Suddenly, I find myself at a loss to understand the underlying questions in the climate discussion. Kemp asks, “could anthropogenic climate change result in worldwide societal collapse or even eventual human extinction?” This begs the question, has abrupt climate change occurred in the past, and, if so, what we the causes? Of course, those answers are, respectively, yes, and “scientists hypothesize…” In other words, we think we know, but we’re not certain. It is that lack of certainty that causes my dilemma.
So now we wonder which scenario is plausible, or implausible. I am an engineer by training, and 50 years of practice, and I believe those questions to be rhetorical rather than practical. I can understand their value from a research perspective, but for society the questions lead to a form of brain-fog that causes us to neglect the essence of climate change. Climate models, and discussion of model outcomes, is beyond the comprehension of a vast majority of the electorate.
From a geologic perspective, the earth’s oxygen-rich atmosphere has occurred for several billion years (http://dx.doi.org/10.1126/sciadv.1600134 ). The statistical probability of climate change is 100 percent, but what is the statistical probability of “abrupt” climate change? Until we can answer that question, we cannot possibly know with certainty if our current anthropogenic activities will hasten or delay the next event.
Again, as an engineer, the questions we need to ask are how can we adapt to the various scenarios. Is a “worst-case” scenario appropriate? I don’t believe it is. I do believe we should appropriate steps to adjust emissions in a manner that does not violate the Iron Law of Climate Change. Shouldn’t we focus on the 100 percent chance item, rather than the 1 in a million or so item?
I read somewhere that there are very real threats to civilization that cause death and destruction daily. We work so hard to protect future generations against a potential threat that we forget about the real world and its wars, and starvation, and misery. Even in affluent and secure countries there is daily high carnage and needless but very real and continuous smoking/drinking and drug-related sufferings and deaths. We argue, as societies, page after page and year after year about hypothetical future risks and spend billions to avert them while all of these ugly, painful, deadly real things are around us daily and accepted as the cost of doing life’s business.
Of course, all of this is strictly my opinion, and expresses more frustration than I had intended. It is meant in no way to discredit or disparage other points of view.
Good post. Your letter was well thought out, concise and tactful.
The original paper "Climate Endgame" was clearly an attempt to debunk your work and the work of others on focusing on reasonable scenarios before it gets traction. Frankly it's their job and in their best interests to promote alarmism and extensive, never ending work on highly unlikely scenarios.
I don't know any of the authors or their qualifications but a quick check shows that they work primarily at places like: "The Center for the Study of Existential Risk" (Oxford), "Centre for Environment, Energy and Natural Resource Governance" (Cambridge), "Center for Health and the Global Environment" (U of Washington), "Future of Humanity Institute" (Oxford), "Fenner School of Environment and Society", The Australian National University. The more fear and anxiety about doomsday they can create makes it easier for them to angle for increased "research" funding and ongoing employment.
Thanks. I think that there is a valid case to make for a focus on extreme scenarios in exploratory research. I do not think that case is to be found in arguments that apocalyptic outcomes are likely (eg, current policies) or significant probability (eg., 5%). Nor is it the case that extreme scenarios are ignored, quite the opposite.
Someone commented on my post: "Without the link to GAT how would you settle on an "appropriate" CO2 concentration?" I respond as follows:
How is GAT measured? How do we know empirically how well we are actually doing? How do we know which GAT is "appropriate"? Realistically, we can't manage the world to achieve some hypothetical GAT. We can't "settle" on an appropriate GAT or an appropriate CO2 concentration. That is just academic. There is no supreme power in the world controlling emissions or GAT. We must simply do the best that we can to reduce emissions.
The problem facing humankind is working in the real economic world to reduce future emissions as much as possible, while providing needed world energy supplies, and supporting a robust world economy. It is not a question of achieving any particular GAT, but rather, how to achieve the best future while maintaining a robust present. Economics, not science will determine that. For example, the 40% of world emissions is emitted by China, Russia and India – heavily based in coal – is one part of the total picture. These countries will continue to emit while it is economically to their advantage. The world needs to focus on emissions and find ways to make it economically advantageous to reduce emissions. If the world reduces emissions as much as it can, that is the best we can do, and whatever the GAT turns out to be, we'll have to live with it. Ah, but you might say: "How will we know how much privation we must endure if we don't know the resulting GAT resulting from any scenario?" Without doomsday predictions, we won't make the sacrifices we need in the present to protect the future." The answer is that much of the world is not tuned into "privation" or "sacrifice" and countries that bear the brunt, regarding themselves as responsible and responsive will only suffer as we now see in Europe. The world is not unified. Emerging nations and rogue nations don't have privation and sacrifice in their dictionaries. The world has to find ways to make it to their advantage to reduce emissions. The best we can do is reduce emissions as much as we can, using CO2 concentration as a measure of our progress, and accept whatever the subjective GAT turns out to be.
For me, KDXL22's third point already raises issues with these bozos' "research integrity." Focusing on the highly unlikely leads to policies that are highly likely to have highly unlikable unintended consequences!
These recent studies are couched in terms of future global average temperature (GAT). There are two steps along the way. Starting with a model for global emissions per year, one step is converting that to increases in atmospheric concentration of CO2 per year. The second step is converting atmospheric concentration of CO2 to increases in GAT. I think that reasonable rough estimates can be made of the change in atmospheric composition for any scenario of future emissions per year. For example, I estimate that each Gt of carbon emitted leads to a ~ 0.25 ppm increase in CO2 concentration if half of the carbon ends up as atmospheric CO2. The conversion of that increase in CO2 concentration to an increase in GAT cannot be done by any tangible algorithm, and while great, complex, global models attempt to do this, they are stymied by the complexity of the earth system as well as the fact that though important, CO2 is not the sole arbiter of the earth's climate. I'm not even sure that one can define what the GAT is? So, I think that the entire field of climate projection for the future is built on shaky foundations. Instead of dealing with ultimate GAT resulting from a future scenario of emissions, we should be content to deal in measurable quantities: the annual emissions, and the CO2 concentration. Each scenario would then be characterized by the expected CO2 concentration say in year 2100, and the goals of various activists might be couched in terms of limiting the concentration in year 2100 to some appropriate figure (e.g. 500 ppm). The problem with characterizing the future in terms of GAT growth is that the link between concentration and GAT is a very shaky one. Emissions and concentration are measurable quantities. GAT is difficult to define, even more difficult to measure, and yet even more difficult to predict from models. We can characterize the sources of emission and develop plans to reduce emissions in key spheres of activity, across many regions and nations, monitoring the CO2 atmospheric concentration as the pulse of our progress, and that provides a rational approach to gradually reducing emissions. But relying on unbelievable estimates of GAT known only to those who run gigantic computer programs adds an element of subjectivity that is not necessary.
I have another suggestion for how to compute the likelihood of a bad outcome defined as observing any one of 1 of 5 events each having a probability of 1%. How about 1-(0.99e5). This would be close to 5%, but not reached by adding them. To me your solution of 0.01e5 is the likelihood of all 5 unlikely events not the likelihood of 1.
They make a cascade, which means the next step is dependent on the outcome of the previous step. In other words, they are presenting a logic that says every step has to be completed in order to reach the final outcome. The only way of treating this mathematically is how Roger presents it. They are probabilities in series, and then you multiply them. This is very basic.
Right. However, based on their “cascade” in order to catastrophe to occur a disaster must occur in each level of the cascade, just as a matter of logic. For instance, start at the top. If emissions are not extreme then there is no basis for the follow on consequences that lead to catastrophe.
Persuasive. I’m a little unclear on their probability of catastrophe cascade though. In their construct are they saying a catastrophic outcome in *any* of those categories inexorably leads to a catastrophe in the whole system?
Yes, see my discussion on Twitter just now with Oliver Morton. I’m not sure any of us understands what they were getting at. In my opinion there is a problem here beyond just communication, but the communication makes it difficult to parse.
I contend though, that we need to go to net zero, in the next couple of hundred years, which is purely commando economy and not optimal in any way. If we stop increasing our CO2 emissions and level out at the existing discharge of 10 GtC, the CO2 will not increase and the temperature very little.
The climate now is good for most things. And we can save a lot of money by still using Coal etc. for heating and transport
I made a mistake. The existing emission of 10GtC is stored: half ind the atmosphere , the other half in the sea and in the biosphere. So we have to go down to 5GTC before CO2 levels out at the existing level. If we continue to emit 10 GT,C the atmospheric content woul likely to go 5-500 ppm .efore it levels out.
One thing that puzzles me is that whatever amount of CO2 we purportedly have emitted, the atmosphere ends up showing an increase that is proportional to half of what we emitted. It was like that in the 1990's and it is like that now, even though we are emitting much more today in terms of absolute emissions. There is something that takes away half of our emissions and leaving the rest for the atmosphere. This something is capable of increasing its uptake of carbon in step with our increased emissions. In all probability, it is increased biological activity that does this but I cannot seem to find any good description of this anywhere.
The thing is, we have only seen this response to a system where our emissions steadily increase. We have never seen the response to a reduction in our emissions, bar what happened in 2020/2021.
To use your numbers, if it holds true today that we emit 10 GtC and the atmosphere absorbs 5GtC of our emissions, does it follow that if we reduce our emissions to 5 GtC then the atmosphere would absord 2.5 GtC? If so, why? Since biological activity seems to be the reason half of the carbon is absorbed elsewhere than the atmosphere, why would it reduce its uptake of carbon in step with our reduction? Why would it not continue to increase uptake, or at least try to stay where it is?
A very good question. The kinetics looks to be a rate parameter times (actual CO2 concentration in the atmosphere - Long term equilibrium = app. 280 ppm)
Suddenly, I find myself at a loss to understand the underlying questions in the climate discussion. Kemp asks, “could anthropogenic climate change result in worldwide societal collapse or even eventual human extinction?” This begs the question, has abrupt climate change occurred in the past, and, if so, what we the causes? Of course, those answers are, respectively, yes, and “scientists hypothesize…” In other words, we think we know, but we’re not certain. It is that lack of certainty that causes my dilemma.
So now we wonder which scenario is plausible, or implausible. I am an engineer by training, and 50 years of practice, and I believe those questions to be rhetorical rather than practical. I can understand their value from a research perspective, but for society the questions lead to a form of brain-fog that causes us to neglect the essence of climate change. Climate models, and discussion of model outcomes, is beyond the comprehension of a vast majority of the electorate.
From a geologic perspective, the earth’s oxygen-rich atmosphere has occurred for several billion years (http://dx.doi.org/10.1126/sciadv.1600134 ). The statistical probability of climate change is 100 percent, but what is the statistical probability of “abrupt” climate change? Until we can answer that question, we cannot possibly know with certainty if our current anthropogenic activities will hasten or delay the next event.
Again, as an engineer, the questions we need to ask are how can we adapt to the various scenarios. Is a “worst-case” scenario appropriate? I don’t believe it is. I do believe we should appropriate steps to adjust emissions in a manner that does not violate the Iron Law of Climate Change. Shouldn’t we focus on the 100 percent chance item, rather than the 1 in a million or so item?
I read somewhere that there are very real threats to civilization that cause death and destruction daily. We work so hard to protect future generations against a potential threat that we forget about the real world and its wars, and starvation, and misery. Even in affluent and secure countries there is daily high carnage and needless but very real and continuous smoking/drinking and drug-related sufferings and deaths. We argue, as societies, page after page and year after year about hypothetical future risks and spend billions to avert them while all of these ugly, painful, deadly real things are around us daily and accepted as the cost of doing life’s business.
Of course, all of this is strictly my opinion, and expresses more frustration than I had intended. It is meant in no way to discredit or disparage other points of view.
Very interesting, thank you.
Good post. Your letter was well thought out, concise and tactful.
The original paper "Climate Endgame" was clearly an attempt to debunk your work and the work of others on focusing on reasonable scenarios before it gets traction. Frankly it's their job and in their best interests to promote alarmism and extensive, never ending work on highly unlikely scenarios.
I don't know any of the authors or their qualifications but a quick check shows that they work primarily at places like: "The Center for the Study of Existential Risk" (Oxford), "Centre for Environment, Energy and Natural Resource Governance" (Cambridge), "Center for Health and the Global Environment" (U of Washington), "Future of Humanity Institute" (Oxford), "Fenner School of Environment and Society", The Australian National University. The more fear and anxiety about doomsday they can create makes it easier for them to angle for increased "research" funding and ongoing employment.
Thanks. I think that there is a valid case to make for a focus on extreme scenarios in exploratory research. I do not think that case is to be found in arguments that apocalyptic outcomes are likely (eg, current policies) or significant probability (eg., 5%). Nor is it the case that extreme scenarios are ignored, quite the opposite.
Looks like Kemp et. al. replied to your letter. They completely ignore the math mistake, which is introductory probability and statistics 101.
It is unclear whether it is a calculation mistake or a mistake in the functional form, but clearly something is not right there!
Someone commented on my post: "Without the link to GAT how would you settle on an "appropriate" CO2 concentration?" I respond as follows:
How is GAT measured? How do we know empirically how well we are actually doing? How do we know which GAT is "appropriate"? Realistically, we can't manage the world to achieve some hypothetical GAT. We can't "settle" on an appropriate GAT or an appropriate CO2 concentration. That is just academic. There is no supreme power in the world controlling emissions or GAT. We must simply do the best that we can to reduce emissions.
The problem facing humankind is working in the real economic world to reduce future emissions as much as possible, while providing needed world energy supplies, and supporting a robust world economy. It is not a question of achieving any particular GAT, but rather, how to achieve the best future while maintaining a robust present. Economics, not science will determine that. For example, the 40% of world emissions is emitted by China, Russia and India – heavily based in coal – is one part of the total picture. These countries will continue to emit while it is economically to their advantage. The world needs to focus on emissions and find ways to make it economically advantageous to reduce emissions. If the world reduces emissions as much as it can, that is the best we can do, and whatever the GAT turns out to be, we'll have to live with it. Ah, but you might say: "How will we know how much privation we must endure if we don't know the resulting GAT resulting from any scenario?" Without doomsday predictions, we won't make the sacrifices we need in the present to protect the future." The answer is that much of the world is not tuned into "privation" or "sacrifice" and countries that bear the brunt, regarding themselves as responsible and responsive will only suffer as we now see in Europe. The world is not unified. Emerging nations and rogue nations don't have privation and sacrifice in their dictionaries. The world has to find ways to make it to their advantage to reduce emissions. The best we can do is reduce emissions as much as we can, using CO2 concentration as a measure of our progress, and accept whatever the subjective GAT turns out to be.
For me, KDXL22's third point already raises issues with these bozos' "research integrity." Focusing on the highly unlikely leads to policies that are highly likely to have highly unlikable unintended consequences!
Hi John, I have a light touch on moderation here, but I'd appreciate if you could edit to remove the name calling. Thanks!
Gee, and I thought I was being overly generous to them. I'll re-submit.
Thank you🙏
These recent studies are couched in terms of future global average temperature (GAT). There are two steps along the way. Starting with a model for global emissions per year, one step is converting that to increases in atmospheric concentration of CO2 per year. The second step is converting atmospheric concentration of CO2 to increases in GAT. I think that reasonable rough estimates can be made of the change in atmospheric composition for any scenario of future emissions per year. For example, I estimate that each Gt of carbon emitted leads to a ~ 0.25 ppm increase in CO2 concentration if half of the carbon ends up as atmospheric CO2. The conversion of that increase in CO2 concentration to an increase in GAT cannot be done by any tangible algorithm, and while great, complex, global models attempt to do this, they are stymied by the complexity of the earth system as well as the fact that though important, CO2 is not the sole arbiter of the earth's climate. I'm not even sure that one can define what the GAT is? So, I think that the entire field of climate projection for the future is built on shaky foundations. Instead of dealing with ultimate GAT resulting from a future scenario of emissions, we should be content to deal in measurable quantities: the annual emissions, and the CO2 concentration. Each scenario would then be characterized by the expected CO2 concentration say in year 2100, and the goals of various activists might be couched in terms of limiting the concentration in year 2100 to some appropriate figure (e.g. 500 ppm). The problem with characterizing the future in terms of GAT growth is that the link between concentration and GAT is a very shaky one. Emissions and concentration are measurable quantities. GAT is difficult to define, even more difficult to measure, and yet even more difficult to predict from models. We can characterize the sources of emission and develop plans to reduce emissions in key spheres of activity, across many regions and nations, monitoring the CO2 atmospheric concentration as the pulse of our progress, and that provides a rational approach to gradually reducing emissions. But relying on unbelievable estimates of GAT known only to those who run gigantic computer programs adds an element of subjectivity that is not necessary.
Without the link to GAT how would you settle on an "appropriate" CO2 concentration?
I have another suggestion for how to compute the likelihood of a bad outcome defined as observing any one of 1 of 5 events each having a probability of 1%. How about 1-(0.99e5). This would be close to 5%, but not reached by adding them. To me your solution of 0.01e5 is the likelihood of all 5 unlikely events not the likelihood of 1.
They make a cascade, which means the next step is dependent on the outcome of the previous step. In other words, they are presenting a logic that says every step has to be completed in order to reach the final outcome. The only way of treating this mathematically is how Roger presents it. They are probabilities in series, and then you multiply them. This is very basic.
Right. However, based on their “cascade” in order to catastrophe to occur a disaster must occur in each level of the cascade, just as a matter of logic. For instance, start at the top. If emissions are not extreme then there is no basis for the follow on consequences that lead to catastrophe.
Persuasive. I’m a little unclear on their probability of catastrophe cascade though. In their construct are they saying a catastrophic outcome in *any* of those categories inexorably leads to a catastrophe in the whole system?
Yes, see my discussion on Twitter just now with Oliver Morton. I’m not sure any of us understands what they were getting at. In my opinion there is a problem here beyond just communication, but the communication makes it difficult to parse.
Hi Roger
Good post:
I contend though, that we need to go to net zero, in the next couple of hundred years, which is purely commando economy and not optimal in any way. If we stop increasing our CO2 emissions and level out at the existing discharge of 10 GtC, the CO2 will not increase and the temperature very little.
The climate now is good for most things. And we can save a lot of money by still using Coal etc. for heating and transport
I made a mistake. The existing emission of 10GtC is stored: half ind the atmosphere , the other half in the sea and in the biosphere. So we have to go down to 5GTC before CO2 levels out at the existing level. If we continue to emit 10 GT,C the atmospheric content woul likely to go 5-500 ppm .efore it levels out.
One thing that puzzles me is that whatever amount of CO2 we purportedly have emitted, the atmosphere ends up showing an increase that is proportional to half of what we emitted. It was like that in the 1990's and it is like that now, even though we are emitting much more today in terms of absolute emissions. There is something that takes away half of our emissions and leaving the rest for the atmosphere. This something is capable of increasing its uptake of carbon in step with our increased emissions. In all probability, it is increased biological activity that does this but I cannot seem to find any good description of this anywhere.
The thing is, we have only seen this response to a system where our emissions steadily increase. We have never seen the response to a reduction in our emissions, bar what happened in 2020/2021.
To use your numbers, if it holds true today that we emit 10 GtC and the atmosphere absorbs 5GtC of our emissions, does it follow that if we reduce our emissions to 5 GtC then the atmosphere would absord 2.5 GtC? If so, why? Since biological activity seems to be the reason half of the carbon is absorbed elsewhere than the atmosphere, why would it reduce its uptake of carbon in step with our reduction? Why would it not continue to increase uptake, or at least try to stay where it is?
A very good question. The kinetics looks to be a rate parameter times (actual CO2 concentration in the atmosphere - Long term equilibrium = app. 280 ppm)
https://www.drroyspencer.com/2019/04/a-simple-model-of-the-atmospheric-co2-budget/