Global warming is one of those topics I've always withheld judgement on because it's such a huge topic that, even ignoring politics, it seems very hard to get a solid understanding of the science.
As IBM and Microsoft were aware, to win with FUD (fear, uncertainty, doubt), they didn't have to convince you to buy their product, they just had to convince you not to buy the competitions', thereby starving the competition out.
Cigarettes, seat belts, lead paint, asbestos, Linux, climate change. They've all been fought using the same tactics.
There is no need to ascribe malice to those that disagree with you about climate change issues. Reasonable people can disagree, especially on an issue that is complex and depends on predictions about future conditions.
It doesn't depend on predictions. Climate change is happening; the models are just to try and figure out how much warmer it's going to get, and how soon.
Most of the real changes would require many people to die.
Going back, worldwide, to a pre-industrial civilisation would be a possibility. However, We're looking at 90% fatality rates to get to that point due to lack of food production. Probably would be even less than that. No chance in happening, unless forced by catastrophic event.
Another possibility is tremendous rationing. We see what having even simple healthcare is in this country. Instead, we would look at rationing everything in the world, including determining who lives and dies by how much resources they use (carbon debt). With 2 high input/output countries, good luck on this happening either.
And we have to look at all the smaller ways to handle global warming. One proposal is to dump megatonnes of lime in the ocean?! Or perhaps this carbon credits scam: it makes the rich richer (Well, where DOES the money go?). At least sources of energy like solar or wind make sense... Until you realise the amount of energy to make a solar panel is greater than the total energy you will get out of it. Bummer.
And underlying is this assumption that global warming == CO2 peak.. I've heard of this law called thermodynamics. CO2 may trap heat, but heat generated anywhere is still heat. We generate heat via our vehicles, electric grid, generators, burn piles, and other human sources of heat. Is this not a direct consideration as well?
A study about a year ago claimed that we could get to half the reductions required to keep us under +2C just by halting all fossil fuel subsidies worldwide (not that that's feasible).
A little googling on solar panel energy cost found, for example, these links:
which claim that energy payback of solar panels takes about three years.
For various reasons I think we'd do better with advanced nuclear, such as integral fast reactors or molten salt reactors. Either would be safer than what we have now, proliferation-resistant, and produce a lot less nuclear waste with a much shorter lifetime. We could have a very energy-rich future without carbon emissions.
As for carbon credits, a lot of economists advocate a "fee-and-dividend" scheme. Instead of trading credits, major source emitters simply pay a flat fee per ton of carbon, passing on the cost to the rest of us. The money is divided among all citizens, equal amount per person. Most people come out ahead (since the median emissions are less than the average), but since the dividend is constant and the new cost variable, everyone has more incentive to conserve.
I've seen calculations for the contribution of waste heat and it's very minor compared to greenhouse gases. However, if our energy usage continues to grow exponentially on this planet it'll be a real problem in a couple hundred years:
http://physics.ucsd.edu/do-the-math/2011/07/galactic-scale-e...
That's complete crap. All it takes is a reasonable investment in technologies like wind and solar.
For example, I'm in Australia. Aluminium smelting here consumes about 1/3 of the electrical energy produced. Mostly that's done with hugely subsidised brown coal. If there were a serious effort to use solar power to smelt Aluminium instead, you'd see a big reduction in the CO2 produced, and it'd likely be cheaper than the current situation.
But there isn't the political will to get it done - primarily because of idiotic arguments like yours.
That's true - but when there's risk of something, you need to buy insurance. Alt. energy, geo-enginnering, and small CO2 taxes are all ways of hedging that we should deploy in some combination. We can't say "don't know, therefore do nothing".
There's also a risk we'll have another ice age. Global warming might be considered insurance against that eventuality. Whatever the perfectly optimum average planetary temperature for human life is, it is exceedingly unlikely that it was achieved in 1990. (And actually seems pretty likely that it's a fair bit warmer than now.)
Do we also need to actively "buy insurance" against meteor impacts?
We, as a society, are really bad a planning for these things collective/non-pressing problems, and the data being hard to come by just adds to this. Your thinking is pragmatic only in a short term sense. Do you really need fire insurance? Only 1:1000 homes ever burn - I guess not then...
The risk for very bad global warming problems (mass despeciation or crop loss, etc. in 100 years) even at 10% is a really bad consequence that requires some form of insurance.
The risk for rock impact in 100 years is about 0.0002%, and we've got people tracking a ton of near earth objects and even some NASA work on how to deflect them. That's probably about $20M in insurance - probably too low, but yes we should care about these things.
It's possible to spend too much on "insurance" and we almost certainly do that when it comes to climate. The general scientific consensus is that the next half-century of warming generally makes humanity better off. So this risk you presumably think we need to be actively trying to mitigate - as opposed to just doing the sort of stuff we do for asteroids - is predicated on the idea that current trends continue for a century. But 50-60 years is quite a long time when it comes to technology. Even if the models are right, by the time global warming starts to be a net negative rather than a net positive, our grandkids will have a lot better options to deal with it then than we do today. Our kids and grandkids will be richer, smarter, better fed (due to longer growing seasons and CO2 fertilization), and MUCH better informed than we are today. The best thing we can do for them is give them an economy that hasn't been crippled due to paranoia that any one PARTICULAR threat we can foresee today will doom their world.
There are an infinite number of potential threats that aren't hurting us now but that we can imagine might hurt our grandkids. Only some of them will actually turn out to be a serious problem. Spending trillions or even billions of dollars of resources combatting any one such threat reduces their potential ability to address all the others.
There's nothing wrong with caring about things like global warming, but there's quite a lot wrong with trying to "fix" it now. To put it in programming terms: "Premature optimization is the root of all evil."
This is a thoughtful response, and shows that we're actually pretty close. The science is mixed on whether humanity is better off I think - there are winners and losers for sure. However, despeciation gets me most as it seems most irreversible, that and whether or not there are other feedback loops that we're not thinking about.
I agree that it's overhyped, but the question here in on the margin. If we could take the insurance payment and put it towards alt./nuclear energy research we'd be killing a few birds.
I don't see how our economy would be crippled by the "insurance" even at $1T in debt funding (remember how big TARP and the Treasury's loans were). Currently, we're putting far far more than this into overspend on the military ($1.5T/year) - which is defending resource and shipping interests far more than anything else. Couldn't we just solve those problems with similar funding?
I agree, and what annoys me is that as a supporter of public-school science education via www.ncse.com and other organizations, I find myself being dragged into debates I didn't want to participate in. NCSE in particular has decided to lump criticisms of climate education in with their original cause, which was to do something about widespread attempts on the part of religious people to subvert public-school biology curricula.
Strong evidence for anthropogenic climate change exists, to the best of my understanding. However, it's roughly 3.4 gazillion times weaker than the evidence for evolution by natural selection. There are reasons why I am actively interested in helping with one fight but not the other... but the organizations that I support as proxies in the fight don't appear to understand or care. I'm not even sure how annoyed I should be about this state of affairs, frankly.
"Let us look at the predictions made in 1990: they predicted an increase in temperature of 0.3C per decade, plus minus 0.1C. Yet less than 0.2C of warming per decade was observed. To be blunt, they got it flat wrong."
How much less than 0.2C was it? 0.3 +/- 0.1 means it can be anywhere from 0.2 to 0.4, so unless it was grossly under 0.2, it wasn't flat wrong.
"there are alternatives to banning oil production: solar radiation management. We might be able to setup space mirrors or otherwise increasing the reflectivity of our atmosphere." ... "But keep in mind that we will soon run out of cheap oil in any case."
In other words, "Fear not, for we might just come up with some miracle technology or something. It might just all work out in the end so just stay the course and don't worry!"
If you're going to call out others for poorly reasoned arguments, at least have well reasoned arguments of your own.
Despite all the green efforts, CO2 output from using fossil fuel still grows. Western economies barely, if at all, made a dent in their fossil fuel consumptions. Upcoming economies like China use more and more fossil fuels. I think its fair to say that we are nowhere near having a solution to reduce global CO2 output.
I think the bigger thing to pull from that quote centers on the word "prediction". The IPCC FAR didn't really make any predictions. Instead, it made a series of projections (twelve of them, to be precise) based on running various sets of hypothetical parameters through a climate model.
Some of those projections were based on very pessimistic scenarios (very high CO2 sensitivity, very high future CO2 emissions). They projected very high increase in temperature. Other projections were based on more optimistic scenarios. They projected a lower increase in temperature.
Now, the number that the author quotes (0.3C +/- 0.1C per decade) corresponds to the range for the block of projections that assumes a very high climate sensitivity - a sensitivity that turns out to be high. Since the projection was based on a scenario that did not correspond to reality, if we assume that the models were reasonably accurate then it should not surprise us that the projection didn't end up corresponding to reality, either. So there's an interesting omission. . . and an example of how what someone doesn't say can say as much about their position as what they do say.
Another omission that speaks volumes is highlighted if you compare the projections for scenarios that ended up being more representative of how those parameters did evolve over the next two decades to what global temperatures have been doing over that time period. Far from the alarmist claims of gloom and doom that the author would like to suggest were made, the projection for the scenario that ended up most closely corresponding to what actually happened slightly underestimated the actual change in temperature. But by an amount that was quite small in comparison to the error bars, so an honest and scientific analysis of the TAR projections would force us to conclude that in this particular respect the IPCC's analysis was actually dead on.
Poorly reasoned arguments are certainly unfortunate. . . but if you really want to make your position look bad, try blatantly cherry-picking the data.
Well, he used the same numbers that the IPCC FAR report's "Policymaker's Summary" section chose to use, which repeatedly uses the term "is predicted" (admittedly, qualified with "based on current model results" or specific scenarios.) The explicit purpose of that section was to suggest possible policy options based on a set of long-term events stated to be found _with certainty_ (their words) by the working group that were going to occur.
Let's say that I am a "policymaker" and not a climate scientist, and I sensibly choose to defer to experts how I should plan for any potential climate change. That's the section that was written specifically to me for that purpose, by the working group--with the same numbers that this guy quotes.
Those cases where the BaU figures are used are disclaimed with the qualifier, "Under the Business-as-Usual scenario", and the ensuing discussion is pretty clearly intended as an exposition on how we should not have wanted to mimic the BaU scenario because, as the model projections for that scenario indicate, there was compelling reason to believe that the consequences would have been bad.
Again, cherry-picking things out of context does not a compelling argument make. Unless the people you're trying to compel can be trusted to not check your sources too carefully.
And of course the corollary to that is, if the topic is controversial you should always thoroughly fact-check your sources.
(Speaking of, I couldn't find the phrase "with certainty" in either of the three FAR policymakers summaries. . . perhaps you meant "with confidence"? Regardless, it's skirting the edge of dishonesty to use the phrase "their words" to describe something that isn't their words.)
"We might be able to setup space mirrors or otherwise increasing the reflectivity of our atmosphere. Obviously, we cannot hope to both keep pumping CO2 in the atmosphere, and correct for it afterward: this could quickly become economically infeasible."
In the context of killing the planet and the entire human race, we still mention a plan could be "economically infeasible".
A run a way process in which we loose all life on earth is not something I have read a lot of proof for. However, adapting to a changed climate may not be a pretty phase for the human race.
>We're not going to "kill the planet," or the entire human race. Humans adapt. That's what we do. That's why we won.
That is very close to the "fine tuning" argument often employed by creationists. It is entirely possible that humans have, in fact, merely been phenomenally lucky to have survived so far, and that the fact that we find ourselves as members of such a phenomenally lucky species is easily explained by the anthropic principle.
If you take all of the intelligent critters out there in the universe that are capable of discussing their adaptability, a pretty big chunk of those critters are going to be members of civilizations that got very lucky over the ages and managed to produce large cumulative populations, while a very small proportion will be critters from short-lived civilizations. Indeed, holding constant whatever level of adaptability we actually have, it is reasonable to assume that we are among the luckier of the civilizations at that level.
So no, we can't just assume that we're going to leap over any hurdle that comes our way.
Except that there are civilisations which have died out due to overexploiting their environment, or the climate changing and killing off their agriculture.
That's two risks. You can add others now which haven't previously been possible, such as disturbing your ecosystems to the point where there's another mass extinction looming, or changing the climate to the point where you spark famine, wars, lack of water, etc.
That's where all of the rational discourse backed by evidence (aka. science) is pointing.
I tend to give people the benefit of the doubt, and read economic arguments like that in the sense of "we will need to spend a large chunk of our civilisation's energy/material output to do this", not "Oh noes! My small, green pieces of paper!"
The thing that I worry about with the economic arguments is that the climate and/or ecosystems will reach a tipping point before we get to the point where we need to do something, and that chunk will be > 100%.
Among the many annoying half-truths of this article is that he doesn't make any attempt to discuss why the predictions from the 1990s were wrong. For example, we now know that dumping of heat and CO2 into the oceans is a bigger effect than originally modeled. In other words, thinking about the big picture, things are worse than was thought (because it means measures like putting up giant sky-mirrors aren't going to reverse committed warming as easily as they might).
Had the OP read the New Yorker article, he might also understand that his references to "forbidden thoughts" are entirely unjustified. The thoughts he expresses are actually quite conventional, and widely discussed in scientific circles.
I always find it troubling when someone with a computer science background assumes that their knowledge naturally transfers to something like climate science. Climate science is about probability and huge multi-factor models - and most importantly - models that change as new knowledge and data appear. It's just my opinion, but that seems to me to be the opposite of what most computer scientists are taught. Obviously computer scientists incorporate new data and new thinking - but I don't think it's quite the same as constructing a statistical model of how climate works. And obviously there are exceptions - but in general, I think it's just a different mindset.
I agree that knowledge of computer science doesn't transfer directly to climate science. However, there's a lot about computer science that may help understand how climate models yield probabilistic outcomes (Monte Carlo, anyone?).
On the subject, be a little careful equating probabilistic models and statistical models. At least in my experience, "statistical models" generally describes statistical fits to past data. Think of a polynomial whose parameters are chosen by least-squares fit to previous observations. These fits can be used to interpolate or extrapolate (to the future, or to different combinations of inputs). However, they don't encode any domain knowledge, and they can be extremely brittle when pushed outside the regimes from which they were derived.
Most climate models, on the other hand, are deterministic, physics-based models. Some of the sub-models may have statistical fits in them, but the modeling is largely driven by some conceptual framework of how the physics drive the dynamic system (e.g., expressing conservation of mass and energy, and how spatial differences drive transport).
Those deterministic climate models can be driven probabilistically-- meaning the uncertain inputs and parameters can be sampled from distributions of likely values. Aggregating all the results gives a probabilistic prediction. A lot of climate reports take this one step further, and aggregate results across multiple models. Most computer scientists will immediately "get" this part, even if they don't understand all the physics represented in the individual packages.
Excellent points - and important not to generalize about model construction.
I have a degree in computer science, and most of my training consisted of solving problems where there was a single outcome - e.g. "does the compiler work?" When I moved into financial modeling, I really had to shift from the idea of one answer to the idea that the real world is exceptionally complex and difficult to model. I had to get comfortable with the idea that I might have an idea of the outcome based on a model, but, as you point out, this is probabilistic outcome.
I found this article to have all the marking of someone with a similar background trying out a different profession without understanding probabilistic modeling, and, as you rightly point out, not understanding the deterministic portion of the models - the domain knowledge.
Moreover, I think he expresses doubt in any model that doesn't turn out the right answer 100% of the time - to me, not understanding the complexity of the systems that are being modeled and definitely not understanding that it is, as the name "climate science" implies, science - you experiment, you gain new knowledge from new data, etc.
38 comments
[ 2.9 ms ] story [ 89.7 ms ] threadThe deniers don't have to be proven correct to prevent action on climate change, they just have to put out enough FUD to create doubt.
Cigarettes, seat belts, lead paint, asbestos, Linux, climate change. They've all been fought using the same tactics.
Going back, worldwide, to a pre-industrial civilisation would be a possibility. However, We're looking at 90% fatality rates to get to that point due to lack of food production. Probably would be even less than that. No chance in happening, unless forced by catastrophic event.
Another possibility is tremendous rationing. We see what having even simple healthcare is in this country. Instead, we would look at rationing everything in the world, including determining who lives and dies by how much resources they use (carbon debt). With 2 high input/output countries, good luck on this happening either.
And we have to look at all the smaller ways to handle global warming. One proposal is to dump megatonnes of lime in the ocean?! Or perhaps this carbon credits scam: it makes the rich richer (Well, where DOES the money go?). At least sources of energy like solar or wind make sense... Until you realise the amount of energy to make a solar panel is greater than the total energy you will get out of it. Bummer.
And underlying is this assumption that global warming == CO2 peak.. I've heard of this law called thermodynamics. CO2 may trap heat, but heat generated anywhere is still heat. We generate heat via our vehicles, electric grid, generators, burn piles, and other human sources of heat. Is this not a direct consideration as well?
A little googling on solar panel energy cost found, for example, these links:
http://solarbus.org/documents/pvpayback.pdf
http://www.theoildrum.com/node/3910
which claim that energy payback of solar panels takes about three years.
For various reasons I think we'd do better with advanced nuclear, such as integral fast reactors or molten salt reactors. Either would be safer than what we have now, proliferation-resistant, and produce a lot less nuclear waste with a much shorter lifetime. We could have a very energy-rich future without carbon emissions.
As for carbon credits, a lot of economists advocate a "fee-and-dividend" scheme. Instead of trading credits, major source emitters simply pay a flat fee per ton of carbon, passing on the cost to the rest of us. The money is divided among all citizens, equal amount per person. Most people come out ahead (since the median emissions are less than the average), but since the dividend is constant and the new cost variable, everyone has more incentive to conserve.
I've seen calculations for the contribution of waste heat and it's very minor compared to greenhouse gases. However, if our energy usage continues to grow exponentially on this planet it'll be a real problem in a couple hundred years: http://physics.ucsd.edu/do-the-math/2011/07/galactic-scale-e...
For example, I'm in Australia. Aluminium smelting here consumes about 1/3 of the electrical energy produced. Mostly that's done with hugely subsidised brown coal. If there were a serious effort to use solar power to smelt Aluminium instead, you'd see a big reduction in the CO2 produced, and it'd likely be cheaper than the current situation.
But there isn't the political will to get it done - primarily because of idiotic arguments like yours.
Do we also need to actively "buy insurance" against meteor impacts?
The risk for very bad global warming problems (mass despeciation or crop loss, etc. in 100 years) even at 10% is a really bad consequence that requires some form of insurance.
The risk for rock impact in 100 years is about 0.0002%, and we've got people tracking a ton of near earth objects and even some NASA work on how to deflect them. That's probably about $20M in insurance - probably too low, but yes we should care about these things.
http://abcnews.go.com/Technology/story?id=98208&page=1
There are an infinite number of potential threats that aren't hurting us now but that we can imagine might hurt our grandkids. Only some of them will actually turn out to be a serious problem. Spending trillions or even billions of dollars of resources combatting any one such threat reduces their potential ability to address all the others.
There's nothing wrong with caring about things like global warming, but there's quite a lot wrong with trying to "fix" it now. To put it in programming terms: "Premature optimization is the root of all evil."
I agree that it's overhyped, but the question here in on the margin. If we could take the insurance payment and put it towards alt./nuclear energy research we'd be killing a few birds.
I don't see how our economy would be crippled by the "insurance" even at $1T in debt funding (remember how big TARP and the Treasury's loans were). Currently, we're putting far far more than this into overspend on the military ($1.5T/year) - which is defending resource and shipping interests far more than anything else. Couldn't we just solve those problems with similar funding?
Strong evidence for anthropogenic climate change exists, to the best of my understanding. However, it's roughly 3.4 gazillion times weaker than the evidence for evolution by natural selection. There are reasons why I am actively interested in helping with one fight but not the other... but the organizations that I support as proxies in the fight don't appear to understand or care. I'm not even sure how annoyed I should be about this state of affairs, frankly.
They produce a nifty State of the Climate report every so often, which is pretty easy to understand. There's one just been released, in fact: http://www.csiro.au/Outcomes/Climate/Understanding/%7E/media...
Disclaimer: I currently work for the Bureau of Meteorology in Melbourne - as a programmer though, not a scientist.
How much less than 0.2C was it? 0.3 +/- 0.1 means it can be anywhere from 0.2 to 0.4, so unless it was grossly under 0.2, it wasn't flat wrong.
"there are alternatives to banning oil production: solar radiation management. We might be able to setup space mirrors or otherwise increasing the reflectivity of our atmosphere." ... "But keep in mind that we will soon run out of cheap oil in any case."
In other words, "Fear not, for we might just come up with some miracle technology or something. It might just all work out in the end so just stay the course and don't worry!"
If you're going to call out others for poorly reasoned arguments, at least have well reasoned arguments of your own.
Some of those projections were based on very pessimistic scenarios (very high CO2 sensitivity, very high future CO2 emissions). They projected very high increase in temperature. Other projections were based on more optimistic scenarios. They projected a lower increase in temperature.
Now, the number that the author quotes (0.3C +/- 0.1C per decade) corresponds to the range for the block of projections that assumes a very high climate sensitivity - a sensitivity that turns out to be high. Since the projection was based on a scenario that did not correspond to reality, if we assume that the models were reasonably accurate then it should not surprise us that the projection didn't end up corresponding to reality, either. So there's an interesting omission. . . and an example of how what someone doesn't say can say as much about their position as what they do say.
Another omission that speaks volumes is highlighted if you compare the projections for scenarios that ended up being more representative of how those parameters did evolve over the next two decades to what global temperatures have been doing over that time period. Far from the alarmist claims of gloom and doom that the author would like to suggest were made, the projection for the scenario that ended up most closely corresponding to what actually happened slightly underestimated the actual change in temperature. But by an amount that was quite small in comparison to the error bars, so an honest and scientific analysis of the TAR projections would force us to conclude that in this particular respect the IPCC's analysis was actually dead on.
Poorly reasoned arguments are certainly unfortunate. . . but if you really want to make your position look bad, try blatantly cherry-picking the data.
Let's say that I am a "policymaker" and not a climate scientist, and I sensibly choose to defer to experts how I should plan for any potential climate change. That's the section that was written specifically to me for that purpose, by the working group--with the same numbers that this guy quotes.
Again, cherry-picking things out of context does not a compelling argument make. Unless the people you're trying to compel can be trusted to not check your sources too carefully.
And of course the corollary to that is, if the topic is controversial you should always thoroughly fact-check your sources.
(Speaking of, I couldn't find the phrase "with certainty" in either of the three FAR policymakers summaries. . . perhaps you meant "with confidence"? Regardless, it's skirting the edge of dishonesty to use the phrase "their words" to describe something that isn't their words.)
"We might be able to setup space mirrors or otherwise increasing the reflectivity of our atmosphere. Obviously, we cannot hope to both keep pumping CO2 in the atmosphere, and correct for it afterward: this could quickly become economically infeasible."
In the context of killing the planet and the entire human race, we still mention a plan could be "economically infeasible".
Jeez.
We're not going to "kill the planet," or the entire human race. Humans adapt. That's what we do. That's why we won.
Such hyperbole endangers vital rational discourse on the subject.
That is very close to the "fine tuning" argument often employed by creationists. It is entirely possible that humans have, in fact, merely been phenomenally lucky to have survived so far, and that the fact that we find ourselves as members of such a phenomenally lucky species is easily explained by the anthropic principle.
If you take all of the intelligent critters out there in the universe that are capable of discussing their adaptability, a pretty big chunk of those critters are going to be members of civilizations that got very lucky over the ages and managed to produce large cumulative populations, while a very small proportion will be critters from short-lived civilizations. Indeed, holding constant whatever level of adaptability we actually have, it is reasonable to assume that we are among the luckier of the civilizations at that level.
So no, we can't just assume that we're going to leap over any hurdle that comes our way.
But when it comes to extinction, we can be pretty sure that's not going to happen.
Is 90% dead much better ? Probably not for the economy.
That's two risks. You can add others now which haven't previously been possible, such as disturbing your ecosystems to the point where there's another mass extinction looming, or changing the climate to the point where you spark famine, wars, lack of water, etc.
That's where all of the rational discourse backed by evidence (aka. science) is pointing.
The thing that I worry about with the economic arguments is that the climate and/or ecosystems will reach a tipping point before we get to the point where we need to do something, and that chunk will be > 100%.
Ultimately, he advocates technological fixes. For a more critical, but highly readable, look at this subject, see Michael Specter's "The Climate Fixers" from the New Yorker this past May: http://www.newyorker.com/reporting/2012/05/14/120514fa_fact_...
Had the OP read the New Yorker article, he might also understand that his references to "forbidden thoughts" are entirely unjustified. The thoughts he expresses are actually quite conventional, and widely discussed in scientific circles.
Seems legit.
On the subject, be a little careful equating probabilistic models and statistical models. At least in my experience, "statistical models" generally describes statistical fits to past data. Think of a polynomial whose parameters are chosen by least-squares fit to previous observations. These fits can be used to interpolate or extrapolate (to the future, or to different combinations of inputs). However, they don't encode any domain knowledge, and they can be extremely brittle when pushed outside the regimes from which they were derived.
Most climate models, on the other hand, are deterministic, physics-based models. Some of the sub-models may have statistical fits in them, but the modeling is largely driven by some conceptual framework of how the physics drive the dynamic system (e.g., expressing conservation of mass and energy, and how spatial differences drive transport).
Those deterministic climate models can be driven probabilistically-- meaning the uncertain inputs and parameters can be sampled from distributions of likely values. Aggregating all the results gives a probabilistic prediction. A lot of climate reports take this one step further, and aggregate results across multiple models. Most computer scientists will immediately "get" this part, even if they don't understand all the physics represented in the individual packages.
I have a degree in computer science, and most of my training consisted of solving problems where there was a single outcome - e.g. "does the compiler work?" When I moved into financial modeling, I really had to shift from the idea of one answer to the idea that the real world is exceptionally complex and difficult to model. I had to get comfortable with the idea that I might have an idea of the outcome based on a model, but, as you point out, this is probabilistic outcome.
I found this article to have all the marking of someone with a similar background trying out a different profession without understanding probabilistic modeling, and, as you rightly point out, not understanding the deterministic portion of the models - the domain knowledge.
Moreover, I think he expresses doubt in any model that doesn't turn out the right answer 100% of the time - to me, not understanding the complexity of the systems that are being modeled and definitely not understanding that it is, as the name "climate science" implies, science - you experiment, you gain new knowledge from new data, etc.