An informed analysis of IPCC errors

The link to the document on LaRouche's website is:

http://www.larouchepub.com/eiw/public/2007/2007_20-29/2007-25/pdf/33-37_725.pdf

It's an interview, not a piece of analysis.

A link to a copy on some other website was posted earlier in the thread.

 

OK, that's a fair question. I composed a full answer off-line, to try to get this right. What can I say, I'm a (social) scientist, it's my nature to try to get it all down on paper.

I guess you really just want a quick, bottom-line response. It boils down to a generic answer (just below, same answer you'd get for any question about science) and the details around that (further below).

Generically, it’s just like any other science: Either show data that materially contradict the general circulation models (and I’ll narrow that down below), or, if the data remain consistent with the general circulation models, then provide an alternative explanation that’s better than the current model (and I’ll narrow that down below too).

And, to be clear, I don’t mean having somebody slap some crap on the page, as typically done here, and say, there, you’re wrong. I mean, a thoughtful analysis by somebody who knows that they’re doing.

THE FRAMEWORK

Let me limit this to global average temperature trend from the IPCC ensemble of general circulation models. The actual evidence is broader than that, and those models predict vastly more than just the temperature trend. (And, in fact, the good match to the observed, in all those ways, is something that makes the models a lot more believable). If you want to see all the things these models predict, and how the matchup looks (e.g. ) across areas, over time, through the depth of the atmosphere, for precipitation and so on, you can read the IPCC's last report, here, though I guess a lot of that will read like mush to most people:

Chapter 8: Climate Models and their Evaluation

For example, by and large, they don't model the El Nino well. But they're working on that. And that just doesn't much matter for making average predictions over long time scales.

I think its worth taking three sentences to make sure people understand those general circulation models. Those models are not fitted to the temperature data -- they aren't some statistical fit to the data. They predict an atmospheric temperature and a temperature trend, from first principles, based on the laws of physics (and some rules-of-thumb where they can't model it down to that level of detail.) Basically, they just calculate, blind, what they think the temp trend ought to be, based on incoming solar radiation and all the rest of it. They literally predict conditions in the atmosphere, at short time intervals, for small(ish) blocks of atmosphere, for the entire surface of the earth. The trend you see is the average of many runs of those calculations, for modestly different starting values for the models.

And, also to be clear, they don't generate any trend in and of themselves. If you leave out greenhouse gases, the models just predict random variations around a constant temperature, as shown by the blue areas here (no GHG increase) versus the red areas (with GHG increase) (from the 2007 IPCC report):




And, if you could measure it well, you’d really rather look at surface heat content rather than temperature. But temperature is what everybody looks at.

Let me also put aside all the chatter about how none of the data are right, how (e.g.) scientists across the world have somehow conspired to mis-measure temperature, and so on. That’s essentially just noise by people who, upon examination, seldom have any understanding of what they’re talking about.

That’s not an assertion that the data are perfect, it’s an assertion that none of the amateurs out there has produced a more reliable set of data than the actual scientists have.

This is also not an assertion that the models are perfect. They disagree with one another, so we know there’s not one best solution.

The real question is not yes/no right/wrong. The real question is one you always face with observational (non-experimental) data: Given what we risk by action and inaction, are the models reliable enough and accurate enough that it would be prudent to act on their predictions? Or are they so clearly wrong, or so obviously questionable, that it would be imprudent to act on the model predictions.

That said, I’ll discuss it as global temperature trend (not “the vast array of actual model predictions”), and yes/no (not “reliable enough to be useful”) below.

And, I realize that “the media” get in your face with boogey men all the time. I guess I also need to state why I think this particular boogey man is worth paying attention to.

1: Future temperature change outside the range predicted by the general circulation models.

OK, with that as the frame, practically speaking, if we saw cooling that went beyond the model estimates plus uncertainty, by a significant margin, then that would cause me to change my mind. It would show that the models I am depending on for my understanding of the world are grossly at odds with the data, in a direction that matters for my decisionmaking. So, it’s not simply cooling, it’s really, at this point, failure to warm as fast as (the average of the ensemble of models used by the IPCC) predicts, plus or minus their uncertainty.

Here's how things looked at the end of 2009, for predictions of global average temperature:



Graphically, if the observed time series for global mean temperature pierced the gray envelope of uncertainty, on the downside, that would make me change my mind. And not just a blip (because, in principle, 1 time of out of 20 you get a "statistically significant" result purely by chance, at the 5% probability level.) For example, you can see one "hot" year that hit the edge of the envelope here. But something prolonged, so that there's no uncertainty that it was unexplained by the existing models of climate. That would certainly prove that my world-view was wrong, at least.

I guess I should specify that by cooling, I mean cooling as measured by the (largely) ground-based observations. The GISS timeseries, the Hadley timeseries. By contrast, "the satellite data" jump around too much to suit me, and different sources that interpret the satellite data have a pretty strong disagreement about basic stuff, like, what's the trend.

At this point, I was going to go into a big song-and-dance about how many years you'd need and blah blah blah, but just eyeball the graph.

Even that wouldn't necessarily mean the theory of GHG-caused global warming was wrong, it would mean that those models are wrong. But I'd accept that as demonstrating that something was deeply wrong about the theory that GHG increases are causing global warming.

2: Alternative, better explanation of the data.

Second, if somebody could provide 1) a convincing alternative explanation of what else is actually causing the warming while, at the same time, 2) explaining why C02 doesn't cause the warming, then I might change my mind on the basis of that. But, basically, 2) isn't going to happen (see Alric above). And all the ones that I've seen for 1) have been pretty half-assed -- most aren't even logically consistent, let alone consistent with the data.

That goes, by the way, for variations on “it’s a natural cycle”. That’s an interesting area for research, I’m not dismissing it. But, fine, if you want to claim that, then, in my book, to claim that this shows it’s not GHGs, then you also have to explain what is wrong with the calculations showing warming from GHGs. Otherwise, you end up with more warming that actually occurred. In other words, a complete alternative explanation must always have two parts: Here’s the real cause, and here’s where all these guys went wrong in calculating that (e.g.) C02 was a greenhouse gas. Ffor the alternative explanation to be preferred, you have to work all that out and demonstrate it empirically, not just raise the possibility that there might be something left out. You have to prove that your alternative explanation is better.

So let me rephrase that. I'm not talking about some half-assed model of correlation between this and that (sunspots and temperature). I'm talking about a fully-worked-out alternative model of how the atmosphere works, based on physical principles, that explains why the earth is warming from some alternative other than GHGs, and explains why GHGs do not actually cause warming. And that alternative model would have to explain the world better than the current one.

What's a good example of what I'd accept there? Well, that paper that came out last month about variation in stratospheric water vapor was pretty good. It was one shot, based on scant data, hasn't been validated, and blah blah blah, but that was genuine news to a lot of scientists. It didn't "disprove global warming", in fact, it included manmade GHG warming and gave a modestly better fit to the data. (So it actually strengthened the case.) But it did show that there are materially important aspects of climate that have not yet been pinned down. You have to be open to the idea that maybe somebody will find one that overturns the whole shebang, though, per Alric's point, that's pretty doubtful.

3) Why pay attention to this boogey man.

Here’s the nutshell view.

We have these models that did a good job of explaining variation in climate in the last century.

So, it’s not that big a stretch to say, they worked OK last century, let’s see what they say about this century.

I think that's a fairly reasonable thing to do. And, then, is there anything a prudent person might want to worry about?

Because, basically, the issue isn't what's already happened. Right now, the total energy imbalance from man amounts to little over a tenth of a percent of the net energy input to the earth (sunlight is about 1300 watts/square meter, net forcing is maybe 2 watts/square meter.) The real issue is, what’s likely to happen if we continue on this path, and really drive the system hard with GHGs. The reason people focus on C02 is that it cumulates and there's no way to get rid of it, at the scale we're producing it, as far as we know now, and it’s the long-term impact of the buildup that’s the issue.

OK, here’s a nice one, taken from this publication in Science:

Model Projections of an Imminent Transition to a More Arid Climate in Southwestern North America -- Seager et al. 316 (5828): 1181 -- Science

If you want to read the whole thing, you need to register.

Let me put up the graph first, then tell you what it means. The blue line is projected precipitation changes for the US Southwest, defined here as everything west of the Mississippi and south of the Nebraska border. Everything to the left and down from the NE corner of Kansas.


The key fact is that a value of –0.09 is what brought on the '30s dustbowl. There was one drought in the ‘50s where it hit –0.13. But, basically, where the blue line dips below 0.09, what this graph tells you is that the median projection, from all the models considered by the IPCC, is that the US Southwest is, at that time, probably a dustbowl.

So, the models are saying a couple of things. If we keep doing what we’re doing, it becomes increasingly likely that we’ll make the US Southwest into a permanent dustbowl. And, the even-money bet for when it flips to dustbowl conditions is 2050. And, though you can’t see it on the graph, but could read it in the underlying article, roughly speaking, there’s a one-in-four chance that could happen by 2030.

To be clear, it doesn’t say 100% chance of dustbowl. It says the even-money bet starts at 2050, and beyond that, it’s a better-than-even-money bet that it’ll happen.

So, the models say, keep doing what you’re doing, and it’s a pretty good bet you’ll take about a quarter of the US lower-48 land mass and convert it to a dust bowl.

The way this gets written up is, if you want to see what we're going to look like by mid-century, take a look at what's happening in Australia now. Not sure how valid it is, but they've certainly had a drought of epic proportion, with dust storms to match.


Me, I think that’s worth paying attention to. I'm not talking about boohoo the polar bears are toast (pity, but not key in my life). I'm talking about, what are my grandchildren going to eat when they get to be my age?

And that’s not even close to the worst risk, and certainly not the only risk, that you get when you run the models forward far enough. But isn’t that enough to make you take note of the issue?

So at the end of the day, you pays your money and you takes your chances. You look at the best available evidence. A whole lot of very smart people think they've figured this out. What they say forms an internally consistent view of the world, that is consistent with the laws of physics, and that appears to match observations (to within the uncertainty of the models and data.) The models that show those bad outcomes for decades out also show that we need to start cutting back now if we want to avoid them. The probable downsides of inaction appear pretty costly. Any reasonable analysis of the issue (e.g., the US CBO) shows that the likely cost of cutting back use of carbon-based fuels is relatively modest (e.g., 2% of GDP by 2050). I’d pay 2% of GDP to avoid a US Southwest dust bowl, and I don’t even live there.

Basically, I think it's literally cheaper to do the preventive maintenance now, than to try to do the repairs later. What’s it worth to lose a quarter of the US lower 48 land mass, permanently, to dustbowl conditions? I think it’s cheaper, and certainly less risky, to cut carbon output now.

 

Tochatihu put that in a recent thread about C02. I couldn't get it to play but my browser is out-of-date.

A23A

Tochatihu: My apology, it was my browser, plays OK in Internet Explorer, I'm running it now. Thanks for the link.

Skipping through it, I see what you mean. I'll run it in one piece this weekend when I have the time. The guy clearly knows his stuff, and the explanations are something that anyone could understand. Very nice. I see now why you were pushing this.

Basically, there are no more significant anomalies between C02 and temperature. I'd stumbled across the ordovician glaciation myself, I hadn't realized the most recent research appears to explain the remaining ones.

This guy is a phenomenally good speaker.

 

Knock yourself out, for what it's worth. It was a serious question, I thought it deserved a serious reply.

 

OK, the credit card analogy was strained.

But my takeaway from this is that I seem to approach these things with a more humble attitude than most.

When a world-renowned expert gives a plenary lecture at the annual meeting of the American Geophysical Union, talking to an audience of experts, ... my first instinct is at least to pay attention to what he said before I dismiss him as being too stupid to understand what he's talking about. I mean, like, nobody in the audience was smart enough to call him out on this?

What caused the rise? You must have missed the part where he put up the fast-Fourier-transform of the the temperature changes, labeled the peaks in the frequency domain to show you what part of changes in earth's orbit those corresponded to, then went on to discuss how the initial changes in temperature were unambiguously caused by orbital changes.

How did you miss the part of the slide on that card analogy where he labeled the start of the warming "orbits".

Maybe you missed the part around 4:30 where he, in essence, gently mocked somebody who sent him an email about that exact point? And got a brief round of applause? And brought that exact quote (
"This one scientific fact alone ...) when he discussed the lag.

Illogical speculation? Did you see the part where they took general circulation models, reprogrammed them for those periods, and got a much reduced warming without the C02. But replicated the warming with it. That's a physics-based calculation of what the temperature should have been.

The cartoon version is: in the paleo data, orbital changes initiate warming/cooling, and C02 feedback enhances the change. Without the feedback, you can't explain the size of the change. Orbital forcing along can't explain it. With C02, you can. It's that simple. You only get half the warming with orbital changes alone.

Are you worried about the timing? Why? Today's situation is completely different with respect to the cause of the increased C in the air.

Then, in the ice ages, we were looking at carbon that was being driven into or out of the biosphere by changes in temperature initiated by the orbital forcing. E.g. driven out of solution in the ocean (just like warming up a can of Coke).

Now, at present, that's not the source of the carbon. We know the source of the carbon in the current issue. It's us. We've got receipts. Statistics of trade tell us total oil, natural gas, coal sales. Translate that to gigatons of C burned, and it's more than the annual increase in C in the atmosphere. Nature remains a net sink. The ocean isn't off-gassing slowly, as during the ice ages, it's soaking up the excess carbon that we've put into the air.

So the timing issue, for the ice age cycles, is irrelevant to the modern situation. Why? Because then, they were looking at the slow release of C due to temperature changes initiated by orbital forcing. Now, we're looking at a vastly faster increase in C due to burning.

Look at the time scales on his charts. In the ice-age cycles, atmospheric C02 varied by about 100 ppm over the course of >100,000 years.

Ice age, from start to finish, 100 ppm C02, >100,000 years. Carbon slowly moved out of biosphere into atmosphere. Required initial temperature rise to initiate it.

We've raised atmospheric C02 100 ppm in a little over 100 years. Carbon being dug up and burned. No need for temperature change to initiate it -- no need to for lag.

On average, C02 is increasing roughly 1000 times faster than the average change over the ice-age cycle, for the past four ice ages.

That slow, gentle, tiny lag over the >100,000 year ice age cycle -- it just isn't relevant to what's going on now.

 

I'll make it simple:

Over the last four ice ages, the feedback from slowly warming temperatures due to orbital changes, to slowly increasing atmospheric C02 levels, appears to account for about half the total warming/cooling that appears over the typical ice age cycle.

My vague recollection is that this is why Milankovitch cycles were originally dismissed. Scientists calculated the warming you'd get and it wasn't enough, by itself, to account for the temperature change. So, the fact that orbital forcing alone was insufficient to explain the (depth of the) ice ages has been known for a long time. The point of that part of the slide show is that the feedback to C02 changes explained the rest.

Yeah, it's even on wikipedia:

[ame=http://en.wikipedia.org/wiki/Milankovitch_cycles]Milankovitch cycles - Wikipedia, the free encyclopedia[/ame]

"The effects of these variations are primarily believed to be due to variations in the intensity of solar radiation upon various parts of the globe. Observations show climate behaviour is much more intense than the calculated variations."

So nobody said that C02 changes caused the ice ages. At least not anybody smart. What they said is that they amplified them. The root cause is orbital variation, as was stated clearly in that lecture. But the fact that C02 amplifies them is evidence that changes in C02 warm or cool the planet.