What's up guys, Rosh here, and welcome to another episode of Skeptics Say the Dumbest Things. The video we're looking at today is part of an ongoing collaboration between skeptic sweetheart Naomi Seipz and notorious skeptic and English aristocrat Lord Christopher Monckton. The series is called Climate of Freedom and essentially features Naomi quizzing Lord Monckton about climate science in an attempt to fact-check what she perceives as pseudoscientific climate alarmism. Now, I've put a link to the original video in the description below, and I encourage you all to go and watch it beforehand so you can check if I've fairly represented their views. So Naomi sets the context of this discussion by talking about an article she's recently seen in the Guardian.
So today I found an article in the Guardian claiming that modeling suggests climate is considerably more sensitive to carbon emissions than thought. So they base this claim on new climate models or refined climate models and say that those models show that carbon emissions have an even more detrimental effect on the atmosphere and on our climate. So let's have a look at a very important question.
Who needs climate models anyway? Naomi then goes on to tell us how most of the predictions about man-made global warming are based on climate models which she calls silly little computer games and she rounds off with a question. If all we want to know is how much warming our sins of emission may cause, do we really need climate models at all?
So let's address that question. Do we need climate models at all? Perhaps we should start with an explanation of what climate models are.
Essentially, they are a mathematical representation of the climate system. The various drivers of the climate are represented in the model by complex equations derived from the laws of physics, and they are constantly refined as our understanding of the climate develops. Over the years, more and more climate processes have been added to the models, and the grids they use have become increasingly detailed.
But are they useful? Well, if a model could accurately simulate the climate system, then I think even Naomi would agree that it would be useful. So I suspect what she's really asking here is, are models accurate? It's a common line of argument among climate contrarians that models are woefully poor at simulating the climate. You may well have come across graphs like these, which appear to show how inaccurate climate model projections have been.
But often these diagrams are specifically designed to mislead you. This one, for example, compares model results for global average temperature with some cherry-picked balloon datasets. It deliberately misaligns the data to visually exaggerate small differences. It ignores the uncertainty ranges around the model projections and uses a whole host of other misleading techniques.
Unfortunately, there are many similar graphics bouncing around the blogosphere. But when we honestly compare like with like, observed warming and model projections match up pretty well. We can therefore be confident that they simulate the Earth's climate system sufficiently accurately to be at least partly useful to us. Imperfect? Of course, but they're pretty much the only tools of prediction we have.
So, to answer Naomi's question, do we need climate models at all? Well, If we want to maximise our understanding of the climate system, then yes, yes we do. But let's see what Christopher has to say.
No, Naomi, we don't need models. All we need is real world data. Oh, well, I guess I'm wrong.
But in all seriousness, this assertion that all we need is real world data is interesting. The implication, of course, is that real world data and climate models are mutually exclusive. That somehow the data and equations used in climate models are fabricated out of thin air. But of course, that's not the case. The equations used in climate models are derived from real world data and the laws of physics.
But let's see how Christopher expands on this. We know by measurement that from 1850 to 2020 there's been about 0.9 degrees of global warming, which isn't much. I love how casually Christopher says that. 0.9 degrees may not seem like much, but when talking about the average global temperature, it is significant. The difference between an ice age and a warm period can be as little as 4 or 5 degrees, so 0.9 degrees is like 20% of that.
And as I've said before on this channel, it's not the magnitude of warming, but the rate that is alarming. After an ice age, it normally takes a thousand years to warm by a degree. So warming by that much in a century is pretty damn fast.
Now if we were to double the CO2 in the air, the extra radiation per unit area, which tells us how much warming we can expect, would be about 3.45 watts per square meter. Close, but not quite. The current best estimate for the forcing caused by a doubling of CO2 is 3.7 watts per meter squared.
But at least Christopher is in the right vicinity. Now the extra radiation caused by all of our activities and all the changes to greenhouse gases, land use, deforestation since 1850 is about 2.9 watts per square meter. This estimate also seems a little high.
The last IPCC report put the total radiative forcing since 1750 at 2.29 watts per meter squared, but that was nearly a decade ago and is somewhat out of date. As we'll shortly see, Christopher cites a 2018 study by Lewis and Currie which does revise the IPCC estimate up to 2.8 watts per meter squared, which is closer. But this is the value for a change in forcing since 1750, not 1850 as Christopher is claiming. The value given by Lewis and Currie for this period is only 2.5 watts per meter squared.
But again, at least Christopher is in the right ballpark. but 0.8 of that 2.9 hasn't yet caused any warming because the vast heat capacity of the ocean causes the delay in the warming coming through. It's true that around 90% of the extra heat from global warming has been absorbed by the oceans and this causes a delay between the increase in energy and a rise in surface temperature.
Unfortunately I couldn't verify Christopher's values here, but once again they seem a little high compared to the values in the studies I looked at. But let's give him the benefit of the doubt. Now if we multiply 0.9 degrees by the ratio of the 3.45 watts per square metre CO2 forcing to the 2.1 watts per square metre of man-made forcing since 1850 that has caused warming to date, we can work out just how much warming a doubling of the CO2 in the air will cause.
and the warming will be about 1.5 degrees per CO2 doubling. Now don't worry if you don't like maths, I'm not going to go into tons of detail here. But I would like to point out that estimates for man-made radiative forcings are at least partially derived from climate models.
You know, the very things that Christopher supposedly doesn't need. But that aside, there's nothing obviously wrong with this equation. In fact, it's almost identical to the one found in the Lewis and Currie paper which Christopher cites, and it's used to calculate the same thing.
The warming response to a doubling of CO2, which is known as climate sensitivity. So far, so good. As I've already alluded to, the values Christopher uses in this equation are different from those in Lewis and Currie, but despite this, he arrives at the same value that they do.
1.5 degrees. So, that's settled then. Christopher is right.
climate sensitivity is 1.5 degrees and the warming projected by models is exaggerated, right? Well, no. Unfortunately for Christopher, it takes more than a single study or calculation to decide scientific debate.
Dozens of studies have attempted to calculate climate sensitivity, each with their own methodologies and their own flaws, and the majority put the figure between 1.5 and 4.5 degrees. Indeed, this is the range suggested by the IPCC. Does this mean that Lewis and Currie are wrong? No. Climate sensitivity could be 1.5 degrees, but based off the broader body of research, it is likely to be higher.
By cherry-picking a low estimate of climate sensitivity based on a single study and ignoring the greater body of research, Christopher is misleading his audience. There are also major limitations to his method of calculation. For example, it relies predominantly on instrumental data, which is presumably what Christopher means when he says real world data. As the name suggests, this data is taken from instruments like thermometers and it only extends back about 150 years.
But we don't know if the warming over this period is representative of how the climate will respond over longer time scales. In fact, there is evidence which suggests that many of the feedbacks which amplify warming and would contribute to a higher climate sensitivity have not fully kicked in yet. This may explain why studies which use instrumental data alone invariably produce lower estimates for climate sensitivity than those which use paleoclimate data. or models, both of which can look at much longer periods of time.
But of course, none of these methods is perfect. Each has advantages and disadvantages, and the fact that there is no perfect methodology is all the more reason to look at the research as a whole rather than picking just one study or methodology and deciding that all the others are wrong. Now, back to you Naomi. Quite frankly, 1.5 degrees is a lot less than 4.1 degrees, which is what climate models from the IPCC claim to predict, or at least they used to, because apparently this article suggests that climate sensitivity is even higher.
So is 1.5 degrees really all? Naomi, I'm not sure where you're getting this 4.1 degree value from. The average value of the climate models in the last IPCC report was just 3.2 degrees, and the IPCC best estimate for climate sensitivity is also around 3 degrees. Now let's address this article. To summarize, some climate models have made some adjustments in the way they model cloud feedbacks because our understanding of cloud dynamics has improved.
Once these adjustments were made the climate sensitivity projected by these models was as high as 5 degrees. This is over three times higher than Christopher's value and half a degree higher than the IPCC probable range. This value may well be too high. and it will be interesting to see the response papers which come out in the future. But one of the reasons it is being taken seriously is that the same models which predict this high sensitivity have also managed to accurately predict short-term weather, which reinforces the validity of the model.
Yes, Naomi, even though the real world warming since 1990 has been less than half what the UN's climate panel predicted that year, they predicted it confidently. The modelers have paid no attention to their previous failures. Let's have a look at this claim in more detail. The 1990 models were crude by modern standards. They focused on warming caused by greenhouse gases and ignored short-term variables like aerosol pollution.
But while this may seem like a major limitation, we have to remember that short-term variables are, by definition, short-term. They cancel themselves out over longer timescales. With this in mind, let's compare what the IPCC predicted in 1990 with what actually happened.
Given the simplistic nature of the model, their predictions hold up remarkably well. While they did slightly overestimate warming, this was largely because man-made greenhouse gas emissions did not increase as quickly as the modelers had assumed. And that's not the fault of the model, by the way.
Our greenhouse emissions are a product of policy decisions, not the physics of the climate system, which is what the model is designed to replicate. When researchers ran the model with this in mind, using data which reflected actual greenhouse emissions over this period, then the projection lines up almost perfectly with observations. You can see that here, represented by the light blue line.
But even without these corrections, observations are still well within the range of uncertainty presented within the original 1990 So no, Christopher, warming is not less than half what the UN predicted in 1990. In 1990 they predicted 3 degrees warming from doubled CO2. Now they predict 4.1 degrees instead of the 1.5 degrees that measurement shows they should predict. There's that 4.1 degree value again. When Naomi first said it, I assumed she'd made a mistake, but the fact that Christopher has also said it makes me think that there must be something I'm missing. So I did a bit of investigating and came across this article written by Christopher about the CMIP6 climate models.
These are the most up-to-date models used by the IPCC. Christopher's article makes the claim that the mean estimate for climate sensitivity of these models is 4.1 degrees. But when I looked at the literature, I couldn't find this backed up anywhere.
Now, CMIP6 models do have a higher average than previous models, with a mean estimate of 3.7 degrees, compared to a mean of 3.2 from the previous generation. But that's still not the 4.1 degree value Christopher is giving. All I can think is that at the time of writing his article, fewer models had released their data, and he was therefore not looking at the full range of estimates. Anyway, Monckton's article outlines pretty much the same argument that he's making here, that climate sensitivity calculated from instrumental data is at odds with climate sensitivity derived from models. Therefore, the models are wrong.
And it makes the same logical error, assuming that climate sensitivity calculated from instrumental data must be accurate. But as we've seen, there is plenty of evidence to suggest otherwise. At the very least, there is enough to cast doubt on this assumption. How can we check if 1.5 degrees is correct?
The predicted extra radiation from doubled CO2 is about the same as the predicted extra radiation this century from all our emissions. So we might expect 1.5 degrees warming this century. And sure enough, the measured warming in the real world rather than in the models since 1979 is indeed equivalent to 1.5 degrees per century.
Yes, Christopher, but we don't know if the warming of the last few decades is representative of warming over longer timescales, do we? That's the whole problem with using exclusively instrumental data. This is also a textbook example of circular reasoning. Christopher claims that instrumental data alone is sufficient to calculate climate sensitivity, and his evidence for this is that the instrumental data demonstrates his calculation to be accurate. And of course we know we can trust instrumental data because instrumental data alone is sufficient.
See the problem? We still don't know if instrumental data will give us an accurate estimate applicable over longer time scales. But even if Christopher is right, an extra 1.5 degrees warming will still take us to 2.5 degrees above pre-industrial levels, which will still have major negative impacts on both natural ecosystems and human civilization.
You know what I find truly astonishing? The fact that we don't derive our knowledge about climate sensitivity from real world data, but rather from silly little computer games, so you can't call it knowledge at all. Yes Naomi, I'd find that astonishing too, if it were true. But it's not, is it?
Studies using instrumental data, like Christopher has used here, are included in the IPCC report, so it's not like they're being ignored. So are studies using paleoclimate data, which I would argue is also real-world data. After all, it's come from analysis of rocks and ice cores on Earth. You know, the real world.
And models are based on physics, also from the real world, and are constantly being tested against real-world data too. All in all, there's a whole lot of real-world data going into estimating climate sensitivity. What you've just proven is that climate models that cannot even predict the weather a week from now cannot predict the longer-term climate either. No, Naomi, he's done nothing of the sort. All he's shown is that estimating climate sensitivity from instrumental data produces a different result from estimates which use paleoclimate data or models.
This doesn't mean that model or paleoclimate estimates are wrong. any more than it means that instrumental estimates are wrong. And by the way, if you reread the article that started this whole discussion, you'll find that the models in question actually can predict short-term weather, so the implication that they're inaccurate just doesn't hold up.
Anyway, Christopher goes on to reiterate his demonstrably incorrect claims that models are woefully inaccurate and thus totally pointless, before coming back to his core point. If all we want to know is how much warming we may cause... Models are pointless.
The real world data is enough, and it shows beyond any reasonable doubt that the warming that we may call will be small, harmless, net beneficial. Good for us, good for our fellow creatures on this planet, good for plant life, good for the planet as a whole. Well...
That's quite the leap of logic. Even if Christopher's claims that warming will be small were true, he's provided no evidence to support the idea that it will be beneficial. This also flies in the face of the brunt of scientific literature in this area, but since he provides no evidence for these claims, I'm not going to waste my time refusing them. And Christopher keeps saying that the real world data is enough, but I'm not sure if even he knows what he means by this.
After all, we've already seen that his version of real world data actually includes some data from models. Anyway, at this point in the video, Naomi goes on to reiterate everything that they've said so far. That climate models are pointless, that the scientists behind them are lying to you, you know, the usual sceptic spiel.
She eventually finishes up with this. As always, I want you to remember that I don't want you to panic because there's no need to panic, there's only a need to be more reasonable. I want you to think. And frankly, I think she's spot on. There is no need to panic.
It's unproductive at best and self-defeating at worst. And we should all try to think more, and that involves fact-checking everything you are told, both from scientists and ageing English aristocrats with no formal training in the subject they're talking about. And that goes for you internet people, too. I've linked all my sources in the description, so feel free to fact-check anything I've said.
I'm only human, so I may well have made mistakes. But anyway… That seems to be the end of Naomi's video, so I guess I'll wrap it up here. I hope you've enjoyed watching, and if you did, don't forget to show your appreciation with a like and a comment. I really appreciate it. And if you want to see more of my content, don't forget to subscribe and smash that bell icon.
There's still plenty of sceptics out there and plenty more to come from me, so stay tuned. Until next time, goodbye.