Another week, another ado over nothing.
Last Saturday, Steve McIntyre wrote an email to NASA GISS pointing out that for some North American stations in the GISTEMP analysis, there was an odd jump in going from 1999 to 2000. On Monday, the people who work on the temperature analysis (not me), looked into it and found that this coincided with the switch between two sources of US temperature data. There had been a faulty assumption that these two sources matched, but that turned out not to be the case. There were in fact a number of small offsets (of both sign) between the same stations in the two different data sets. The obvious fix was to make an adjustment based on a period of overlap so that these offsets disappear.
This was duly done by Tuesday, an email thanking McIntyre was sent and the data analysis (which had been due in any case for the processing of the July numbers) was updated accordingly along with an acknowledgment to McIntyre and update of the methodology.
The net effect of the change was to reduce mean US anomalies by about 0.15 ºC for the years 2000-2006. There were some very minor knock on effects in earlier years due to the GISTEMP adjustments for rural vs. urban trends. In the global or hemispheric mean, the differences were imperceptible (since the US is only a small fraction of the global area).
There were however some very minor re-arrangements in the various rankings (see data [As it existed in Sep 2007]). Specifically, where 1998 (1.24 ºC anomaly compared to 1951-1980) had previously just beaten out 1934 (1.23 ºC) for the top US year, it now just misses: 1934 1.25ºC vs. 1998 1.23ºC. None of these differences are statistically significant. Indeed in the 2001 paper describing the GISTEMP methodology (which was prior to this particular error being introduced), it says:
The U.S. annual (January-December) mean temperature is slightly warmer in 1934 than in 1998 in the GISS analysis (Plate 6). This contrasts with the USHCN data, which has 1998 as the warmest year in the century. In both cases the difference between 1934 and 1998 mean temperatures is a few hundredths of a degree. The main reason that 1998 is relatively cooler in the GISS analysis is its larger adjustment for urban warming. In comparing temperatures of years separated by 60 or 70 years the uncertainties in various adjustments (urban warming, station history adjustments, etc.) lead to an uncertainty of at least 0.1°C. Thus it is not possible to declare a record U.S. temperature with confidence until a result is obtained that exceeds the temperature of 1934 by more than 0.1°C.
More importantly for climate purposes, the longer term US averages have not changed rank. 2002-2006 (at 0.66 ºC) is still warmer than 1930-1934 (0.63 ºC – the largest value in the early part of the century) (though both are below 1998-2002 at 0.79 ºC). (The previous version – up to 2005 – can be seen here).
In the global mean, 2005 remains the warmest (as in the NCDC analysis). CRU has 1998 as the warmest year but there are differences in methodology, particularly concerning the Arctic (extrapolated in GISTEMP, not included in CRU) which is a big part of recent global warmth. No recent IPCC statements or conclusions are affected in the slightest.
Sum total of this change? A couple of hundredths of degrees in the US rankings and no change in anything that could be considered climatically important (specifically long term trends).
However, there is clearly a latent and deeply felt wish in some sectors for the whole problem of global warming to be reduced to a statistical quirk or a mistake. This led to some truly death-defying leaping to conclusions when this issue hit the blogosphere. One of the worst examples (but there are others) was the ‘Opinionator’ at the New York Times (oh dear). He managed to confuse the global means with the continental US numbers, he made up a story about McIntyre having ‘always puzzled about some gaps’ (what?) , declared the the error had ‘played havoc’ with the numbers, and quoted another blogger saying that the ‘astounding’ numbers had been ‘silently released’. None of these statements are true. Among other incorrect stories going around are that the mistake was due to a Y2K bug or that this had something to do with photographing weather stations. Again, simply false.
But hey, maybe the Arctic will get the memo.
Nick Gotts, BlogReader and others — I am a (retired) computer scientist and have often and continue to write undocumented one-off computer programs which become more and more impenetrable as the experimental results are developed. Such programs were and are never intended to be read by anyone else.
So Gavin, bless his patient soul, has the right of it completely: go roll your own.
And as for perceptions, any rational observer will conclude that this matter is indeed a tempest in a teapot.
Steven Mosher (#392) wrote:
Actually I haven’t studied any Moore although it would be possible for me to use someone’s argument without knowing it – if I had run into it in the past. Things tend to get rather thoroughly integrated if you think about then too much. Likewise, being a good Aristotelean, I tend to think that there is very little which is new under the sun. Whatever we might do already exists as a potentiality of human action.
But the “hand” obviously goes back to Descartes in his room by the fire. And who knows? He might have been recalling Plato’s “The Republic,” one of my favorite books, specifically the scene in the cave.
It pays to read the classics. Philosophers tend to recycle things a fair amount, oftentimes as a tribute to earlier philosophers.
Well, I am sure you can see why.
As a libertarian, you would in all likelihood be strongly anti-Communist. Likewise, he prides himself on his education, having obtained a PhD. He is strongly opposed to acknowledging the scientific basis of climatology, quite passionate about it, actually, and willing to go to great lengths against that which he opposes.
Finally, he has a background in deconstructionism. In some ways, not that unusual for a libertarian nowadays, but this is one of the areas which have claimed some familiarity.
*
I wouldn’t call myself a libertarian but a classical liberal. Same thing, I suppose. And a large part of my motivation in the past was the result of my opposition to totalitarianism. I tried infiltrating a cult that had used brainwashing in the past at one point, for example.
I also have passing familiarity with Alvin Plantinga. If I remember correctly, he posed a self-referential argument against evolutionary biology a while back. I suspect he knew better. Some poor creationist tried to spring it on me at one point – without attribution. There is some good analytic philosophy out there, though. For example, I like some of the work that has been done on the relationship between coherentialism and foundationalism.
Re#400: “Critics, must come up with their own way in confirming or denying the validity of GT measurements.”
I agree. Those who are truly and scientifically skeptical of results need to carry out the work and present it to the scientific community. It seems like the ‘auditors’ are claiming that they don’t have to do this, but at the same time claiming they are only doing what scientists should (i.e., be skeptical of conclusions).
Well, guess what, we have a framework for doing that in a formal way and it’s called science. Why can’t the auditors just propose, design, and conduct science like everyone else?
In fact, here’s the link to the AGU conference in December, abstracts are due Sept. 6th.
http://www.agu.org/meetings/fm07/
Responding to someone else, Hank Roberts (#396) wrote:
Debate certainly has its place in our society, but I myself prefer dialogue.
Debate and dialogue are about as different as night and day. In debate, there exists a genuine tendency almost to the very core of it to think in terms of opposition, to adopt an us vs. them view of the world and thereby place pragmatic concerns above truth itself.
In contrast, dialogue is a cooperative endeavor in which the “us vs. them” is regarded essentially as a passing illusion. The goal of dialogue is the discovery of the truth, and in it one recognizes the fact that by cooperating with others and sharing one’s insights one stands a far better chance of understanding reality than if one were working alone.
As such I believe the proper intent and orientation of science and dialogue are the same.
Gavin’s response to #367: [Response: Again, you miss the point. It is better for someone to come up with the same answer by doing it independently – that validates both approaches. – gavin]
If they differ?
[Response: Then you investigate further. -gavin]
But, but, that means they might actually have to do some work, rather than sit around trying to poke holes in the work of others.
A rather obvious reason why auditors and skeptics are not jumping at the chance to build their own climate models is because there are not millions of dollars to fund the computers/servers, programmers, and statisticians required such a project. Correct me if I’m wrong, but someone who asks the U.S./Canadian/Any-European government to fund a model that is being conducted by a “skeptic” would not be receiving any funding.
[Response: Actually, I don’t know about that. Presumably they would start with one of the existing models, they’d have to demonstrate some proof of concept of their idea and that they were capable of doing the work involved. Red flags for the funders would be if there was some pre-determined conclusion before any experiments were done, or if the literature was extremely selectively cited in the proposal. To get funding, you need to deal with the best arguments from the mainstream, not cherry-picked talking points. It’s not too high a hurdle if people are serious. – gavin]
Re #401
It may well be that, once released as it is, the “code” proves itself quite unusable, i. e. too difficult to be decrypted by the average, and also somewhat better than average climate auditor…
(Just usable to be rerun as it is to obtain uselessly identical results…)
But saying to somebody that perhaps you have kept waiting for months:
“even if you ask for it, I will not give my uncommented code to you,
and that’s BECAUSE I’m worried that you would lose your precious time in vain attempts of mastering its complexity”
has a quite hollow ring in it.
Even a non-paranoid is thus brought to imagine that some additional untold reason MUST stay there behind.
It could be the desire, easy to understand, to fully own the result of one’s work, or also the fear to open easier grounds for totally stupid and factious arguments…
Possibly there are cases in which these elements somewhat justify a “more reserved” attitude,
but for figures as these, deemed to be weighty enough to justify fundamental changes in the lifestyle of the entire mankind, it seems to me (and I suspect to many others) these petty problems are unmeasurably irrelevant,
So the minimum acceptable to make it easier for the “auditors” and to silence skeptics, should be releasing together with the results:
input data, algorithms explanation AND runnable code.
Anyone participated in this?
“Through the program for Climate Model Diagnosis and Intercomparison, LLNL provides the
international leadership to develop and apply diagnostic tools to evaluate the performance of climate
models and to improve them. Virtually every climate modeling center in the world participates in this
unique program.” http://www.ofes.fusion.doe.gov/FusionDocuments/07SCOverview.pdf.
gavin> It is better for someone to come up with the same answer by doing it independently – that validates both approaches.
How does this compare with the methods used to find the error in the satellite temperature data?
[Response: That is the approach that was taken then. An independent analysis was done (RSS) making a whole set of different assumptions to the original UAH analysis. It ended up giving a different answer. At which point people started combing through the codes for reasons why. It was only when there was an independent emulation to compare to that the problem (with the LECT correction) was discovered. After all, people had been saying that there was something fishy about the UAH analysis for years prior to that. – gavin]
[ Again, you miss the point. It is better for someone to come up with the same answer by doing it independently – that validates both approaches. – gavin ]
I think that validates both people made the same assumptions and / or mistakes.
[ Benson: undocumented one-off computer programs which become more and more impenetrable as the experimental results are developed. Such programs were and are never intended to be read by anyone else. ]
I’m hoping that climate modeling software is more than just a one off hobby project. Implying that they can’t be understood by someone else makes me think of junior programmers that wave their hands when trying to describe why their program does something when they don’t really know.
Or to put it another way: a fudge factor here, a smoothing of data there, a rejection of rural sites there and pretty soon what you’re modeling is what you want to see.
[Response: The GISS Climate model code is available: http://www.giss.nasa.gov/tools/modelE , the issue with the GISS urban adjustment is orders of magnitude simpler. – gavin]
Hank Roberts wrote (#396) in response to Steven Mosher (#392):
Hank,
He isn’t saying that I confused him with Steven W. Mosher before, but that other people had confused him with Steven W. Mosher before.
He states:
As such, judging from his post, I have made this mistake only once.
But you are right.
Back when he was giving Ray Ladbury such a hard time and I had made that comment about people with ideological commits. It got him so riled that he turned his attention to me, so I wrote a post in which the first three sentences were (No Man is a (Urban Heat) Island, Comment #193):
The Population Research Institute headed by Steven W. Mosher is devoted to population growth, so at that point I was trying to get his attention. Then my last sentence in that post was:
Not that subtle, I must admit, but I was trying to tell him that not only did I know who he was, but that I had the evidence. Other than that, it was a pretty innocuous post. After that we didn’t hear from him for several weeks, so I kind of assumed I was right.
But then he came back and after a bit focused on me rather intently.
At the end of two long posts, I responded #240:
(Note: the sentence with the brackets was garbled due to my exhaustion at the time, so I have just cleaned it up.)
Anyway, he disappeared for a couple of days after that. But what I had to say just recently (#378) would have made him nervous if he knew how to read between the lines. And if left unchallenged, it would have gotten other people interested, and they would have started digging.
Who knows?
If they had done some digging, they might have discovered that what I was hinting at involved things that would cast Steven W. Mosher in a very bad light.
You must admit there are some similarities – just as I pointed out in #402. Personally, it wouldn’t surprise me at all if Steven W. Mosher were obsessed with RealClimate – given that Steven Mosher’s little war on the scientific status of global warming. But I might regard it as cause for some concern.
Anyway, I am glad that our Steven Mosher isn’t that Steven Mosher.
It sets my mind at ease.
A rather obvious reason why auditors and skeptics are not jumping at the chance to build their own climate models is because there are not millions of dollars to fund the computers/servers, programmers, and statisticians required such a project. Correct me if I’m wrong, but someone who asks the U.S./Canadian/Any-European government to fund a model that is being conducted by a “skeptic” would not be receiving any funding.
If the skeptics had any prospects of posing a serious challenge to the current scientific consensus, they sure as h-e-double-hockeysticks would have gotten tons of funding from Exxon or the Saudis or whomever.
Given that Exxon and Co have been very generous in terms of funding scientifically incompetent hacks, you can be sure that they would not hesitate to throw truckloads of money at *competent* scientists who could make a serious scientific case against global warming.
But in spite of the *billions* of dollars of cash they have on tap, the big oil companies have not been able to find *any* competent climate scientists to fund.
That should tell you something…
Nitpickery: Drop the comma to get the web page (the blog software underscores punctuation sometimes, breaking links)
http://www.giss.nasa.gov/tools/modelE
These docs, linked from that page, answer a lot of questions
http://www.giss.nasa.gov/tools/modelE/HOWTO.html#part0_3
But gee, Gavin, can’t you make it run on a Palm Pilot or an iPod so every child can play? …. sorry ….
I see some repetitive misconceptions in the blog arguments between skeptics and the rest of us. One is a confusion of weather, particularly local weather, with overall trends in climate. Another is a tendency to talk about science independently of the world around us. It’s clear to me that it’s warmer on average now than it was 40 years ago, and I know older people were saying the same thing to me 40 years ago. One simple example is Martha’s Vineyard’s main harbor which in the early 1900s froze over hard; I don’t know when this stopped. Also, climate change appears to include polarization of warm and cold. This year in the US and Europe (and maybe elsewhere) we’ve seen extremes on both sides. In the US those above the jet stream see cool, those on it see wild, and those below it see hot. If the hot is hotter and the cold is colder, that’s a change that is part of the same phenomenon.
About the northern European summer, I’ve heard La Nina mentioned. It made me think of other readings on the subject of how the Gulf Stream mechanism is being broken down, though I get the impression it is not being blamed for the summer about to be past. Here’s the latest I saw on this, not from the primary source, but a reasonable summary nonetheless:
http://news.bbc.co.uk/2/hi/science/nature/6946735.stm
For those of you who prefer an excerpt:
“Writing in the journal Science, they say it may now be possible to detect changes related to global warming.
The Atlantic circulation brings warm water to Europe, keeping the continent 4-6C [1C=1.8F] warmer than it would be otherwise.
As the water reaches the cold Arctic, it sinks, returning southwards deeper in the ocean.
Some computer models of climate change predict this Atlantic Meridional Overturning Circulation, of which the Gulf Stream is the best-known component, could weaken severely or even stop completely as global temperatures rise …
Last year the same UK-led team published evidence that the circulation may have weakened by about 30% over half a century.”
One common thread I see in the reams of calculations meant to prove that data adjustments call theory and observation into question is a tendency to overlook the connection of all parts of the planet with all other parts, and the way their patterns play with each other. As an artist, I love the whorls and spirals, and can’t help looking at them as all connected.
I’m always surprised that people can ignore all those telling pictures of changed glaciers everywhere, and the wild weather and related deaths, problems with water and wars for scarce resources, drought, wildfires, and say something nasty and in many cases untrue about Al Gore as if that were some kind of answer.
RE: # 383: “going fast now” sorry not to provide a link:
look at the lasee 4 days of this animation of the Jetstream:
Re 382 (Hank)
Everything in 381 can be found in IPCC
However, I confirmed some of the information in the IPCC by going to the produce market and talking to fruit growers. I looked in my journal for when my fruit trees bloomed. I checked to see how ripe my apples are this week. I talked to USDA. I talked to ag guys with mud on their boots. All good data points , (if put that in the context of Sampling Techniques by Cochran. I still use the 3d edition.)
The truth is not in peer reviewed journals, the truth is in nature. You are not likely to find a peer reviewed journal telling you that you can melt a hole for ice fishing with a dog turd, or that you can melt a channel across an ice covered pond with a handful of coal dust so you can get to town in your wooden canoe. That is what every boy should learn by doing. Peer reviewed journals never talk about the really important things in life. And, global warming is about the really important things in life.
Three years ago when I was saying that in my experience with snow and ice, the Greenland Ice Cap was likely to disintegrate faster than the models and IPCC were suggesting, and everyone thought I was a kook. Now, it seems that I may have been right. See (http://www.dailyindia.com/show/166067.php/Greenland-ice-cap-meltdown-to-cause-22ft-floods). I look at (http://stratus.ssec.wisc.edu/products/rtpolarwinds/) and consider how those winds are blowing across open and warm water, and I think 300 years might be a little optimistic. Ice turns to rubble in the rain. (Ask an ice climber – someone that looks at 20,000 feet of ice every year as if his life depended on it. Ja, I’ve done a little ice climbing.) And, as long as the ice is there, it is going to support a thermal low that will pull air (full of latent heat) off the North Atlantic Drift. Sometimes peer review confuses conventional wisdom and being correct. An unstated assumption in the conventional wisdom about the melt of the Greenland ice was that the Arctic sea ice would remain intact. It is not.
Regarding the issue of fresh water under the Arctic Ice (as constrained by the Arctic gyres), a quick Google brings up articles such as http://ams.confex.com/ams/pdfpapers/84555.pdf. Then, thinking through the physics from different starting conditions brings the conclusion that Arctic Ice is different from Antarctic Ice, because the Arctic Ice is floating in fresh water. (Fresh water delivered by all of the rivers that drain into the Arctic.) If you Google, you can find other people have come to the same conclusion. I have not checked for peer review, but the physics works and I am confident that it is correct. I trust my physics more than I trust peer review. I trust regardless of the fact that the role fresh water under Arctic ice is ignored in the NSIDC pages on Arctic sea ice. Regardless of the fact that I do not see it in the community climate models. This bit of physics is enough to explain why the ice is melting faster than the models predicted. On the other hand, maybe it is in the models and I just do not see it. Or, maybe by 2008, we will have a climate model that says the Arctic ice will be gone by 2010. What else do you suppose it will say?
Today, I am looking at (http://pm-esip.msfc.nasa.gov/amsu/index.phtml?2) and (http://www.osdpd.noaa.gov/PSB/EPS/SST/sst_anal_fields.html) and asking what they can tell me about how Antarctic deep waters are interacting with deeply submerged Antarctic ice. Interesting physics.
Did Darwin cite a bunch of journals or did he; go, look, measure, and think?
I believe that every instrument goes out of calibration way too often. I believe and that every piece of software has bugs in it. in the fall, I carry a fruit knife, because I know our apples have worms in them, but I eat them anyway. I believe that the deer are out in the orchard eating my apples right now!(Sound of hooves on the walk.)
Further to #407, I’ve wondered about this: I suspect that if they were sufficiently motivated a company like ExxonMobil could get a climate model up and running pretty quickly. They have large scale computing, they have experts in computer modelling, they can download modelE to get themselves started,you’d probably want to employ someone with experience in climate modelling and a couple of similar ‘postdoc’ level people… but what would they do with it?
I’m unconvinced of the value of self-appointed non-expert auditing; as a practising scientist I find multiple evidence lines (i.e. satellite or borehole reconstructions) far more convincing than a heavily audited single reconstruction. I’d also be interested in complete re-analyses based on the same data.
$405 &
I guess science and accounting DO have something in common (re #225 & 245). As a bookkeeper I would add things up, but I’d add them up again to make sure my adding was correct. If I got a different sum, I’d add them up a third time. Sometimes I got 3 different sums. :)
Re #401 [I am a (retired) computer scientist and have often and continue to write undocumented one-off computer programs which become more and more impenetrable as the experimental results are developed. Such programs were and are never intended to be read by anyone else.]
Then in my view, they are not best practice in science, which is by its nature a collective endeavour, and should strive for as much transparency in method as possible. I know what you describe is very often done, I’ve done it myself, and I don’t want to suggest it seriously undermines the fine work done by Gavin and colleagues – but maybe, if they had felt they should make sure all the code they use is suitable for the public domain, this error would have been noticed earlier, either by them or by someone else.
RE412: Hi Tim.
Maybe I’m W after all. If you wiki here is a hand you
will see why I thought you were cribbing a GE Moore argument.
on snarky days my version goes “here is a finger!”
I come and go here so sometimes won’t see all your
threads.
Sorry if you think I’m singling you, you just happen to be entertaining and engaging, more so than most.
Cheers
It’s cost effective for the fossil fuel brigade to fund sceptic organisations. Very cost effective.
Spending money on simulations would just prove the AGW viewpoint and if they didn’t I dare say the climatologists would love to get their own back.
http://www.exxonmobil.com/Corporate/Files/Corporate/gcr_contributions_public06.pdf
RE #418 & Exxon running a climate model. I, for one, wouldn’t trust their code, esp if the results disproved GW. With millions of lines of code, I imagine they would be able to slip a few sneaky things in.
There’s been a lot of talk about “heroes” — sportsmen, people caught in life-threatening situations, and such — but the climate scientists are my heroes. When this site opened, I was actually a bit surprised that not only were there scientists untempted by oil money who had not gone over to the dark Exxon side, but were unafraid of attacks by the unknowledgeable, wrongly motivated, and (perhaps in some cases) Exxon-funded.
I regret having read some comments on another blog (linked in some post above) against Hansen. I was saddened (but not surprised) by the meanness and viciousness. The same people who would participate in destroying other’s property and person through global warming, and adamently deny they had anything to do with it (against all evidence that they had), would not be above such verbal wickedness.
James Hansen and the other climate scientists, and those valiantly struggling to reduce GHGs and to encourage others to do so, are my true heroes in this day and age.
Okay, the scientists made an insigificant error, which changes nothing when corrected (and they did correct it). I imagine if they later found 1887 to be the warmest year in the U.S., or even the world, it probably wouldn’t disprove AGW, since AGW is about trends, not single numbers.
When the new warmest year comes out, say, in 2008, what will the scientists’ response be? “A single year’s stats do not prove anything, even if they do fit the trend [which has already indicated global warming].” As much as I would like to see more people convinced about global warming, I respect this scientific approach.
[[By trying to hide the mistake under the carpet our dear Real Climate scientists helped the other side’s extremists to misunderstand the news.]]
If you think your opponents can’t honestly believe their position, but must secretly agree with you and be lying for evil reasons, then you don’t understand the issue at hand.
Nobody tried to “hide the mistake under the carpet.” That’s an accusation of lying, and is itself a lie.
[[The truth is not in peer reviewed journals, the truth is in nature. You are not likely to find a peer reviewed journal telling you that you can melt a hole for ice fishing with a dog turd]]
Well, that’s the problem with science, all right. It just hasn’t devoted enough attention and analysis to dog turds.
Timothy.
Care to comment on the sensor supplier change I noted in #387?
It’s a peer reviewed paper. Sensor supplier changed.
TMAX goes up, Sensors changed back in the 1990.
USAF also had a problems with this supplier and
the particular sensor in question.
Will this undermined AGW. No. Will this undermine the credibility of “authorities” and diminish thereby the force of appeals to authority?
Timothy. Renounce the bad sensors.
#413: Given that Exxon and Co have been very generous in terms of funding scientifically incompetent hacks, you can be sure that they would not hesitate to throw truckloads of money at *competent* scientists who could make a serious scientific case against global warming.
But in spite of the *billions* of dollars of cash they have on tap, the big oil companies have not been able to find *any* competent climate scientists to fund.
Flimsy argument, as it cuts both ways too easily:
If AGW were true, then you can bet every government in the world would be falling all over themselves to ensure it didn’t blossom into a larger problem.
But in spite of the *trillions* that world governments have on tap, they have opted to do nothing about it. Hence, it must not be true.
Pretty weak, eh?
re: 427
Everything isn’t symmetrical.
Re #426 [But in spite of the *billions* of dollars of cash they have on tap, the big oil companies have not been able to find *any* competent climate scientists to fund.
Flimsy argument, as it cuts both ways too easily:
If AGW were true, then you can bet every government in the world would be falling all over themselves to ensure it didn’t blossom into a larger problem.]
It’s your analogy that’s flimsy. The oil companies (primarily Exxon) that are funding and supporting denial that AGW is a serious problem, must be assumed to want to convince governments and publics of this position. If there were competent climate scientists who had (for example) knowledge of some important new negative feedback mechanism they believed might seriously alter future climate outcomes, and wanted to incorporate in a new model, funding them would be a highly cost-effective approach to doing so. I can see no reasons for them not to do so if any such climate scientists existed.
Governments, on the other hand, are subject to enormous pressures against taking effective action to curb AGW. First, direct pressure from the oil companies and other special interest groups such as power companies, auto manufacturers, building firms and airlines. Second, because many of the measures they would need to take would be highly unpopular, e.g. raising gasoline prices and taxes on air travel. Third, because measures taken by any one government would be insufficient to limit the problem, and many such measures would, if taken by one government, put its economy at a disadvantage relative to those of other states: hence all the attempts at international agreements.
Brad DeLong called Tobin Harshaw who was the ‘Opinionator’ at the New York Times: http://delong.typepad.com/sdj/2007/08/tobin-harshaw-o.html
Some years back I recall a climatologist posting, here I think, that he worked on very sophisticated climate models — with significant research that is proprietary, that couldn’t be talked about — for the petroleum industry, and commenting that all the large oil companies have used them for years. They understand what happened, that’s where the oil came from.
They’re modeling the climate in which the sediment _accumulated_ to be able to predict where to drill _now_ for oil, accounting for continental drift.
#429: The funding given by environmental groups to groups showing AGW exists far exceeds that given to other groups by oil companies. In any case, this argument is irrelevant (ad hominem); if exxon funded scientists can show that AGW doesnt exist, one should be able to find where their model(s) differed in assumptions and argue against those assumptions decisively.
My concern, in general, is that journal articles favorable towards the idea of AGW have been published for quite a long time. This is because, whether you admit it or not, a climatology journal is more likely to publish a paper showing the grand importance of climatology over a paper which concludes that the climate is more or less unpredictable and/or uncontrollable [I am not suggesting that this is the sole basis for publication, merely that is is an ever-present bias].
Then, modelers funded by environmental groups or governments under political pressure to “do something” about AGW are making hundreds of microassumptions about the data favoring AGW (with out without citing the aforementioned published papers). When the data comes from various datasets with multiple adjustments made to most rows, the possibility for conscious or unconscious data manipulation increases greatly; anybody who has ever tried to beat a rival model or make a political point (usually that you, the modeler, or your dept. needs more funding or is justified in existing) has felt this constant temptation. Other assumptions that greatly worry me are that AGW modelers claim to be able to predict in, say, decade-sized bins how much the avg. global temperature will increase, but will admit if pressed that they cannot say much about a given year, and can say virtually nothing about any given region of a planet in a given year. If that is the case, how are assumptions about the damages and benefits of global warming worldwide being made… are they being conjured out of thin air? are they completely unreliable MLEs? Policy is being made on this very point: that global warming is bad, and that the bad outweighs the good.
Finally, I am curious how these models are cross-validated (or if they are validated at all) and what type of model is typically used to fit the data.
This is my first foray into sharing my thoughts on AGW, hopefully they aren’t old and tired points/question.
#432
Carl, could you please clarify what you mean when you say the following?
“Finally, I am curious how these models are cross-validated (or if they are validated at all) and what type of model is typically used to fit the data.”
Re #417 [Did Darwin cite a bunch of journals or did he; go, look, measure, and think?]
The dichotomy is a false one. Although there were a few scientific journals in Darwin’s time, journals did not have the central place they do today in a scientific community that is very much larger. However, Darwin was highly active in an international network of scientific correspondence, without which his work, and in particular the marshalling of evidence for “The Origin of Species”, would not have been possible. See http://www.darwinproject.ac.uk/.
Re#431, Hank says: “They’re modeling the climate in which the sediment _accumulated_ to be able to predict where to drill _now_ for oil, accounting for continental drift.”
I’m not sure exactly what you’re referring to, so correct me if I’m wrong. This isn’t really climate modeling in the sense this thread has been discussing. This is more about modeling the how, why, when, and where of sediment transfer and accumulation. Climate indeed has significant effects on sediment flux (e.g., more precip –> more erosion –> more runoff –> more sediment pumped into a basin, and so on). And, yes, where there are large accumulations of sediment, you find hyrdrocarbons.
Sedimentary researchers are borrowing concepts from GCMs for modeling sedimentary systems. For example, see the CSDMS project. Atmospheric and climate scientists are probably the leaders in quantitatively understanding, and therefore modeling, the complex interactions of various forcing, feedbacks, teleconnections, and the like.
The fact that oil companies model complex systems should come as no surprise…and to say they don’t have ‘competent’ people is also ludicrous. They realize the strength of the scientific case for AGW, they’re not idiots…the focus is on misinforming and confusing the public re policy. The independent ‘auditors’ are just one piece of that puzzle. I would like to see the auditors produce some real science and have it be ‘audited’ by climate scientists….I have a funny feeling that’s not gonna happen.
I agree wholeheartedly with Lynn Vincentnathan(#423). Dr. James Hansen is my hero also. I trust him.
My trust is not shared by the likes of a Jack Kelly, who writes editorial opinions in the Toledo Blade(today, for instance, 8/18/07). He rips Dr. Hansen, Al Gore, and who knows who else, who report on the severe global warming we’ve already experienced, and will experience, if we don’t cut our burning of fossil fuels drastically. His trash in today’s Blade is about the fifth such one he’s written in the last year or so. How The Blade allows him to write such dribble is a mystery to me, but it does seriously cause me to question continuing my subscription to that paper. The parts I need can be read at the local library.
It’s amazing the denial of AGW in this area. The local rag is partly to blame. The convenient denialists grab onto an article like Kelly wrote, and turn off any furthur reading, or research.
Facts like the Arctic ice hitting it’s lowest point in recorded history this summer, and the projection that it’s likely to be ice free in the summer of 2030 get lost in the orgy of celebrating over NASA’s very minor mistake. Also thrown out is the fact that the global temperature from 6/06 to 6/07 was the warmest ever.
432, the flat-earth theory gets even less funding than the anti-global-warming theory. Does that mean we should be skeptical about the idea that the earth is not flat?
Exxon and the coal industry have billions or trillions to lose, if there were even a 5% chance that funding scientific work to disprove global warming would pay off then they could sink in billions. The fact that that do nothing more than relatively little PR funding speaks volumes.
Re #427: [If AGW were true, then you can bet every government in the world would be falling all over themselves to ensure it didn’t blossom into a larger problem.]
Not to get off into politics, but I really think you need to take a more cynical^H^H^H objective view of government. Consider for instance the cost of levee and bridge maintenance, and why governments don’t fall all over themselves to ensure those don’t blossom into larger problems…
#437: already said that this was an irrelevant point. perhaps oil companies know that nobody will accept the results of a team that they fund precisely because any result will be discarded as biased from the outset. But, again, it does not matter who funds what, only what the assumptions and processes involved in the modeling are.
#433: For example: You can, using a neural network and enough variables, fit 100% of the data in any dataset given enough complexity in the NN architecture (you can achieve something close to this in other models as well). However, such a model would almost certainly predict the future with very little accuracy. This is because the variability in observed data (i.e. station temperature records) leads to artificial associations between the input variables and what level the observed data “happened to be at” in your data set. Until your completed model is used to predict already known values from another data set that weren’t used in building the model, or (better yet) until you wait to see how the predictions compare to the actual future values, the model cannot be trusted. I have not even directly asked a climate modeler, but I have seen statements online that imply that this process doesnt occur… obviously one shouldn’t judge based on heresay, so I am asking a more knowledgeable group about the modeling process.
Btw: “the modeling process” may differ widely from model to model, in which case I’d appreciate hearing some things about the variety of what’s out there.
to simplify/add to my last post: what temperature predictions more than 5 years out have been at all accurate? Could a model now predict the mean global temperature for 2010-2015 accurately? If it cannot, it is completely worthless.
Furthermore, this prediction should come from a model TODAY. You cannot continuously feed new data into a model without destroying your ability to evaluate it’s accuracy. Assume the model is bad. As the predictions begin going astray, new data will re-adjust the parameters so that the predictions are accurate once again. Acceptable alternatives are using a validation data set (not used in building the model) or by simply using the predictions from an older model and comparing them to current temperatures.
Is there data available to compare predicted temperatures for 2005-2007 from a late 90s (or any other arbitrary time periods) model to actual temperatures ? This is actually rather important. If the model fails to work, this would be a very troubling sign. If it is consistently accurate, one removes the question of predictive power and leaves only the question of whether or not the data itself represents what we are trying to measure.
[Response: See Hansen’s 1998 projections. -gavin]
Re #432 “Policy is being made on this very point: that global warming is bad, and that the bad outweighs the good.” Really? News of this policy development seems to have escaped me.
Ahh one last thing. Gavin, The two charts you show in the article have different Axes.. one shows anomalies between -1.5 and +1.5 and the other between +-.6.
This is not your “beck” moment, but it might be interesting to avoid the “chartsmanship” issue.
Just a thought.
[Response: This makes no sense. They are different quantities, with different levels of noise. Why should they be on the same scale? Global means have much less variability than regional ones, or local ones. Each scale is chosen to show the whole record in as much detail as is practical. (Plus they are taken directly from the GISS website and so weren’t drawn by me in any case). – gavin]
#439
I am familiar with neural network and the phenomnon you describe is not something that is particular to them. You can laso fit a simple polynomial to match any finite set of data points, given a high enough degree of the polynomnial(s). Thi is indeed a problem when it comes to things like statistical models use in e.g. regression.
Climate model are _not_ statistical models.
A climate model is a first principles model, like for example the models used to study air flow around an aeroplane wing before it is built. These models are based on the fundamental laws of physics not statisical fits. One starts out with the laws of mechanics describing fliund flow, heat transfer and so on, then set up the basic geomtry of the system. For a climte model this would the earth, it topography and the presceence of water and other materials at an initial time. Then one lets the model run and reads of data as time pases in the model. The accuracy of a model is judged for example by starting it up with the known weather and climate in 1965 and then comparing the models prediction with the observerd climat in the following years.
If the model does not do well one does not just change some parameters to make it fit the data, as one would have do nein a statistical model. Instead one looks at what part of the actual physics one has left out and tries to add that without making the model so complex that our computers can’t run it.
The climate models used today do give correct predictions for the 1970’s if you start them out at some time in the 1960’s.
#441: I think that goes without saying; we would want to increase global warming if the positives (defined as some socioeconomic metric) outweighed the negative, or at least would not care so much to try and stop it. Or, alternatively, if AGW is shown to not exist in addition to showing that NGW (N = natural — not sure of the jargon used in this field yet) does exist:
1) The phenomenon could be ignored (if found to be a net positive)
2) Resources could be spent to move people out of coastal areas, to begin buying farmland further north in canada, etc (if found to be a net negative).
It is most likely that global warming, regardless of the cause, will cause both positive and negative effects in most areas with some distribution of net positive and net negative effects across the globe. I find it ridiculous that many people assume that GW is a net negative across the globe.
#432 #439
Regarding the decade predictions with a lack of predictions for specific places and points in time.
Part of the problem here is that “climate” is essentially a description of the average temperatures, humidities, amounts of rain and so on in an area during a year. Large averages like this are much easier to predict than the effects at a specific place.
A simple everyday example of this is the bubbles in a pot of boiling water. From physics we know that when water is heated enough it will boil. So if we turn up the heat on the stove we can predict that the water in the pot will begin to boil. Predicting at which precise point at the bottom of the pot the first bubble will form when the water begins to boil is much much harder.
Tea or coffee, Steven?
Re #444 [It is most likely that global warming, regardless of the cause, will cause both positive and negative effects in most areas with some distribution of net positive and net negative effects across the globe. I find it ridiculous that many people assume that GW is a net negative across the globe.]
Do you have specific evidence of anyone “assuming” that, rather than arguing it? There are very good reasons to expect it, the most general of which is that social, technical and ecological systems are adapted to current climatic conditions, and rapid change in any direction is likely to be highly disruptive. For more specific reasons, see for example the IPCC AR4 Working Group III Summary for Policymakers (http://www.ipcc.ch/SPM040507.pdf) or the Stern report (http://www.hm-treasury.gov.uk/independent_reviews/stern_review_economics_climate_change/stern_review_report.cfm).
BlogReader — The one-off projects are not ‘hobbies’. The results are the basis for scientific papers.
Nick Gotts — Actually, this technique IS the best scientific practice. Rather than taking the (large amount of) time to polish and document the program, it is more productive to write and polish the paper.
I have never heard of a case in which a referee wanted to see the computer code to complete his task of checking that the submitted paper was, at least on the surface, correct.
[[The funding given by environmental groups to groups showing AGW exists far exceeds that given to other groups by oil companies.]]
Oh, really? What are the respective numbers, and what is your source?
#443: Thanks, I did not know this (re: first principles model vs. statistical model). So, then, the modeling process is to find which of the known physical laws are required to be inputs in the model given a desired level of accuracy. Should infinite computing power be available, the model would be perfect (assuming perfect input data).
All laws used in current modeling are or have been assumed to be accurate. Climate skeptics, then, can challenge the truth of the physical laws, human errors in the modeling, or the input data only.
Do the current models predict the 1980s, 1990s, 2000s (so far) well? I would be concerned only if error continued to rise decade by decade (indicating a rising trend that does not exist).
#445: Complete agreement here. I am very annoyed when MLEs of where the first bubble will appear are touted as accurate because the overall “averaged” model is accurate, though (ex. hurricane forecasts).
#447: I would argue that canadian tundra // Siberia will undoubtedly become more valuable if local temperatures rise. This is a silly example, but there are many areas where the net effect of warming would be positive even when considering the fixed cost of relocating persons/industries. Even then, we should rather consider [Global warming costs – global warming benefits] and compare it to the costs of reducing greenhouse gases; if it is cheaper to do nothing, that is what should be done. It’s better to spend $1T to move everyone off the shoreline, out of floodplains, and out of deserts than it is to spend $2T so people can stay put (this assumes that money can represent sociopolitical concerns, and also that spending $2T would actually do anything for the climate). The world isn’t going to stay in a constant state on any political/economic/social front, regardless of global warming’s effect. I will look at the IPCC’s assumptions sometime in the next few days and comment back here if the thread is still alive.