that we're expanding the amount of carbon stored in trees to take up some of the carbon we're putting in the air. But of course there are also large sources and sinks of carbon dioxide that are, have nothing to do with us. And it's important when you think about net zero, we focus on the fact the ones that we have control over because you know, if you had net zero but you were including natural sinks that we have no control over, what that means would not be, we would be stabilizing the climate. That means we would be stabilizing the amount of carbon dioxide in the atmosphere and that would lead to a continuously warming climate. So what we need to stabilize the climate is to have net zero human-caused emissions.
Fischetti:
Claudia, this is, this actually makes me think of something you said earlier on about, you know, what goes into the models and, and this point about only being able to control what we can control. So do the models, though, take into account some of these natural causes like thawing permafrost. The, like you know, the, the glaciers that everyone is always focused on. How much of that is taken into account and how much agreement is there about that as well.
Tebaldi:
So every generation of models include more processes representing the entire earth system. And in fact we stopped calling these things climate models and now we call them earth system models because we are attempting to represent all these processes that, you know, represent the carbon cycle like both Bob and Kate were talking about. So we have process representation in our models that try to model how trees, vegetation, soil, cycle the carbon from the atmosphere into themselves and, and out again. We do the same with the oceans that are a big repository of carbon as well. And we have biogeochemistry in the ocean represented. I think Bob mentioned that representing ice sheet is a very difficult enterprise. And so models right now don't include interactive ice sheets. There is an attempt, but the, the level of computational effort in representing these things is, is kind of overwhelming our capacity to run these models for hundreds of years of model time.
Glaciers are represented. And what I would say about agreement is that like every, you know, new effort, it requires time to study and confront, compare and draw conclusions that are robust, that cross models, and others that are not. And we have efforts, international efforts in the community to facilitate that. So we have these couple modeling and comparison projects that prescribe experiments that are standard across climate modeling centers so that the model output can be compared and lessons about what is robust and what needs to be worked out in the community can be drawn.
Kopp:
So one thing as Claudia was talking about the different generations of, of models, right? How, how they keep adding new elements. So they grow into increasingly complex earth system models. And they think this is important to understand. If you look at projections from multiple generations of, of climate models or earth system models over time is that, you know, for certain things we, we, we seem to be narrowing the spread of projections. But for many others the spread is remaining pretty consistently broad. And that's not because we're not learning things, it's because we're taking things, in many cases, we're taking things that were not previously part of the system that we were modeling and bringing them in. So effectively you're converting sort of things that were from the model perspective unknown unknowns into sort of known unknowns. And so your overall assessed spread isn't shrinking that much because you're, you're incorporating more and more features of the real world into the model. And so that's, that's sort of expanding the spread even as we might be learning about some of the other elements that have been in the models for a longer period of time.
Fischetti:
I had two questions I kind of wanted to conclude with and have you all speak to and this is, this is one of those two, which is, how certain are we about all the things we've been talking about here? You know, there's certainly misunderstanding. There's misinformation. I don't want to necessarily get into that so much, but this question of uncertainty is always nagging scientists who are trying to constantly improve it and it also gets used out there in the real world for political purposes and others. So could we, could you just talk about this a little bit?
Calvin:
Yeah, so that's a great question. I think we don't know precise thresholds for different effects right now because of a lot of the things we've mentioned earlier. But we generally know the direction of change. We know kind of what's processes are important and how they might affect the system. So we can't say exactly how warm it'll be if we follow a particular emissions path. But we do have a sense of what's possible in terms of future warming and what matters. And some of the things that we don't know are, are things, cause we're out of sample of our observations, either because we have incomplete observations or because we haven't necessarily seen this in the past. And so some of the questions on how emissions translate into temperature, like will the carbon uptake by land and ocean saturate, at what point, how much more will they absorb and how much carbon will be left in the atmosphere. We do know that higher warming levels result in higher risks to human and natural systems. We know that delaying mitigation makes it more costly. I would say one of the biggest uncertainties right now is what society will do. We have an influence over which emissions path we're on and that makes it, you know, we have a say in our future world. And that makes it very uncertain.
Fischetti:
Claudia, Bob?
Kopp:
I don't think I can say it better than Kate did, but I would say that, you know, all models are incomplete. That's why they're models, right? So that, so no model is perfect, right? If you have wanted to build a perfect model, it would end up being as complex as the real world and take as long as the real world to run. So what models can do is provide powerful insights, right? So they say, well, if we have this representation of core processes, and in the case of climate models, some of those core processes are physics that we've been studying, the scientific community has been studying for well over a hundred years and are very well known. We can say, well, what is the, you know, what is the spread of plausible futures, plausible current conditions we might see. So current generation of climate models does a pretty good job of, for instance, reproducing historical climate changes when they're fed with the historical changes in greenhouse gas and aerosol concentration.
So that means we can have a pretty good degree of confidence if we feed them, say, a counterfactual history without those human emissions. And they consistently tell us that human emissions are responsible for essentially all of the observed global warming since the middle of the 20th century. But as we've talked about a number of times, we do get into these out of sample problems. And you know, if you look at the geological record in many cases, climate models tend to underestimate the warming you see during past warm periods, which perhaps highlights the fact that they're incomplete and maybe some of the processes that they're missing are ones that will become increasingly important if we push the climate system too far out of sort of the realm of, of recent experience against which the models have been developed and, and calibrated. So the geological record provides sort of a key tool because that's the only place we can turn for really looking at out of sample behavior. But the geological record is also noisy and so, so there's issues comparing. But it does give us reason to be concerned that the further we push the climate system, the more likely it is we'll end up getting a larger response than we think we will.
Tebaldi:
I just want to add one thing to all these. And of course I agree with everything that was said. And it's the fact that, you know, we are here, we have been talking about climate models only in the perspective of climate change, but these climate models have been developed to look at processes in our system sometimes independently of the problem of climate change. There is an enormous community out there of climate scientists that use these models to understand our system right now and, and there is an enormous amount of work that goes on to confront these models with observations. So what I mean to say is that this climate models are, you know, subject to an enormous level of scrutiny and development and they are useful to run experiments that we cannot run, of course with the real system, also independently of the problem of climate change.
And, and they, they are a representation of our system that allow us to understand how the system works. And I get a little frustrated when, you know, we received these very facile criticisms about what models are good for or bad for. I would like to show, you know, the thousands of papers that are published almost every ear that probe these models with respect to all sorts of things that don’t even mentioned climate change. So these are really important tools to understand our system. Even if you are a skeptic and you're not worried about climate change. And nevertheless they are, you know tested and tried extensively, and they are understood for their strengths and their weaknesses. And the serious climate scientists don't, don't take them to length that are not appropriate.
Fischetti:
That's a good point. So it's the same very same climate models that are telling you, hey, is El Nino is going to be a strong this this summer? Does that mean it's going to be warmer than usual in certain parts of the country? I mean, models are what creates your daily weather forecast that you see on TV and hear on the radio. It's a good point that it's not as if these models are some sort of isolated exercise used to see what the temperatures are going to be in 2050. It's a still a continuum of work that builds on itself and that we use now every single day for lots of things.
Kopp:
Yeah. And in some cases you know, for instance, NOAA has moved to using the same, what's called the dynamical core of the model, both for weather forecasts and for, for climate projections. Right. So, so, you know, in some cases they're sort of philosophically similar tools, but in some cases the actual code base now is overlapping too, what we use for our short term weather forecast and long-term climate.
Fischetti:
Great. So, so here's the last question. We're trying to address the public and we're trying to address people in Congress who are listening. For them, what one or two bottom-line messages can we give to all of our folks in Congress that they can give to their constituents and other government leaders, to industry people that they talk to about climate change?
Kopp:
So I would say first of all, climate change is real. We humans are responsible for it and it's having damaging impacts on our health and our economy today. And these impacts are only going to get more severe with every ton of greenhouse gas we emit. The only way to stabilize the climate is to bring our net emissions of carbon dioxide to zero and to sharply reduce emissions of other greenhouse gases. The faster we do this the less harmful climate change there's going to be. So just as we look at the last few months, just as with Coronavirus, delaying mitigation runs the risk of dramatically escalating harms.
Calvin:
I can go next. So I would just say when I'm thinking about this, I see a lot of discussions, we mentioned this earlier, that suggests that the decision we have is between very high and very low warming. So like if we don't keep temperatures well below two degrees C, then you know, the world is doomed or that we're committed to high levels of warming. And I think just as we mentioned before, there's a whole range of possibilities in between very high and very low. I also see a lot of discussion about the cost of mitigation and how it’ll impact energy or food prices. But I think it's important to also remember that not mitigating has costs and impacts on prices. And so I would just say if I were to summarize what we know in sort of one statement, I'm borrowing this from the IPCC special report on 1.5, but every degree matters, every year matters, every action matters.
Tebaldi:
Well, I just want to mention one thing that we didn't talk about, which is this idea that now the science allows us to determine how much climate change right now has contributed to the occurrence or the severity of extreme events that impact us directly. You know, be it [Hurricane] Harvey on Houston or be it, you know, the wildfires in the West. So there are analyses out there that are able to tell us how much those events have been made more damaging by climate change. So it is no longer an issue, you know, for polar bears, even if it is an issue for polar bears still. We are experiencing the effects of climate change right now, like Bob said, and we even have the science to tell us by how much climate change is responsible for this. So I guess if you are on the coast of the Atlantic or the Gulf, you may be worried about the next hurricane. If you are in the West, you may be worried about the next wildfire season. And so you should be concerned about climate change because these things have already been linked to climate change. But like Kate was saying, we have the means to start doing something about it and we should start sooner rather than later.
Fischetti:
That's a great point to end on. It's affecting us now in ways we can quantify and we have things we can do to improve. So I want to thank all of you. Bob Kopp at Rutgers, Catherine Calvin at the Pacific Northwest National Laboratory, Claudia Tibaldi also at the Pacific Northwest National Laboratory. Thank you for your time and your insights. And we will continue to talk about this in the months to come.
