Regarding the latest UAH and RSS global temperature data plots Dave B writes: “…could you post a best-fit, to be fair? I don’t have the technology.”
Sometimes I’m tempted to tell people to do the work themselves, after all, I’m overloaded as it is. But, it is the 4th of July weekend, and I’m stuck here in the smoky toasty Sacramento Valley babysitting a bunch of servers until my chief tech support guy comes back from vacation, so what the heck.
I’m not sure what he’s implying by “fair” but it has been my experience that no matter what you put in a graph, or how you graph it, somebody will find fault with it. Below are raw data overlaid with 1st order and 5th order curve fits to show long and short term trends.
Click for large plots
And “to be fair”, and to make everyone happy/angry here is the last 11 years, when the warming trend flattened.
Click for a larger image
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Ed Ried,
Thanks for the link. The simple Sunspot model I mentioned above accounted for volcanic eruptions visually (i.e. where the deviations were below the model fit, I looked for some volcanic eruptions in the same time frame). That one chart has in one place all the scattered volcano data I found in my brief google-quest. I’m keeping the chart for reference.
Overall, the approach of the linked chart “appeals to my intuition” as a fledgling “climate aficionado”. Of course, “appeals…” does not mean its true — but, I like to start with something intuitive and see if it stands up to progressively closer scrutiny.
Anthony, yours is a very worthwhile site for information sharing.
Nice write up Anthony!
The Engineer.
I have a BS. MS. and Ph.D. in Mechanical Engineering. My Ph.D. topic was related to turbulent particulate flows. So, I’m accustomed to the various issues associated with the sorts of random seeming variations one sees in different sorts of transport phenomena.
If you look carefully, to fairly simple statistical tests, of the sort generally taught in undergraduate engineering courses. 🙂
There are reasons I don’t try to immediately apply a fourier transform. Some have to do with the phenomenological explanations suggested by the IPCC.
DaveK (00:04:33) :
“Higher order polynomial regression curve fits are great for smoothing data, and to see trends within the scope of that data. But if you have ever tried to use them to extrapolate outside that scope, you will know the perils of that exercise. Once you exceed the boundary conditions of that fit, all bets are off, and the higher order the polynomial, the less useful it is for predictions beyond the scope of the data, especially when the so-called “dependent variable” (the y axis value) has little dependence on the “independent variable” (the x-axis).”
That’s worth restating, so I did. You probably don’t want to put much faith in the left and right ends of the data either.
It would be amusing to take the same time frame, do a 5th order fit over the middle 2/3s or so of the data but display the curve for the whole time period. I bet it would dissuade people from extrapolating such curves in the future!
Thank you very much for the graphwork, Paul. I have bookmarked your site and will check it often!
And that moving average graph tells me something very useful – which is that we are at an extremely crucial time period *right* *now*. If we are to remain in the relatively stable temperature regime of the last 8 years or so, then temperatures need stop dropping pretty quickly. If we are to see evidence of long term warming, temperatures need to start heading back up rapidly. However, if over the next 6 months temperatures continue to drop at the same rate as they have the last 6 months, then we will plunge back to levels not seen since the low around 1985 and the mid-70’s, in which case I imagine we will begin to see some dramatic observable confirmation.
Any one of those outcomes is possible; but over the next 6 months only 1 of them can turn out to be correct.
I’d like to make two basic points:
1. The logic used to suggest that Mr. Hansen believed in global cooling is incorrect. The evidence is that a man who DID argue in favor of global cooling (Mr. Rasool) used a computer program written by Mr. Hansen. The fact that Mr. Rasool used Mr. Hansen’s program provides zero evidence about Mr. Hansen’s beliefs. Mr. Hansen is no more responsible for the global cooling hypothesis than the manufacturer who supplied the hardware on which Mr. Rasool performed his computations. The statements made here about Mr. Hansen are slanderous. Moreover, they are classic ad hominem attacks: arguing that the AGW hypothesis is incorrect because one of its advocates is an evil person. If Osama bin Laden endorses AGW, that doesn’t make it wrong. If Mother Teresa opposes AGW, that doesn’t make it wrong. Let’s dispense with this National Enquirer gossip and concentrate on the facts.
Second, on the matter of the most useful curve to which to fit the data: we should rely on Occam’s Razor to guide our considerations. We want to go with the simplest hypothesis. If we represent our data with a polynomial, then the complexity of the polynomial is directly commensurate with the degree of the polynomial. In other words, y=ax+b is simpler than y=ax + b(x**2) + c. Occam’s Razor argues strongly in favor of the first order least squares fit over the fifth-order fit. This works in favor of the AGW hypothesis.
REPLY: I’m sorry, your story in number one is just absolutely ludicrous. You’d have to assume that they never talked about it or collaborated on it. If Hansen didn’t believe in global cooling at the time then he’d likely not have allowed his connection to it in any way. University research departments are tight centers of collaboration, and if one scientist is in disagreement of others works in the department, they usually don’t collaborate to create a finished product. These two people were just office doors apart, not on other sides of the world where one found a tool to use created by the other and there was minimal collaboration.
It doesn’t much matter what James Hansen, other commenters, or you believe for that matter, nature will be the final arbiter on the AGW issue.
There were lots of university meteorology departments that started research into global cooling at the time, so it would be no surprise that Hansen and Rasool were following the same path as many other departments at the time. It is your logic that is flawed.
– Anthony
I’d like to respond to Basil’s comments about periodicities. I think they represent a common logical error: scanning complex time-sequenced data for patterns. The most obvious examples of this come from financial analysis. There have been lots of voodoo artists who crunched the data trying to predict the markets. They’d fit complex polynomials or various periodic functions to the data, twiddling the coefficients, until they got a good fit. Then they’d base investment decisions on these models. And surprise, surprise, half of them would be right and make money! The other half would lose money and declare that they just needed to make a few more adjustments. The problem here is that they’re playing numerological games, not serious analysis. If you mindlessly fit functions to data, you get garbage. The important thing is to have an underlying mechanism to justify whatever function you attempt to fit to the data.
Where you can justify a function with a mechanism, then you’re safe to proceed. And sometimes you can use random function-fitting to work backwards in an attempt to discover a mechanism. If your data shows a apparent periodicity, then it’s worthwhile to inquire into the possible causes of that periodicity. If you can’t identify a mechanism, then any attempt to project that periodicity into the future is vulnerable to the possibility that you’re looking at some sort of statistical artifact.
There are some periodicities in climate whose underlying mechanisms have been worked out. I don’t question them. What I am warning against is numerology as opposed to scientific analysis. GIGO.
Ophie,
It is your colleagues at “real”climate that are the numerologists.
It is they that not only are scanning complex time-sequenced data for patterns but also are injecting those patterns into the data for them to find.
Sorry if this is a bit vitriolic, Anthony, but I’ve just been perusing the Australian “Daft” Garnaut report. I just keep on having this mental image of Emperor Canute, naked, telling the tides to stop. How can people be this stupid? Answer, they are not, they are dishonest.
It is your colleagues at “real”climate that are the numerologists.
It is they that not only are scanning complex time-sequenced data for patterns but also are injecting those patterns into the data for them to find.
Fine. So let’s not do it here, OK?
REPLY: Opie, I’ll be the one to choose what does and does not get discussed here, not you. If you want to be able to make those choices yourself, get your own blog.
Discussion of the issues about time sequenced data is of interest to a large majority of people here, and it was you that brought up the issue. Thus the issue and discussion is fair game. – Anthony
I am extrapolating from Oph’s views that the AGW models assume that CO2 is the mechanism. From what little I know about scientific investigation of modeling (I know more about the gold standard scientific method, having used it myself), I would think that the group that came up with the AGW model understood that “A” model is built to explain a theory, not test it. If a theory is being tested, a control model (or even better, several competing models) would be advantageous here (I am also extrapolating from Oph’s view that the CO2 theory has not been proven yet and they are still in the testing mode). That would lead me to suggest to the AGW investigation group that they need several models going at the same time with different mechanistic theoretical assumptions for each.
The research literature is awash with climate theories, not just one climate theory. I don’t understand the eggs in one basket method unless the group is not an investigative group and is instead an agenda group.
Anthony dismisses my argument with speculative comments for which he offers no evidence. For example, he claims that These two people were just office doors apart
If you know that to be true, then surely you can provide us with their office numbers. Or are you just making it up?
[REPLY: Same department, same building at Columbia, strong circumstantial evidence of being just doors away. Office numbers? Heh, funny man that Opie. I challenge you to find and present here the department course catalog from Columbia that year. Office numbers will likely be there.]
f Hansen didn’t believe in global cooling at the time then he’d likely not have allowed his connection to it in any way.
That’s speculation for which you have neither evidence nor logic. You’re just making it up.
[REPLY: Oh the logic is there. Speculation perhaps, but supported by cicrcumstances. ]
if one scientist is in disagreement of others works in the department, they usually don’t collaborate to create a finished product.
You have no evidence that they collaborated. The evidence you have states that Mr. Rasoor used Mr. Hansen’s program. Scientists grab code from each other all the time. Ever heard of good old SPSS? In my research I took some functions from another fellow who had nothing to do with my project. Borrowing code in part or in entirety does not establish collaboration. You’re fantasizing a connection that is not derivable from the evidence.
[REPLY: You mean you used his code without even bothering to tell him or ask permission? Seems like plagarization to me.
The fact that Rasool used code from Hansen says they collaborated at some level, unless you are ready to say that Rasool used it without notice of any kind or even asked for instructions on how to use it. Highly unlikely in a university department envrionment. Again I submit that if Hansen didn’t agree with the research being conducted by Rasool, he probably would not have helped or agreed to provide the Mie scattering code because he would not want his name associated with a global cooling paper.]
It doesn’t much matter what James Hansen, other commenters, or you believe for that matter, nature will be the final arbiter on the AGW issue.
Indeed so!
There were lots of university meteorology departments that started research into global cooling at the time
Yes, indeed — and do you know what their conclusions were? They pretty much trashed the idea. Remember the NAS, which I mentioned earlier as having never made an erroneous report? In 1975, they issued a report on the question entitled “Understanding Climate Change: A Program for Action”. Basically, they said that “we do not know enough about climate to make predictions”. And by the way, the Rasool paper did NOT predict global cooling — it presented a mechanism by which global cooling was possible.
[REPLY: “They pretty much trashed the idea.” No, the PDO did that for them in 1978, Nature ruled the day. They were well along in the research for cooling and then nature threw the curve ball, they didn’t know why, they started to look for answers, CO2 was centered upon. PDO wasn’t discovered until much later. Nature leads, research follows, it has always been that way. To suggest that they weren’t on that path in the 70’s and that they came to the conclusion without nature’s lead is just ludicrous on your part. ]
This whole story about scientists predicting global cooling is a load of lies. It didn’t happen that way.
[REPLY: Load of lies? To use your own words, “no logic, speculation”. Oh it did happen that way, I was in the middle of it at the time. It’s not lies, its being spun now for the current view. I suppose now you’ll cite the recent Petersen ex post facto paper as “proof”.
Opie, please answer this question: how old are you? ]
Regarding the temperature time series, it is clear that there are a number of variables that have to be factored in to explain it properly.
First and foremost, there is an ENSO index variable that greatly impacts the global time series. Major El Nino in 1997-98, 1987-88, 2000-2005 and 1983-84, minor La Nina events in 2007-08, 1989 and 2000.
Secondly, there are volcanoe impacts with Pinatubo in 1991and El Chichon in 1982.
If one could factor these out in some manner, the time series would have a much different slope etc.
Opie, I’ll be the one to choose what does and does not get discussed here, not you. If you want to be able to make those choices yourself, get your own blog.
You misunderstand. Robert Wood criticized an error that he perceives at RealClimate. I suggested that the error not be repeated here. My verb “do” referred to the error, not the discussion.
Occam’s razor argues for the simplest but ‘simple’ needs to be clarified. A linear trend, while ‘simpler’ is not always at the top of the list of preferred trend decriptors in nature. Exponential/Logarithmic or periodic curves are more common and hence might be a better fit as per Occam.
Great blog! I read it often but this is my first post here.
~peace~
A linear trend, while ’simpler’ is not always at the top of the list of preferred trend decriptors in nature. Exponential/Logarithmic or periodic curves are more common and hence might be a better fit as per Occam.
If you have a model that suggests exponential, logarithmic, or periodic functions, then yes, you should apply that model. If you don’t have such a model, and you are inquiring into the first derivative of the function, then a straight line is the simplest function. Do you have a model suggesting some other function?
Brendan: I looked at your link to the work of T.J. Nelson, reading some of it and skimming the rest. He is to be commended for good organization of material, impressive gathering of relevant data, and clear writing. However, his scientific logic has some weak points. See the following quote:
QUOTE:
“Between 1900 and 2000, atmospheric CO2 increased from 295 to 365 ppm, while temperatures increased about 0.57 degrees C (using the value cited by Al Gore and others). It is simple to calculate the proportionality constant (call it ‘k’) between the observed increase in CO2 and the observed temperature increase:
ln (365/295) = k * 0.57
k = 0.3735
ln (2) = 0.693 = k*deltaT
deltaT = 1.85
This shows that doubling CO2 over its current values should increase the earth’s temperature by about 1.85 degrees C. Doubling it again would raise the temperature another 1.85 degrees C. Since these numbers are based on actual measurements, not models, they include the effects of amplification, if we make the reasonable assumption that the same amplification mechanisms that occurred previously will also occur in a world that is two degrees warmer.”
END QUOTE
Several problems: First he is assuming that 0.57 is the actual increase — but UHI is a concern, not to speak of data manipulation.
Second, he is assuming that all increase is due to CO2, but there could be other causes for temperatures to rise in the 20th century such as land use, solar variation, recovery from the LIA, . . . .
Third, he would get a different result if he based his calculation on the last 20 years rather than the 100 years in the 20th century. Or the results would be even more scary and dramatic if he took 1900 to 1940 to establish his k.
In two REPLYs to a previous comment of mine, one person flatly rejects my statement that subsequent research “pretty much trashed” the idea of global cooling. Yet I cited an NAS report from 1975 that stated that we don’t know enough to make such predictions. The replying person states that only the discovery of the PDO in 1978 put an end to the global cooling hypothesis, yet the NAS report in 1975 — three years before the discovery of the PDO — said that there was no basis to make predictions of warming or cooling. The facts contradict the claim made by the replier.
[REPLY: just because NAS issued a report in 1975 doesn’t mean anybody paid attention to it or dropped all the research about it. Global cooling research went on at many universities into the early 1980’s, NOVA did a show about it in 1979.]
A second replier insists that the global cooling brouhaha did indeed take place as the skeptics claim, and provides as evidence the assertion that he was there. I suggest that a somewhat higher standard of evidence would be appropriate here. Please cite the scientific papers that predicted global cooling. On that question, I did find one interesting tidbit here:
http://www.skepticalscience.com/argument.php?a=1
As far as peer reviewed scientific papers in the 1970’s, very few papers (7 in total) predicted global cooling. What surprises is that even in the 1970’s, on the back of 3 decades of global cooling, significantly more papers (42 in total) predicted global warming due to CO2.
Lastly, you ask how old I am. I’m 58 years old. I assume that you ask the question because you think that I’m some young punk who doesn’t know his arse from a hole in the ground. Well, I was “there” too — in fact, I was in grad school studying physics at the time. And in fact, in my planetary atmospheres course, the instructor spent plenty of time on the evolution of planetary atmospheres, including heating and cooling factors. The starting point was the greenhouse effect, discovered in the 19th century. Venus is the perfect example of the potential magnitude of the effect. My instructor discussed the possibility of the same effect here, and the standard wisdom at that time was that human emissions of CO2 should in fact lead to temperature increases — but that there was no observational evidence to support what the theory predicts. The overall conclusion was “it may well be happening, but we don’t have the data, so we’ll have to wait and see.”
This from a physics professor who was teaching a course on planetary atmospheres. I think it represents the conventional wisdom in 1974. So you can see that I really do know what I’m talking about when I say that these stories of scientists believing in global cooling are a pack of lies.
[REPLY: The “pack of lies” which is your opinion is contradicted by at least 7 papers in the 70’s that you cite above. So are you going to claim those papers were lies now? The scientists wrote them but did not believe in what they wrote? As I predicted, in citing skeptical science you inadvertently cited the Peterson 2008 paper, which is spin central. It was written to quash the idea.
The tendency that you have to use terms on this forum such as “deniers” and “pack of lies” suggests that you are more about passion and less about facts.]
Okay, this “global cooling” discussion is getting way out of hand. Obviously, I’ll refer everyone to RealClimate’s “Global Cooling” article, but another comment is important:
What the hell does it matter if, in the infancy days of climate science, climatologists mistakingly believed that there was going to be an imminent period of cooling? Seriously, what does it matter? I was cleaning out my basement the other day and I came across a PC World article from the month when the very first 1.0 GHz machines went on the market; in that article, the writers suggested that it wouldn’t be long before we hit the 10 GHz benchmark. Obviously, they were wrong, as we are now fully aware that heat issues prevent clocking of single core CPU’s that high, which is why we have moved to a multi-core paradigm in computing.
Should the PC World writers from that month forever hold the stigma that they made an erroneous prediction based on a more primitive knowledge base than we have now? Of course not. Yet you’re doing the same thing.
Referring to the Global Cooling mole is an ad hominem meant to attack the scientists involved in AGW research, and does nothing to refute or further any argument against the AGW theory and its predictions.
The tendency that you have to use terms on this forum such as “deniers” and “pack of lies” suggests that you are more about passion and less about facts.
With all due respect, Anthony, many commenters here enjoy lambasting Hansen and other public AGW-related figures, and love to propagate patently false lies about the AGW theory or its proponents. If you are going to outlaw the use of the word “denier” (which, in the case of your posters who refuse to believe that CO2 can act as a GHG, is a most appropriate term), then it would be courteous of you to outlaw the word “alarmist,” as, to use your own words, that term “suggests [the commenter is] more about passion and less about facts.”
REPLY: The issue with “denier” is it’s relation to the Nazi holocaust. “Alarmist” has no such connotations.
“which, in the case of your posters who refuse to believe that CO2 can act as a GHG, is a most appropriate term”
Let’s see a show of hands: who deosn’t believe CO2 is a GHG?
Don’t think you’ll find anybody. Lots of people here doubt CO2 is the sole cause though, inclding me. – Anthony
The “pack of lies” which is your opinion is contradicted by at least 7 papers in the 70’s that you cite above. So are you going to claim those papers were lies now?
No, they were a small minority. You seem to be assuming that if a single scientist advocated global cooling, then the idea had currency. There are always weirdo ideas floating around the scientific community, largely because (despite the claims of those with a political agenda who are unaware of the truth) scientists are extremely open-minded about scientific issues. If you can make a good case for something, you’ll get it published. And if you can make a compelling case for something that everybody else rejects, then you get the Nobel Prize.
Remember, there were also 42 papers — six times as many as the global cooling papers — promulgating global warming. If you want to base a case on these papers, then you end up concluding that scientists were 6:1 against global cooling. Therefore, stories that scientists believed in global cooling are a crock.
The tendency that you have to use terms on this forum such as “deniers” and “pack of lies” suggests that you are more about passion and less about facts.
Perhaps you did not read my post on this subject, but I used the term “denier” on my first post and as soon as somebody objected to it, I promised not to use the term again — and I have not used it. Meanwhile, many posters here continue to use derogatory terms towards those they disagree with. Are your standards of civility dependent upon the editorial claims of the writer?
I think “pack of lies” is a correct characterization of a story that is false and yet is often repeated despite proffered evidence that it is false. When people make an honest mistake, they stop doing it once they have been corrected. However, the skeptics keep repeating the global cooling falsehood over and over. I think we’re not talking about an honest mistake here, but rather willful disregard for the facts.
Daniel: I actually agree with this statement:
What the hell does it matter if, in the infancy days of climate science, climatologists mistakingly believed that there was going to be an imminent period of cooling?
However, given that the current round of GCMs are unable to predict/foresee/scenarioize, whatever, anything which can be falsified by observations, since the error bounds apparently encompass conditions ranging from current conditions on Mercury to those on Titan, what makes you think that climate science is still not in a state of infancy? Have you not noticed a tendency over the last 20 years to move the bar every year in order to keep a semblance of credibility? Does that not give you pause?
I’ve been following this issue since about 1981, long before Hansen testified, from back in the days when atmospheric scientists used to publish about negative feedback mechanisms stabilizing climate, and before the current groupthink of positive feedbacks.
Positive feedbacks are not intuitively likely because if the Earth’s climate was dominated by positive feedbacks it would have reached a bifurcation point and run away millions if not billions of years ago. End of story. The climate system has been essentially stable for 100’s of millions of years through atmospheric changes, orbital fluctuations, and solar irradiance changes. If positive feedbacks dominated the system, we would not be alive today.
Oops, I didn’t see this earlier:
REPLY: just because NAS issued a report in 1975 doesn’t mean anybody paid attention to it or dropped all the research about it. Global cooling research went on at many universities into the early 1980’s, NOVA did a show about it in 1979.
You seriously believe that nobody paid attention to an NAS report? Wow… all I can say is… careful… well, I suppose that there were skeptics even then who refused to pay attention to the NAS. But the great majority of scientists hold the NAS in very high esteem. To be invited to be a member of the NAS is a major career plum. So YOU may not have paid attention to the NAS report, but just about everybody else who had any interest in the topic certainly did. Jeez, even *I* remember reading a story about that report.
As to “global cooling research”, I think that you’re conflating climatology research with global cooling research. Many of the factors that were cited in the original Science paper, such as aerosols, were legitimate subjects of scientific research. Indeed, they are STILL be studied as part of global warming research. The IPCC AR4 report includes 25 pages on aerosols. Are you suggesting that the IPCC believes in global cooling because it is reporting on research on a topic raised in conjunction with global cooling?
REPLY: No I’m suggesting that I had firsthand knowledge of scientists at universities in the USA that were still pursuing the idea of global cooling as late as about 1981.
In fact I know of one university researcher who dropped what he was doing in synoptic and mesoscale meteorology after the hard winters in the USA in 1977/78 and stared focusing on global cooling issues.
The 1975 NAS report is like the IPCC report is this way; some paid attention, some took it with a grain of salt, and other just flat ignore it as “science by consensus”.
A good example of science by group consensus that turned out to be flat wrong was the consensus of the solar science panel convened by NASA lat year that claimed we’d be seeing the activity of solar cycle 24 by now.
From this story on space.com where they talk about the opposing views solar scientists have for cycle 24 they offer some opinions. NOAA Space Environment Center scientist Douglas Biesecker, who chaired the panel, said in a statement:
Well, obviously March 2008 didn’t happen. We are still waiting. So much for that group consensus. Data, calculations, models, expert panels, and even SWAG wasn’t able to predict natures whims.
Nature rules, nature confounds, nature outwits.
A dumb question about curve fitting, which I’ve been unable to answer using my old math texts or google:
What’s the general equation for a skewed sine wave?
Thanks to anyone that can help.
Daniel Rothenberg wrote: “Okay, this “global cooling” discussion is getting way out of hand.”
I agree, and as I said in the article; “…it has been my experience that no matter what you put in a graph, or how you graph it, somebody will find fault with it. ”
And look what is had turned into.
So we all earn a time-out. Thread closed for the weekend, we all have better and more fun things to do.
You can read NASA’s version of Hansen’s change in position here:
http://www.giss.nasa.gov/research/features/temptracker/
They seem to minimize his predictions of global cooling, but they write:
“Hansen marveled at the power of these airborne particles, known as aerosols. If aerosols can reflect and absorb incoming sunlight, what effect could events like Agung’s eruption have on Earth’s surface temperature? To find out, he plugged what was known at the time about aerosols, greenhouse gases, and how Earth absorbs and radiates energy into some physics equations. His results suggested that the aerosols should slightly cool the planet.”