I found the chart below in the chapter Global Climate Change of the NOAA/NASA CCSP climate change report. (I discuss this report more here). I thought it was illustrative of some interesting issues:
The Perfect Backcast
What they are doing is what I call "backcasting," that is, taking a predictive model and running it backwards to see how well it preforms against historical data. This is a perfectly normal thing to do.
And wow, what a fit. I don’t have the data to do any statistical tests, but just by eye, the red model output line does an amazing job at predicting history. I have done a lot of modeling and forecasting in my life. However, I have never, ever backcast any model and gotten results this good. I mean it is absolutely amazing.
Of course, one can come up with many models that backcast perfectly but have zero predictive power.
A recent item of this ilk maintains that the results of the last game played at home by the NFL’s Washington Redskins (a football team based in the national capital, Washington, D.C.) before the U.S. presidential elections has accurately foretold the winner of the last fifteen of those political contests, going back to 1944. If the Redskins win their last home game before the election, the party that occupies the White House continues to hold it; if the Redskins lose that last home game, the challenging party’s candidate unseats the incumbent president. While we don’t presume there is anything more than a random correlation between these factors, it is the case that the pattern held true even longer than claimed, stretching back over seventeen presidential elections since 1936.
And in fact, our confidence in the climate models based on their near-perfect back-casting should be tempered by the fact that when the models first were run backwards, they were terrible at predicting history. Only a sustained effort to tweak and adjust and plug them has resulted in this tight fit (we will return to the subject of plugging in a minute).
In fact, it is fairly easy to demonstrate that the models are far better at predicting history than they are at predicting the future. Like the Washington Redskins algorithm, which failed in 2004 after backcasting so well, climate models have done a terrible job in predicting the first 10-20 years of the future. This is the reason that neither this nor any other global warming alarmist report every shows a chart grading how model forecasts have performed against actual data: Because their record has been terrible. After all, we have climate model forecasts data all the way back from the late 1980’s — surely 20+ years is enough to get a test of their performance.
Below is the model forecasts James Hansen, whose fingerprints are all over this report, used before Congress in 1988 (in yellow, orange, and red), with a comparison to the actual temperature record (in blue). (source)
Here is the detail from the right side:
You can see the forecasts began diverging from reality even as early as 1985. By the way, don’t get too encouraged by the yellow line appearing to be fairly close — the Hansen C case in yellow was similar to the IPCC B1 case which hypothesizes strong international CO2 abatement programs which have not come about. Based on actual CO2 production, the world is tracking, from a CO2 standpoint, between the orange and red lines. However, temperature is no where near the predicted values.
So the climate models are perfect at predicting history, but begin diverging immediately as we move into the future. That is probably why the IPCC resets its forecasts every 5 years, so they can hit the reset button on this divergence. As an interesting parallel, temperature measurements of history with trees have very similar divergence issues when carried into the future.
What the Hell happened in 1955?
Looking again at the backcast chart at the top of this article, peek at the blue line. This is what the models predict to have been the world temperature without man-made forcings. The blue line is supposed to represent the climate absent man. But here is the question I have been asking ever since I first started studying global warming, and no one has been able to answer: What changed in the Earth’s climate in 1955? Because, as you can see, climate forecasters are telling us the world would have reversed a strong natural warming trend and started cooling substantially in 1955 if it had not been for anthropogenic effects.
This has always been an issue with man-made global warming theory. Climate scientists admit the world warmed from 1800 through 1955, and that most of this warming was natural. But somehow, this natural force driving warming switched off, conveniently in the exact same year when anthropogenic effects supposedly took hold. A skeptical mind might ask why current warming is not just the same natural trend as warming up to 1955, particularly since no one can say with any confidence why the world warmed up to 1955 and why this warming switched off and reversed after that.
Well, lets see if we can figure it out. The sun, despite constant efforts by alarmists to portray it is climactically meaningless, is a pretty powerful force. Did the sun change in 1955? (click to enlarge)
Well, it does not look like the sun turned off. In fact, it appears that just the opposite was happening — the sun hit a peak around 1955 and has remained at this elevated level throughout the current supposedly anthropogenic period.
OK, well maybe it was the Pacific Decadal Oscillation? The PDO goes through warm and cold phases, and its shifts can have large effects on temperatures in the Northern Hemisphere.
Hmm, doesn’t seem to be the PDO. The PDO turned downwards 10 years before 1955. And besides, if the line turned down in 1955 due to the PDO, it should have turned back up in the 1980’s as the PDO went to its warm phase again.
So what is it that happened in 1955. I can tell you: Nothing.
Let me digress for a minute, and explain an ugly modeling and forecasting concept called a "plug". It is not unusual that when one is building a model based on certain inputs (say, a financial model built from interest rates and housing starts or whatever) that the net result, while seemingly logical, does not get to what one thinks the model should be saying. While few will ever admit it, I have been inside the modeling sausage factory for enough years that it is common to add plug figures to force a model to reach an answer one thinks it should be reaching — this is particularly common after back-casting a model.
I can’t prove it, any more than this report can prove the statement that man is responsible for most of the world’s warming in the last 50 years. But I am certain in my heart that the blue line in the backcasting chart is a plug. As I mentioned earlier, modelers had terrible success at first matching history with their forecasting models. In particular, because their models showed such high sensitivity of temperature to CO2 (this sensitivity has to be high to get catastrophic forecasts) they greatly over-predicted history.
Here is an example. The graph below shows the relationship between CO2 and temperature for a number of sensitivity levels (the shape of the curve was based on the IPCC formula and the process for creating this graph was described here).
The purple lines represent the IPCC forecasts from the fourth assessment, and when converted to Fahrenheit from Celsius approximately match the forecasts on page 28 of this report. The red and orange lines represent more drastic forecasts that have received serious consideration. This graph is itself a simple model, and we can actually backcast with it as well, looking at what these forecasts imply for temperature over the last 100-150 years, when CO2 has increased from 270 ppm to about 385 ppm.
The forecasts all begin at zero at the pre-industrial number of 270ppm. The green dotted line is the approximate concentration of CO2 today. The green 0.3-0.6C arrows show the reasonable range of CO2-induced warming to date. As one can see, the IPCC forecasts, when cast backwards, grossly overstate past warming. For example, the IPCC high case predicts that we should have see over 2C warming due to CO2 since pre-industrial times, not 0.3 or even 0.6C
Now, the modelers worked on this problem. One big tweak was to assign an improbably high cooling effect to sulfate aerosols. Since a lot of these aerosols were produced in the late 20th century, this reduced their backcasts closer to actuals. (I say improbably, because aerosols are short-lived and cover a very limited area of the globe. If they cover, say, only 10% of the globe, then their cooling effect must be 1C in their area of effect to have even a small 0.1C global average effect).
Even after these tweaks, the backcasts were still coming out too high. So, to make the forecasts work, they asked themselves, what would global temperatures have to have done without CO2 to make our models work? The answer is that if the world naturally were to have cooled in the latter half of the 20th century, then that cooling could offset over-prediction of temperatures in the models and produce the historic result. So that is what they did. Instead of starting with natural forcings we understand, and then trying to explain the rest (one, but only one, bit of which would be CO2), modelers start with the assumption that CO2 is driving temperatures at high sensitivities, and natural forcings are whatever they need to be to make the backcasts match history.
By the way, if you object to this portrayal, and I will admit I was not in the room to confirm that this is what the modelers were doing, you can do it very simply. Just tell me what substantial natural driver of climate, larger in impact that the sun or the PDO, reversed itself in 1955.
A final Irony
I could go on all day making observations on this chart, but I would be surprised if many readers have slogged it this far. So I will end with one irony. The climate modelers are all patting themselves on the back for their backcasts matching history so well. But the fact is that much of this historical temperature record is fraught with errors. Just as one example, measured temperatures went through several large up and down shifts in the 40’s and 50’s solely because ships were switching how they took sea surface temperatures (engine inlet sampling tends to yield higher temperatures than bucket sampling). Additionally, most surface temperature readings are taken in cities that have experienced rapid industrial growth, increasing urban heat biases in the measurements. In effect, they have plugged and tweaked their way to the wrong target numbers! Since the GISS and other measurement bodies are constantly revising past temperature numbers with new correction algorithms, it will be interesting to see if the climate models magically revise themselves and backcast perfectly to the new numbers as well.
I agree 100%. To anyone who has done some mathematical modeling, it is painfully obvious that climate models have been fudged to fit history. The fit is just too good going backwards (and too poor going forwards). Moreover, climate models with very different assumptions all fit history.
The amount of groupthink and self-deception which is going on in the modeling community is amazing.
And by the way, I think you dignify it by calling it “backcasting.” I would call it vaticinium ex eventu. Prophecy after the fact.
Just a quick question:
If you extend the graph below the 270ppm level, what do those lines do? It seems obvious that they’ll all end up at wildly differing temperatures at the 0ppm level.
I find it amazing that so-called reputable scientists would willingly associate themselves with that sort of rubbish.
Further to my last comment:
The sensitivity graphs all show the greatest slope at around 270ppm. That being the case, are we to assume that, in pre-industrial times, small changes in CO2 levels led to wild swings in global temperatures? I’ve certainly seen no evidence of that.
Consensus on man made warming has been considered settled science for a few years, why are we still tweaking the models. The models already have spoken and furthermore they were “peer reviewed”! Oh wait, did I say models, I meant model, because all model are created equally.
I am always baffled by people who use the argument of “the models prove agw”. It’s such a feat of circular logic I am almost impressed.
Also, I believe the IPCC doesn’t like to call them plugs, preferring to use “flux adjustments” which are later defined as “empirical corrections that could not be justified on physical principles”.
Comedy gold.
Just did a quick google on “Flux Adjustments” and was amazed – simply gobsmacked that anything at all was based on models so adjusted – in many cases the adjustments were quite large, of the same order of magnitude of the modeled fluxes.
It does appear that more modern models are now producing better results without flux adjustments. Which is great, more power to them – they recognized a problem and fixed it properly, by creating a better model – which is what scientists do. What scientists don’t do is patch up the broken model and scare the world with it in the meantime.
I can tell you what happened in 1955. It was a strong La Nina year, so it was a perfect year for them to cherry pick.
This post is as lame as anything I’ve seen here, full of childish claims about climate science, lies, fiction, and horrible misunderstandings of science.
I have done a lot of modeling and forecasting in my life. However, I have never, ever backcast any model and gotten results this good. I mean it is absolutely amazing. – you don’t say what field you’ve done modelling in. It’s certainly not climate science. If you have never got good results from your modelling, perhaps you’re trying to model something that can’t be accurately modelled. Or perhaps you’re really bad at modelling. Whatever, the inadequacies in your attempts to model unrelated phenomena have no bearing on how to assess the qality of climate models.
when the models first were run backwards, they were terrible at predicting history. Only a sustained effort to tweak and adjust and plug them has resulted in this tight fit – bullshit, pure and simple. As ever, you just make up your own facts, when a cursory glance at the literature shows you to be living in an anti-scientific Luddite dream world all of your own.
You regurgitate the same old tired attempt to show that Hansen 1988 has not accurately predicted today’s temperatures. Your attempt relies on comparing apples with oranges and cherrypicking the bit you like the look of. Try comparing annual averages of observations with annual predictions, instead of monthly with annual.
So the climate models are perfect at predicting history, but begin diverging immediately as we move into the future. That is probably why the IPCC resets its forecasts every 5 years, so they can hit the reset button on this divergence – how ignorant are you, really? Try reading Rahmstorf et al 2007, which compares 1990 projections with 16 years of subsequent observations.
What happened in 1955? Not much. Why do you think something should have? Are you even looking at the blue line? It lies just above 32.0F in 1955, and by the time we get to 2000, where does it lie? That would be just above 32.0F.
Climate scientists admit the world warmed from 1800 through 1955, and that most of this warming was natural. But somehow, this natural force driving warming switched off, conveniently in the exact same year when anthropogenic effects supposedly took hold – you have to try quite hard to be such a moron, really. ‘Switched off’? Who ever said that? Try living in the real world for a bit, will you? If you think that anthropogenic effects are thought to have begun in one single year, you are even more woefully ill-informed than I’d thought possible.
The sun, despite constant efforts by alarmists to portray it is climactically meaningless, is a pretty powerful force – constant efforts? Show us one, to prove you aren’t just making stuff up yet again. And please, learn the difference between a climate and a climax.
This graph is itself a simple model – and a wrong one. You’re taking an equation to calculate the equilibrium response and thinking that you can use it to calculate the transient response. You can’t. Your graphs do not show what you think they show.
aerosols are short-lived and cover a very limited area of the globe. If they cover, say, only 10% of the globe, then their cooling effect must be 1C in their area of effect to have even a small 0.1C global average effect – you ever heard of Pinatubo? Do you know what happened to global temperatures after it erupted? Do you know what this says about the lifetime and cooling effect of aerosols? I’ll give you a clue – your beliefs about them are not even remotely close to the truth.
So, to make the forecasts work, they asked themselves, what would global temperatures have to have done without CO2 to make our models work? The answer is that if the world naturally were to have cooled in the latter half of the 20th century, then that cooling could offset over-prediction of temperatures in the models and produce the historic result. So that is what they did. Instead of starting with natural forcings we understand, and then trying to explain the rest (one, but only one, bit of which would be CO2), modelers start with the assumption that CO2 is driving temperatures at high sensitivities, and natural forcings are whatever they need to be to make the backcasts match history. – utter, utter, utter bullshit. How pathetic to even think this sort of crap, let alone write it down, let alone publish it for all the world to see.
Scientist says, “aerosols are short-lived and cover a very limited area of the globe. If they cover, say, only 10% of the globe, then their cooling effect must be 1C in their area of effect to have even a small 0.1C global average effect – you ever heard of Pinatubo? Do you know what happened to global temperatures after it erupted? Do you know what this says about the lifetime and cooling effect of aerosols? I’ll give you a clue – your beliefs about them are not even remotely close to the truth.”
Based upon an examination of the graph, the general dip in global temperature average attributable to volcanic eruptions lasts for roughly five years at most. By your own words in relationship to time scales for climatic variation, this is insignificant with regard to its effect on temperature averages. If ten years of a flat trend in temperature since 1998 is insignificant, than five years is also insignificant.
Scientist, I do modeling in compressible fluid dynamics, which is hardly unrelated to climate. I suspect you have no idea what the challenges are in modeling fluid systems or have any real clue about the state of the science today. Unfortunately, neither do a lot of climatologists–most of them are just plain lacking in the mathematical background.
Keith – volcanic aerosols come from a single injection. Industrial aerosols are constantly replenished. The aerosols from a single volcanic eruption can easily spread over a large proportion of the globe, and can easily cause global cooling of a few tenths of a degree. So, the claims that aerosols only cover a small portion of the globe and cannot even cause cooling of 0.1°C are clearly false.
Josh S – why should climate scientists know about compressible fluid modelling, and what is your source for assessing their mathematical backgrounds?
Why do you want to jump in with a second grader pissing in the pool? “Scientist” never provides any scientific refutation, just foul-mouthed, leftist contradiction. There’s nothing worthy of response in her comment.
Skeptic, thanks for getting me over my laziness. I went and read the draft propaganda for myself. On Page 22:
“This pattern of tropospheric warming and stratospheric cooling is consistent with our understanding of how atmospheric temperature should be changing in response to increasing greenhouse gas concentrations.”
Has Hansen discovered some old NASA data tapes regarding the
tropical mid-tropospheric hotspot?
Psyentist,
exactly what mechanism injects the industrial aerosols into the stratosphere, similar to volcanic eruptions, that allow them to be globally distributed and “hang around” for a few years??
@the so-called “Scientist”
http://www.sciencedaily.com/releases/2008/08/080812160615.htm
Psychentist,
There is a blog called ATMOZ. The guy that runs that blog used to run around claiming to be a scientist too…til it was discovered he was a student in AZ.
Mike C has strong opinions but few convincing refutations.
The simple facts are that Hansen A and B wildly overstated CO2 temperature sensitivity from 1988.
The warming since 1900 mostly occurred from 1980-1998 and then plateaued. Meanwhile, CO2 has
shown a pretty steady line increase since 1880.
It’s pretty hard to believe the AGW predictions without convincing explanations for the poor
temperature vs CO2 tracking over the last 108 years.
Despite emotion and anger, there is simply no way we should accept hindcasting as evidence of predictive potential of climate models. Forecasting is the only acceptable test in science, and it is demonstrable that forecasts of models to date are substantially too high:
The 1990 IPCC FAR central prediction is much too high.
The 1995 SAR central prediction didn’t do badly until the last several years. Now it is statistically well above the historical line, even including any upward biases in the historical record per McKitrik and Michaels analysis.
2001 TAR is also much too high, .
2007 AR4 is too early to tell as a forecast, although Lucia Liljegren has shown that its hindcast from 2001 is already falsified.
Also, although it is only one model, it is fair to criticize the Hansen prediction, since it has been assigned much attention and publicity. Anyone can look at it and tell that is also too high. To deny the evidence in front of one’s eyes is akin to Cardinal Bellarmine’s Jesuits refusing to accept the evididence of their senses when they saw the moons of Jupiter through Galileo’s telescope. Denial of evidence against belief is a very strong human leaning. It will take much much more in the way of disconfirmation to impress the believers and curtail their advocacy.
Dear Fred J.
You mean the projections made by NASA (Hanson model) and presented to congress in 1988 do not replicate measurements taken in the intervening 20 years? Doesn’t that mean that the 1988 model is falsified scientifically?
Keith – volcanic aerosols come from a single injection. Industrial aerosols are constantly replenished. The aerosols from a single volcanic eruption can easily spread over a large proportion of the globe, and can easily cause global cooling of a few tenths of a degree. So, the claims that aerosols only cover a small portion of the globe and cannot even cause cooling of 0.1°C are clearly false.
Josh S – why should climate scientists know about compressible fluid modelling, and what is your source for assessing their mathematical backgrounds?
Posted by: Scientist | August 13, 2008 at 10:28 AM
Let’s see, so much to choose from. First, Josh, I have agree with Scientist here about fluid modeling by climate scientists. Let’s be honest, they can’t figure out the mechanics of clouds for their models, they guesstimate the effects of precipitation, and generally ignore water vapor. Sounds to me like they don’t have any need to know anything about modeling fluids. And compressible fluids no less, no need for that even though fluids under pressure and gases under pressure share many similar properties as even most high school chemistry students know. [/sarcasm]
As for aerosols, considering that even Hansen agrees that the overall emission values for industrial aerosols is falling, one should be able to conclude that the total climatic effect of them is decreasing also. But that was not my point in my statement, and not even the point of the statement of Skeptic.
The point he made is that to lower temperatures around the globe, the aerosols either have to cover a large percentage of the globe or have a larger temperature dampening effect. Most aerosols do not travel over extensive areas of the globe due to the fact that they are heavier than air and tend to settle to the ground. To spread over a large region, they need something on the order of a volcanic eruption to disperse than high enough in the atmosphere for them to cover a wide area. Even so, they tend to settle out in a very short period of time. He never said it could not change temperature 0.1 degrees centigrade. To be accurate, he said it could, just that it would need a high degree of effect based upon the amount of area they would cover based upon short life and limited dispersal. “If they cover, say, only 10% of the globe, then their cooling effect must be 1C in their area of effect to have even a small 0.1C global average effect.”
My point was that the effect of aerosols is short lived, and used your own citation of Mt. Pinatubo as an example. I also used your own oft repeated point that “it takes more than ten years of a temperature trend” to be considered a significant climate/temperature event. The general perceived effect of volcanic eruptions tends to last only about five years, and you accept that as a valid climate/temperature event, so a trend that is twice as long should be just as valid. You can’t have it both ways, so which petard to you wish to be hoist on? Either ten years is valid, or volcanoes do not effect climate significantly with regard to temperature.
Natural cooling such as from volcanic aerosols are not relevant to the 1988 NASA/Hansen congressional testimony. That report unambiguously predicts that CO2 green house warming will rise above natural variability by the 1990’s. The 2008 scientific test of the hypothesis shown in the 2nd and 3rd figures above falsifies the prediction.
It’s interesting reading comment sections in the various blog posts on the climate sciences right now. One can spot the posts of a certain type of mentality (i.e., Mike C); personal abuse, rage, dismissiveness, arrogance, no counter-arguments or evidence, or at best offering a link or two, as if that dismisses all the complex issues raised in the article.
Fred J and Will Nitschkey, both of you need to look at the format of this blog again… the name of the poster is below the line that is below the post… don’t get me mixed up with the poster using the “Scientist” screen name.
I found similar problems with “backcasting” when I made a critical analysis of various climate reports written by Australia’s CSIRO. Please see http://mclean.ch/climate/EE%2017-1_03%20McLean%20ok.pdf
The climate models all rely on the correlation with carbon dioxide for their predictions, never mind that temperatures started rising TWENTY YEARS after carbon dioxide levels were known to be increasing. It strikes me that the output of the models might have been just as close to observed temperatures if a factor for global population, urban population or global wealth was included.
Perhaps the following point is obvious, but I did not notice it being explicitly addressed. A key reason why GCM are able to achieve great backcasts is the ability of aerosols to act as dummy variables in the models. Perhaps several of you have had similar experience as I have had in building and / or evaluating models. You can get great fit by inserting dummy variables, and as a result, there is great emphasis on the impact of the drivers on the dependent variables. (Of course, if you cannot come up with legitimate values of the dummy variables in the future, then your forecasts will be unreliable.)
I have spent considerable time examining the sources of aerosol values that GCMs use. First, I am impressed with the amount of data that is out there on this issue. However, at the same time, aerosols are a gold mine of opportunities to insert dummy variables and hypothesized relationships. One can easily choose values that will produce a great fit and be misled into believing that other drivers (such as CO2) are responsible for changes in the dependent variable.
Researchers should be allowed the ability to update and refine models based upon best available data. For example, some early GCMs forecasted greater warming at both poles. However, as new data of observations are inserted, models now generally forecast warming at the North Pole and delayed warming at the South Pole. Of course, this dichotomy is driven by recent observations forcing relationships to change inside the models, but it probably would be more difficult for the dichotomy to emerge if not for the presence of “dummy variables.”
The author of the blog is quite generous in comparing Hansen’s 1988 forecasts with GISS. Only in climatology would it be acceptable for a forecaster be allowed to validate his model results with his estimate of observed values — when his estimate of observed values are not reproducible nor verifiable. (To some degree his estimate of observed value are verified by its similarity to HadCrut estimate values, but HadCrut also has issues with UHI which apparently have been examined even less than in GISS. And we can get into comparison with satellite data which I am not going to address in this post.)
Take a look at this summary of the hockey “schtik” controversy.
http://bishophill.squarespace.com/blog/2008/8/11/caspar-and-the-jesus-paper.html
Very interesting. Bet we won’t see this discussed in Science or Nature ED pages.
Someone above indicated that the new GCMs no longer use plugs. Well…a rose by any other name is still a rose. The old models used plugs, which we can think of as irrational variables, or variables inserted without physical reason to prevent the models from producing impossible results. The new models use rationalized variables to make the models work. Rationalized variables = irrational variables + an excuse. The excuse is simply a speculation on a physical process with little or no verifying evidence. Water vapor and aerosols are the most notable rationalized variables, but I am sure that there are many others. In the models, all rationalized variables lead to warming with increasing CO2, meaning the intent of inserting the rationalized variable is not to make the model better at predicting climate, but to support the intial assumption of an impending, man-made, global warming crisis. In short, the models are an advocacy tool (propaganda), and worthless for predicting future climate, although they do have some worth as process models, showing us how many of the countless assumptions in the models are incorrect.
The rationalized variable of ‘aerosols’ is key to the hindcasting ability of the models, but this is easily refuted by looking at the Northern and Southern Hemisphere seperately. Unlike volcanic aerosols, which are injected into the stratosphere, spread out globally and hang around for several years, man-made aerosols stay in the troposphere, where they are mostly rained out in a matter of days. Because of this short life, man-made aerosols where primarily confined to the Northern Hemisphere, where they mostly originated during the 20th Century. If the cooling of the mid 20th Century was caused by human aerosol emissions, then the cooling should have been confined to the Northern Hemisphere, just like the aerosols themselves. The Southern Hemisphere, with rising CO2 and little aerosol content should have kept warming. It didn’t! It cooled in lock step with the Northern Hemisphere. The rationalized variable does not fit the observations and should be removed. It is not being removed because this is no longer about science. In fact, it hasn’t been about science for much of the past 20 years!
As a modeller, I’m not at all impressed by the “fit” which has been tweaked to perform relatively well only from 1950 to 2000. And the tweak possibilities is huge since it is based on negative forcing by aerosols (see eg modelE of GISS) whose profiles differ from one model to another so this kind of backcasting has no merit.
The fact that the “fit” is near nul from 1900 to 1950 and the natural simulation (blue curve) is truncated after 2001 is a good signal of cherry picking.
No matter how hard they try to spin the presentation, the science to support the theory of anthropogenic influence on climate is non-existent.
Scientist, climate scientists should care about compressible fluid modelling because, just in case you didn’t notice, air is a compressible fluid. Atmospheric models are essentially vast, very complicated models of a compressible fluid system. Oceanic models use incompressible fluid modeling, and aerosols and precipitation constitute multiphase flow. In other words, climate modelling is fluid modelling.
what is your source for assessing their mathematical backgrounds?
I’m a mathematician by education, and in my PhD work, I have discovered that much of the current state of fluid modelling rests on a shaky (or sometimes non-existent) mathematical basis. For example, when someone models the entire inertial subrange a chaotic dynamical system with regression fit or a stochastic differential equation with experimentally-derived coefficients, there are bound to be significant errors. Since lots of people in fluid dynamics don’t appreciate the mathematical nuances of the Navier-Stokes equations, they pull stuff like that and don’t understand why they can’t get rid of the errors.
These climate models have huge amounts of experimental correlation, far more than we use in the aerospace industry, and that’s just bad fluid dynamics. With models that dependent on regression fits, I didn’t trust their results past three or four years. The data has since confirmed I was right.
Josh,
Absolutely, frickin’ bravo!
A plain demonstration of the difference between those who are paid to create things that actually work, and those who dropped out from reality, generating mounds of fiction furthering a political agenda so the politicians will pay them to write still more steaming mounds.
Those who can do. Those who can’t teach. Those without a clue preach!
Getting back to the topic (sort of)
The question was, “what the hell happened in 1955?” My question is, why should we only see ‘enhanced’ global warming above a certain level of CO2? I can see that at average temperatures of freezing or below, water vapor cannot exist in great quantities, and at CO2 levels below about 200ppm, vegetation starts dying off, so inhibiting any further downward trend in CO2. But why should the ‘enhanced’ greenhouse effect only kick in at above a certain level, and where is this level? Is it supposed to be above 400ppm, above 350ppm, above 300ppm, or where? Wy should the ‘enhanced’ signal not be detectable at any level, down to pre-industrial levels? Surely if the enhanced greenhouse effect exists, we should have been able to measure it by now.
Keith – The general perceived effect of volcanic eruptions tends to last only about five years, and you accept that as a valid climate/temperature event, so a trend that is twice as long should be just as valid. You can’t have it both ways, so which petard to you wish to be hoist on? Either ten years is valid, or volcanoes do not effect climate significantly with regard to temperature. – in the real world, one cannot understand complex systems in terms of childish either/or questions like this. The question is how long one needs to discern the influence of a forcing among the unforced variations inherent in the climate system. Adding 2ppm of CO2 to the atmosphere each year increases temperatures by about 0.2°C/decade. Adding 17 million tonnes of SO2 to the stratosphere decreases temperatures by 0.5°C within a year. Unforced variations are of the order of ±0.5°C annually. Given that one effect is 25 times larger over a single year than the other, and therefore is clearly detectable over a much shorter timescale, your either/or question is entirely meaningless.
climate modelling is fluid modelling – not really. Equations of fluid motion are in climate models, but only as one input, and necessarily one which is integrated over the timesteps involved.
much of the current state of fluid modelling rests on a shaky (or sometimes non-existent) mathematical basis – point us to a publication in which this is discussed, would you?
models that dependent on regression fits – climate models do not depend on regression fits.
Peter – your question doesn’t seem to make sense. The enhanced greenhouse effect, if you wish to call it that, has been detected. It is the reason for the dramatic warming since 1975. Who ever said anything about only seeing ‘enhanced’ global warming above a certain level of CO2?
Non-scientists often have difficulty understanding the forest for the trees, especially in data-rich climatology. The most elegant arguments in physics and mathematics are the simplest arguments. “Backcasting” and “back-predicting” climate using climate codes are circular arguments. Because the codes are data fits to begin with, there is no surprise that they match existing data!
The Climate Skeptic article here seems to understand this but should have said so explicitly. While it is great to point out that James Hanson, PhD has never been able to predict future climate accurately using the feedback formalism that he developed twenty years ago because it is probably wrong, the computer codes that utilize it have so many free (adjustable) parameters that they can fit any data and predict any future their authors desire. This makes them an exercise in virtual reality akin to a Hollywood movie: real in some respects but completely subject to the whims of their Directors. They prove nothing.
Circular arguments and meaningless data fits using large numbers of free parameters are extremely fundamental errors in physics and mathematics.
As to someone in this discussion claiming to be a scientist and substituting profanity for insight, you can draw your own conclusions. Those of us who have always given colleagues the benefit of the doubt regarding honesty are horrified by the bad behavior, clearly not limited to this one instance.
To explicitly address the huge issue of accountability, I think everyone needs to sign his or her name to discussions such as this. My name is Gordon J. Fulks, PhD, and I am a physicist. I can be reached at gordonfulks@hotmail.com.
much of the current state of fluid modelling rests on a shaky (or sometimes non-existent) mathematical basis – point us to a publication in which this is discussed, would you?
OK. You could start with Ruelle & Takens’ 1970 paper On the Nature of Turbulence, compare to the basic formulation of the k-e turbulence model, and draw your own conclusions. If you want to get a little deeper than that, Mittal’s paper in Journal of Computational Physics mentioning the complete inability of Spalart-Allmaras to predict airfoil stall is useful. There’s a start. But what you should really do is read McDonough’s paper, and here’s the full citation:
J. M. McDonough, “On intrinsic errors in turbulence models based on Reynolds-averaged Navier–Stokes equations,” Int. J. Fluid Mech. Res. 22, 27–55, 1995.
He outlines a bit on the fundamental problem with modeling turbulence as a purely dissipative term, which is also one of the core features behind LES, although the focus of the paper is RANS. I would the books Turbulence by Uriel Frisch and Chaos by James Gleick. Combined with Ruelle & Takens, that should get you started.
not really. Equations of fluid motion are in climate models, but only as one input
That is completely the wrong way of looking at it. The climate is a fluid system with a lot of complexities. The momentum and energy equations are the fundamental model; everything else is an addition on top of that.
climate models do not depend on regression fits.
Almost anything in a climate model that is not a PDE is basically a curve fit to experimental data or a “fudge factor”…like the CO2 forcing term and the cloud cover term.
It’s helpful to provide links to papers, if you can. I’ll look them up later, but in the meantime you ought to read some papers about the radiative properties of CO2, going right back to Tyndall and Arrhenius. To describe our knowledge of CO2 forcing as a ‘fudge factor’ suggests that you know nothing at all about even the extreme basics of climate science.
Scientist: “your question doesn’t seem to make sense. The enhanced greenhouse effect, if you wish to call it that, has been detected. It is the reason for the dramatic warming since 1975”
So then, what happened to the enhanced greenhouse effect prior to 1975? Where was the dramatic warming? We had cooling between the ’40s and the ’70s. Oh sorry, I forgot – it was the aerosols which caused that cooling, wasn’t it? Except, in that case, we didn’t have dramatic warming after 1975, temperatures were simply recovering from the aerosol effect, weren’t they? So I’ll ask again: where was the enhanced greenhouse effect?
To imply that the CO2 forcing term in climate models is chiefly based on radiative emissivity suggests that you know nothing at all about even the “extreme basics” (what is an “extreme basic?”) of climate science. But you know and I know that the main issue is this “feedback loop,” not a direct radiative warming. The small amounts of anthropogenic CO2 in the atmosphere aren’t nearly enough to radiate the earth enough to account for the warming trend between ~1960 and 1998.
By the way, since I get my science from print journals instead of blogs and press releases, I don’t have URLs for papers readily on hand. Real scientists look things up based on references. When I publish my note on the RANS equations, though, I’ll be sure to send you a copy, how about that?
The enhanced greenhouse effect, if you wish to call it that, has been detected.
The warming profile in the atmosphere is completely inconsistent with a CO2-enhanced greenhouse effect. So no, no enhanced greenhouse effect has been detected. AFAIK, we don’t know what we’ve detected.
Peter – are you trying to be stupid? Do you understand tha CO2 is not the only thing which affects the climate?
Josh S – I do not know of a single research journal which is not available online. Looks like you don’t know where to find the latest research. Your talk of ‘the CO2 forcing term’ shows that you haven’t got a clue how climate models work, and you try to dig yourself out of your hole by pretending you weren’t talking about the direct radiative forcing but feedbacks instead. So, you know fuck-all about models. What do you know about observations? Also fuck-all. The warming profile in the atmosphere is completely consistent with a CO2-enhanced greenhouse effect.
AFAIK, we don’t know what we’ve detected. – what is this royal ‘we’? On behalf of whom do you admit your collective ignorance? I am sure you didn’t detect anything yourself, so this sentence is just bizarre, really.
It seems to me that hind-casting or back-casting is a lot like a clairvoyant telling you what happened yesterday. Since you already know what happened, you can adjust all the parameters to make the picture match what happened. The real proof of the pudding is to forecast, then wait and see if your output was correct. That is the only proof that matters. Unless of course, you can find a group of modellers who are totally ignorant of the past and remain so throughout the process.
And, in closing, feeding trolls just makes them hang around longer. I agree completely with Gordon Fulks, real names provide more credibility and accountability. My name is really John Nicklin, not a stretch from my “handle” here.
Scientist: Like I said, I get my research from print journals, meaning that I cut and pasted references from my own papers. I didn’t say the journals aren’t online. I could go onto any of the databases and find the URLs for you, but the point is that I don’t feel like it, and if you’re really a scientist, you should have no problem with looking up whatever papers you want to read should you actually intend to read them. Not every journal has online archives going back to the 90s, by the way. However, Ruelle & Takens’ paper is easy to find, as it’s cited in most papers on turbulence theory.
pretending you weren’t talking about the direct radiative forcing but feedbacks instead.
And here I thought the various feedbacks are supposed to be one of the things that CO2 is forcing, you know, like Andrews and Forster mention here:
http://www.agu.org/pubs/crossref/2008/2007GL032273.shtml
But I suppose I was wrong. Either way, the effect of CO2 on the climate is not directly derived from fundamental physical laws. For example, in the paper above, we don’t have any governing equations derived from fundamental physical laws for the formation of clouds. Those models have to be derived via observational correlation, which may very well take the form of a hypothesized ODE with “tweaked” coefficients (like the famous k-e turbulence model). But mathematically, that’s still a correlation…just a few steps fancier than linear regression. And because a turbulent fluid is a chaotic dynamical system, I don’t expect it to work very well. It doesn’t work well with energy dissipation; so why should I expect it to work well with cloud formation? AFAIK, that’s how all the CO2 forcings work.
Of course, I’m talking to someone who thinks that fluids are merely “one input” into predicting climate.
what is this royal ‘we’? It’s the unnamed actor in your passive “has been detected.”
so this sentence is just bizarre The collective “we” is not uncommon in English. For example, one might say “We really love hamburgers in the USA,” even though he himself might not like hamburgers, and not every American likes hamburgers. One also might say “We’ve been fighting in Iraq for a long time,” even though he himself had never been to Iraq. A mother might say to a child, “Are we not going to eat our dinner?” even though her own plate is clean. Another example is the old adage, “We can put a man on the moon, but we can’t solve [fill in your social problem].” Hope that clears things up.
By the way, since the CO2 forcing models end up with a completely incorrect atmospheric temperature profile, that would suggest that there’s a problem somewhere.
Quoting Scientist – “in the real world, one cannot understand complex systems in terms of childish either/or questions like this. The question is how long one needs to discern the influence of a forcing among the unforced variations inherent in the climate system. Adding 2ppm of CO2 to the atmosphere each year increases temperatures by about 0.2°C/decade.”
So, you are saying that a decade IS long enough to detect a climate trend? Your scale of effect in this quote is based upon a decadal scale, and the effect is of a measurable quantity. So, if a decade is long enough for this effect, then our observations that there has not been any perceptible warming over the last decade is valid as well. Thanks for confirming this, Scientist.
Scientist: “Do you understand tha CO2 is not the only thing which affects the climate?”
Good Lord! He’s finally woken up!
Josh S – well if you don’t feel like linking to papers in a useful way, I rather suspect you’re just quoting irrelevant papers in the hope that it will make it look as if you know what you’re talking about. Your stroppy teenager attitude doesn’t give me any reason to believe that you have any science background to speak of.
the effect of CO2 on the climate is not directly derived from fundamental physical laws – yes it is. You misunderstand the way clouds are treated. Their formation cannot be treated directly within current climate models because the spatial resolution available is much larger than clouds. So, they have to be parametrized. If you think you know a better way, I’m sure a journal would love to publish your paper. The parametrization of a phenomenon in a model with limited spatial resolution does not mean that the operation of the phenomenon at higher spatial resolution is not understood.
The point about your bizarre ‘we’ is that it is analogous to saying ‘we don’t know why the Earth orbits the Sun’, or ‘we don’t know why helium is an inert gas’. We do. If you don’t, then your ignorance is nothing to be proud of.
CO2 forcing models end up with a completely incorrect atmospheric temperature profile – no they don’t. This claim is simply a fantasy of the Luddite deniers. Have you read any actual literature on this?
Keith – what a tiresomely thick person you are, really. Do you understand what I explained about natural variations? Do you see that some changes are larger, over a short period, than natural variations? And the other changes are smaller? Which ones do you think we’ll detect, and which ones won’t we detect? If you can get over your childishly black and white approach to science, I’m sure you can grasp this.
Josh S,
“Scientist” is the guy who ignores gravity laws and who repeat everywhere it’s “positive feedback”. So giving him ref. to papers discussing hydrodynamic systems is a complete waste of time.
He is not interested in sincere discussions, only in trolling.
Dear Peter:
The 1988 NASA/Hansen hypothesis presented to congress accounts for natureal sources of warming. It is now falsified by 20 years of measurements. That’s all that need be said about the hypothesis. There remains the question why it is false.
Nick Bottom – your beliefs about the relation between observations and models over the last twenty years are incorrect. Read some actual science.
Scientist, if the data point cannot be detected or observed, then it is a not a data point. Science is supposed to be about observation and the accumulation of measurable facts, then interpreting those facts to explain what has been observed. If the interpretation seems valid, then you see if it can predict an outcome and is replicable. This allows your hypothesis to stand the test of debate.
You are the one providing the statistics that support my view that ten years is a sufficient period for observing a climate trend. You are the one that said that an increase in carbon dioxide can be observed as increasing temperature 0.2 degrees centigrade in a decade, not me. If this is so, then reason says an analysis of a ten year period should be sufficient to provide a trend that is as valid as your statement. This is logical reasoning and the value of semantics. Words do mean things, and you provided the words.
I’m still wondering how Church and White discovered that 7 to 19 micron increase. Did they have some hyper-efficient measuring device to measure a change that is roughly the thickness of one or two red blood cells? Did they have tide gauges with microscopes and microscopic gradations to measure down to the millionth of a meter? It is just such a subtle measurement they were able to accomplish.
“It seems to me that hind-casting or back-casting is a lot like a clairvoyant telling you what happened yesterday.”
I totally agree. It’s like followers of Nostradamus who interpret his quattrains after the fact. The real test is whether you can make predictions without the benefit of hindsight. At this test, the warmists have failed miserably. Just as “psychics” have failed miserably.
And by the way, it’s interesting to note the double standard at work. If temperatures had been rising steadily since 1997, the warmists would be saying “Aha!!! Anthropogenic Global Warming!!” But if temperatures are flat or declining, they claim a 10 year period is insignificant.