Tim Lambert (deltoidblog AT gmail.com) is a computer scientist at the University of New South Wales.
Bob Carter has managed to get a whole bunch of people to sign a letter touting his warming ended in in 1998 claim. Here's what they signed:
there has been no net global warming since 1998. That the current temperature plateau follows a late 20th-century period of warming is consistent with the continuation today of natural multi-decadal or millennial climate cycling.
I bet that they didn't include a graph of global temperatures:
You know, even the CEI admits that this "warming ended in 1998" claim is disingenuous, but look at all the people who signed Carter's disingenuous letter.
And look who Nexus 6 spotted in the list: Edward Wegman. Yes, the man touted as an independent judge of McIntyre and McKitrick arguments against the hockey stick turns out to be a global warming denier.
Environment
Politics
Comments
He got Louis Hissink and Zbigniew Jaworowski, too. Drs. Larry, Curly and Moe Howard were unavailable for comment.
Posted by: jre | December 13, 2007 2:29 PM
Actually thats Dr. Howard, Dr. Fine, and Dr. Howard.
Why this is really interesting is that you often hear those names paged in the background on TV shows and movies.
Posted by: Rob | December 13, 2007 2:34 PM
Thanks, Rob! My wretchedly incomplete education in the fine arts just got a boost!
Posted by: jre | December 13, 2007 3:01 PM
Here is the truth about Global Warming.
http://video.google.com/videoplay?docid=-3309910462407994295
http://video.google.com/videoplay?docid=7299668110171032533
http://www.youtube.com/profile_videos?user=GlobalWarmingFraud&p=r
Posted by: Mike | December 13, 2007 3:19 PM
Thanks, Mike! Now I know the IPCC blames babies for global warming. Evil, evil babies--that's the consensus.
Posted by: Boris | December 13, 2007 4:09 PM
Hmm... no colon. I guess by "here" you are referring either to Deltoid in particular or Scienceblogs in general. Nice endorsement!
Posted by: pough | December 13, 2007 4:32 PM
"WEGMAN: Carbon dioxide is heavier than air. Where it sits in the atmospheric profile, I don't know. I'm not an atmospheric scientist..."
Perhaps the blue on the graph above is supposed to represent CO2 settling into the valleys.
Posted by: JB | December 13, 2007 6:28 PM
Somebody needs to ask Wegman if H2O is heavier than water and if so, where it sits in the water column. I don't know 'cause I'm not a water chemist.
Posted by: Thom | December 13, 2007 6:48 PM
Disappointing to see Don Aitkin as a signatory to something like this. I know Lindzen has given passing support to "global warming stopped in 1988" before, but this letter, with its absurd reference to a "plateau" is a much stronger version of the claim.
Posted by: John Quiggin | December 13, 2007 7:14 PM
Interesting, on the other hand that McIntyre didn't sign. Given that it's published in Canada, you would imagine he was asked.
Posted by: John Quiggin | December 13, 2007 7:17 PM
I was also disappointed to see Reid Bryson on there. Otherwise the list seems to be largely the usual useful idiots.
Posted by: markg | December 13, 2007 7:24 PM
Ha! If you go to Mike's YouTube link for the videos on "Global Warming Fraud," you get a screen full of video thumbnails, all of them purportedly taking on the notion of human-caused climate change. The only video on the page with a multiple-star rating is one by comedian George Carlin, who delivers a rant about the hypocritical self-absorbed yuppies who want to "save the planet," mocking the impact of tiny humanity on an earth billions of years old. The crowd eggs him on with cheers and laughter. Then, partway through his rant, after saying several times that we can't harm the earth, he sticks in the knife by noting that, "The earth isn't going anywhere, although we are."
There's a pause in the laughter, which probably indicates that the people who were most enjoying Carlin's diatribe just figured out he is talking about the planet's survival, not the survival of its ecosystem. Sure, humanity will go extinct, the environment will go into upheaval, and some completely new equilibrium point will eventually be reached. Then I think the people who were probably grim-faced and tight-lipped during the first part of Carlin's piece take over the laughter and the anti-eco crowd who assumed Carlin was a climate-change denialist must grimace as they realize he isn't actually: He's just saying it's no big deal if we foul our nest and die. The nest will raise up some other form of life. It's robust. We're puny.
Posted by: Zeno | December 13, 2007 8:32 PM
Oh boy, on the same day that these skeptics signed onto Bob Carter's editorial, NOAA released this:
NOAA: 2007 a Top Ten Warm Year for U.S. and Globe
Here's a gem from the press release: "This currently establishes 2007 as the eighth warmest on record. Only February and April were cooler-than-average, while March and August were second warmest in the 113-year record."
I think Bob Carter et al. are truly deserving of the "Rake of the Year Award."
Posted by: Thom | December 13, 2007 9:23 PM
Once again, that trendline in the graphic does not support your point. It is not a "since 1998" trendline.
Posted by: nanny_govt_sucks | December 13, 2007 9:39 PM
There is no "since 1998" trend that has meaning in a climatological context. A few years of variability means nothing, one way or the other. Even the most recent year of arctic melting does not constitute a "new trend".
I thought you 'Anti Global Warming" folks were all top-notch statisticians?
Posted by: markg | December 13, 2007 9:47 PM
That's beside the point. Tim is trying to discredit the "since 1998" claim. In order to do so, he should use data and trendlines "since 1998". The graphic he's presented doesn't do that.
Posted by: nanny_govt_sucks | December 13, 2007 10:47 PM
nannygovtsucks said: "Tim is trying to discredit the "since 1998" claim. In order to do so, he should use data and trendlines "since 1998". The graphic he's presented doesn't do that."
Tamino put to rest the nonsense about global warming stopping in 1998
I notice nanny also posted in the comments there, so he is certainly aware that the trend shows warming since 1998 (even if one includes 1998, the peak year of El Nino)
Nanny seems to go from blog to blog recycling his "skepticism".
It gets old after a while.
Posted by: JB | December 13, 2007 11:10 PM
Hmm,
I thought you all were scientists?
No one denies that we are in a warming trend. The only doubt I have whether it is anthropogenic or not.
There are several competing hypotheses as to what is causing our current warming trend. To strip out the BS, here are several of the current debates:
1) There is an UNPROVEN hypothesis that human activity has increased the atmospheric concentrations of some trace chemicals. 2) There is an UNPROVEN hypothesis that changes in solar activity has increased (and decreased) the atmospheric concentrations of some trace chemicals. 3) There is an UNPROVEN hypothesis that changes in the global average temperatures have increased (and decreased) atmospheric concentrations of some trace chemicals. 4) There is an UNPROVEN hypothesis that changes in the concentrations of some trace chemicals have increased (and decreased) global average temperatures.
Most of the comments I see here are mistaking correlation with causation, just as the post regarding the NOAA record is.
Science is not a popularity contest: The number of authorities who subscribe to a given hypothesis has no relationship to its empirical accuracy or validity. Theories require proven, predictable results. They do not require a democratic vote of the authorities. If the opinion of a majority of the authorities was all that was required, we would still think the earth was flat. Empirical proof is required.
If there is an increase in the current warming trends due to manmade increases in CO2 levels, then we might want to do something about it. But what if it is caused by sunspot activity, as some very reputable researchers believe? Do we want to take drastic measures which will kill a lot of people as are currently in vogue?
Think back (if you were alive then) to the Global Cooling predictions of 1975. The proposed solutions were also drastic: things like spreading coal dust across the Arctic and Antarctic. That turned out to be a mistaken prediction. But what if we had actually done it? Where would we be now? Do we have the knowledge--when we cannot even model the effects of clouds in our computer simulations which are the foundation of current IPCC theory--to accurately manage the global climate?
I personally don't think so.
So you can call me a apostate. I believe in climate change; but I do not believe--or disbelieve--in the currently ascribed causes. I will wait for the empirical evidence before making a decision. And based on that, I think we can come to a general consensus on effective means of dealing with it.
Posted by: steveB | December 13, 2007 11:13 PM
Nags, I know you visit 'Open Mind'. I know you saw Tamino's presentation of the trends over the past years. I know you didn't like that the linear trend since 1998 is still relentlessly positive (r = .24 by the raw data used to make up that graph, a slope of +.008 degrees C per year). Not bad, considering the data was cherrypicked to give the opposite impression. Take it out a few years earlier and the slope is much higher (.02 degrees/year), as expected since 1998 was an El Nino year. But even though I know all this, I just don't know why you insist on foisting your own illusions on those of us in the reality-based community. As missionaries for capitalism go, you're making a lot of converts: for anything but. However, we can tell Bob C that the facts go against his nonsense, carefully handpicked though it was. I guess it's just not vintage nonsense. Perhaps if he used a 2-year trend? It's -.062 from 2005 to 2006. On the other hand, it's +.024 from 1999 to now, consistent with the trend over the last 25 years. He loses even when he picks the cards.
Posted by: stewart | December 13, 2007 11:25 PM
SteveB 'If there is an increase in the current warming trends due to manmade increases in CO2 levels, then we might want to do something about it. But what if it is caused by sunspot activity, as some very reputable researchers believe?"
Oh, brother, the sunspots in our eyes are killing us.
Might I suggest you educate yourself a little on the basics before you assume everyone else is "mistaking correlation with causation"? [sic] (should read "equating correlation with causation" or perhaps "mistaking correlation for causation" )
Contrary to your own mistaken belief, the question of the source of the recent warming has been quite extensively considered by climate scientists and they have come to the conclusion that most of it is anthropogenic in origin.
Perhaps you might start with the latest IPCC report.
Posted by: JB | December 14, 2007 12:52 AM
That would be a stupid thing for Tim to be trying to do, and it is certainly not how I interpret his post.
The issue is not whether or not the trend since 1998 is up or down. The issue is that it is statistically invalid to cherry-pick a particular year when the temperature is particularly high and insist upon calculating the "trend" from that particular year. This is an elementary and well-known statistical error. Basically, when you have a noisy signal, you can create the appearance of "trend" in whatever direction you choose if you are allowed to choose as your starting point a time when the value is particularly high or particularly low (why start at 1998? why not 2000?). The statistical rule is that you must either use the entire data set, or make the decision as to what intervals to examine trends over before you look at the data.
Posted by: trrll | December 14, 2007 1:06 AM
Steve: you say you do not believe in the currently ascribed causes for climate change, presumably referring to increases in atmospheric CO2, methane and other greenhouse gasses. I am curious as to what your views are so perhaps you could say where you disagree with the following.
It was established about 150 years ago that CO2 absorbs infrared radiation.
About 100 years ago the effect of a doubling of atmospheric CO2 on Earth's temperature was calculated. Since then, the estimate has been refined but it turns out that the original estimate was not far wrong.
There has been a documented rise in the level of CO2 in the atmosphere over the past 50 years.
Evidence from several sources indicates this is a result of human activity, in particular the burning of fossil fuels.
Given the basic physics, a rise in the level of CO2 in the atmosphere is expected to result in an increase in global temperatures, provided there is no massive feedback mechanism or other process acting in the opposite direction.
No negative process that is remotely close to the magnitude required has been identified.
Given this, a rise in the Earth's temperature is to be expected and in fact the observed rise accords quite well with predictions.
As you seem to disagree with the conclusion, I would like to know exactly which step you feel is incorrect, and the evidence you have to support your views.
You refer to mysterious trace chemicals that you claim have increased and decreased as a result of changes in solar activity. What are these chemicals? Do you have evidence for changes in solar activity? The graphs I have seen show virtually no change.
You claim you require empirical evidence. What experiments do you think need to be done and which data do you think need to be collected? What specific results, if they were to be found, would convince you that anthropogenic greenhouse gasses are resulting in global warming?
BTW Scientific theories are never proven to be correct, at best they merely withstand every attempt to show that they are wrong.
Posted by: Richard Simons | December 14, 2007 1:07 AM
First sentence in the original post.
This is just the first of the provably false statements in Steve's post.
That was easy!
Posted by: dhogaza | December 14, 2007 2:24 AM
Hi all
Tim, which graph do you wish to use, the one on this post or the one from the satillite data on your post:
http://scienceblogs.com/deltoid/2007/11/apictureisworthathousand1.php#more
The latter shows a levelling off in temperture since 2000. Lets just say 1998 was an anomily. I know in New Zealand 2007 was not a special year temperture wise, in Auckland anyway. The graph even suggests that 2007 is quite cool compared to the last few years.
So I am a bit confused on which data I should use. Please elighten me.
Regards from a non warming New Zealand Peter Bickle
Posted by: Peter Bickle | December 14, 2007 3:32 AM
Use whichever dataset you wish. The linear trend starting at 1998 through 2006 is positive for UAH, RSS, Hadley/CRU, and NASA. NASA has the most, RSS has the least. Of these, NASA is the only one to cover the poles, which is the primary reason that its numbers are higher (since that is where the highest anomalies are).
Posted by: cce | December 14, 2007 4:21 AM
Hi all
CCE, from the post cited by Tim I reckon the trend is linear since the turn of the century.
Regards Peter Bickle
Posted by: Peter Bickle | December 14, 2007 4:30 AM
Hey, look dude, only we Americans are allowed to substitute our country for the world when speaking of "global" trends.
Posted by: dhogaza | December 14, 2007 4:52 AM
"Hi all
CCE, from the post cited by Tim I reckon the trend is linear since the turn of the century."
Distinguishing between a linear and a non-linear trend over seven years is quite a feat, especially done by eyeball as appears to be the case here.
Posted by: John Quiggin | December 14, 2007 4:54 AM
Hi all
You can say that John, I made a judgement call. It will not be exact to zero but it will only be a small slope either + or -. All I am saying is that using satillite data there has been no significant warming since the turn of the century based on this.
Regards Peter Bickle
Posted by: Peter Bickle | December 14, 2007 5:11 AM
Of course, by the same argument I might suggest that, based on the the 2007 winter, winter arctic ice mass is asymptotically heading to 0.
Hmm, I wonder if I can write that up and get it into GRL...
Posted by: markg | December 14, 2007 5:46 AM
Peter B - I didn't know eyeballing a graph was an effective way to determine statistical significance. (And didn't you mean to say 'from 1998', not 'from the turn of the century'?)
Tamino looked at this canard a few months ago by plugging the data into a spreadsheet:
http://tamino.wordpress.com/2007/08/31/garbage-is-forever/
His (or her?) conclusion: whether you use the GISTEMP or HadCRU data, the slope from 1998 to the present is positive, and the trend is statistically significant.
Posted by: Tony Lee | December 14, 2007 6:30 AM
Hi all
Not often I get 3 replies with out abuse, (I am tough skinned though:)), I must be improving my manners!
All I am saying is that the satillite data looks pretty flat lined for me since 2000. I have no qualms about the rise from 1979, but this rise has now settled down since 2000. Tell me I am wrong based on this graph, but it is a horizontal line from 2000 onwards.
From this I am going to say the rate of GW increase has decreased in the first 7 years of this century. What is the slope of the trend line since 1999 or 2000, it will be a lot shallower than from 1979 to 2007.
Also, it looks like ice levels in the artic have returned to normal and the Antartic is not warming, except for the Peninsula. There is more to what is happening than meets the eye, IMHO.
Regards Peter Bickle
Posted by: Peter Bickle | December 14, 2007 7:13 AM
steveB, your constant repetition of "UNPROVEN" shows that you don't know what science is. Science cannot prove, only disprove. If science finds the truth, it is incapable of proving that what it has found is the truth. Science is only capable of proving that wrong ideas are indeed wrong.
Go learn.
Posted by: David Marjanović | December 14, 2007 7:28 AM
Peter, it's really not possible to draw any particular conclusions from short term changes in any data time series. It is apparent for instance that in the last 40 years while the temperature has been trending upwards there are plenty of local minima. If global satellite derived temperatures stay as they are for, perhaps, another 10 years we will have some reason to speculate. Increasing CO2 emissions suggest that this is unlikely.
Arctic ice? it freezes every winter; this is unsurprising. However, that ice is thin and more likely melt quickly in the following summer. (I hope the lack of change in arctic ice in Auckland is reassuring to you.) Frankly noone has any basis on which to evaluate how much of an outlier the 2007 melt season is until the peak of the 2008 melt season. All we know for sure is that 2007 was a long way off the beaten track.
Posted by: markg | December 14, 2007 8:10 AM
Peter seems to forget that New Zealand is surrounded by ocean, which will greatly moderate changes in temperature.
A bit snarky, I know, but isn't it always amazing that confused people or denialists or sceptical people are the first to lecture scientists on how science works, and the last to accept the results when science is actually done.
Posted by: guthrie | December 14, 2007 8:19 AM
Really? You should have. It went from 0.02 deg C in September 1979 to -0.06 deg C in May 2006. How can you call that a rise?
Posted by: Chris O'Neill | December 14, 2007 8:22 AM
Why do people keep responding to Peter Bickle?
You show him the data and he just responds with the same denial and/or just moves the goal posts: From "The latter [satellite data] shows a levelling off in temperture since 2000" to "I am going to say the rate of GW increase has decreased in the first 7 years of this century".
Most people would be embarrassed to keep posting the same rubbish -- and being shown wrong -- time and gain.
Posted by: JB | December 14, 2007 8:50 AM
I'm confused. Why is this different from the previous "A picture is worth a thousand words" post? Is there data/documentation publically available for this one, too?
Posted by: bill r | December 14, 2007 8:58 AM
steveb posts:
[[1) There is an UNPROVEN hypothesis that human activity has increased the atmospheric concentrations of some trace chemicals.]]
Well, first of all, science doesn't deal in "proof" -- only evidence. But there is overwhelming evidence that humans have raised the level of carbon dioxide in ambient air. Hans Suess first detected the isotope signal of fossil fuel carbon in air back in 1955.
[[ 2) There is an UNPROVEN hypothesis that changes in solar activity has increased (and decreased) the atmospheric concentrations of some trace chemicals.]]
Huh? What?
[[ 3) There is an UNPROVEN hypothesis that changes in the global average temperatures have increased (and decreased) atmospheric concentrations of some trace chemicals.]]
Are you talking about the way CO2 goes up or down with Milankovic cycles? I don't think any climatologist doubts it at this point.
[[ 4) There is an UNPROVEN hypothesis that changes in the concentrations of some trace chemicals have increased (and decreased) global average temperatures.]]
It's as close to completely factual as any scientific theory can get. The only reason the Earth isn't frozen over is because of greenhouse gases -- its equilibrium temperature, from solar constant and albedo alone, is well below freezing.
Posted by: Barton Paul Levenson | December 14, 2007 9:02 AM
Could someone please explain how to read the graph to me. Obviously I can see that at some point there is a + in the temperature but what is the 0 mark a reference to? Some average?
Thank you
Posted by: apy | December 14, 2007 9:39 AM
apy,
Yes, the 0 line is an average centered on the decade around 1980. It is a reference line. The graph isn't measuring temperature as such, but the temperature anomaly in comparison to the reference period.
Posted by: luminous beauty | December 14, 2007 10:26 AM
Repeat until you are blue in the face:
Volume, not surface area, matters.
Volume, not surface area, matters.
Posted by: dhogaza | December 14, 2007 11:41 AM
Yep! I don't get it! With some people you can put the charts and graphs right in front of them and they still won't believe! Life can be strange! Dave Briggs :~)
Posted by: Dave Briggs | December 14, 2007 1:02 PM
Reading Carter's pulled quote again, it appears he is using the term "net global warming" since 1998. One wonders what the definition of "net" is in this case and whether Tim's graphic showing a trendline since 1850 is any refutation of Carter's quote. Perhaps by "no net global warming since 1998" he meant no years since 1998 have exceeded 1998 temps?
Posted by: nanny_govt_sucks | December 14, 2007 1:11 PM
For anyone who cares to bother, a modest deconstruction of the statement appears on my blog. Advert over. :)
Posted by: fergus | December 14, 2007 1:22 PM
Forget 1850, the trendline since 1998 refutes Carter's quote. And if he's focussing on the year 1998 specifically, then he might as well be arguing that 1998 was the end of strong El Ninos, since any El Nino comparable to 1998 on top of current temps would unambiguously shatter 1998's record (ignoring NASA's analysis, which shows 2005 and preliminary 2007 as being higher than 1998). It's also amusing that the analysis that shows the flattest temperature trend since 1998 (RSS), also shows the steepest trend of all since 1979.
From these observations, and using the same standards, we might come up with our own headlines, such as "El Ninos stopped in 1998" or "The Urban Heat Island Effect stopped in 1979."
Posted by: cce | December 14, 2007 1:36 PM
nannygovtsucks said: "Perhaps by "no net global warming since 1998" he meant no years since 1998 have exceeded 1998 temps?"
And perhaps the moon is made of string cheese?
There is just one problem with your theory: the very next sentence in Carter's letter:
"That the current temperature plateau follows a late 20th-century period of warming"
There is no plateau in temperature since 1998. Plateaus are flat. The trend is upward, as shown by Tamino.
You know that. I know you know because I can see that you commented on Tamino's post about the upward trend since 1998.
Your effort to defend Carter in this case is truly pathetic.
Posted by: JB | December 14, 2007 2:18 PM
"Also, it looks like ice levels in the artic have returned to normal..."
Well, actually, they havent. even surface area anomaly is still at -1.2 million km2 - a level that in the satellite record was seen once for a short time in 1995, and then often in the last 4-5 years. The arctic is refreezing (duh - its fricking cold up there in the winter) but is still way below the norm up until the last few years.
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.anom.jpg http://arctic.atmos.uiuc.edu/cryosphere/
Not to mention that area is only one measure - and the loss of thick multi-year ice is extensive, and the late-freezing winter ice is likely to end up thinner that normal 1-year ice. There is a HUGE loss of ice volume, way above that shown by the measures of ice area.
Posted by: Lee | December 14, 2007 2:39 PM
Has there been any warming since 1998?
No.
Whatever the reason is irrelevant. The statement is correct.
Has CO2 increased at a terrific rate?
Yes.
It's not a very convenient position for the Climate Faithful, is it?
Posted by: Jack Lacton | December 14, 2007 3:33 PM
@Jack Lacton:
Learn about statistics, outliers and anomalies before trying to pull the stunt of claiming that 1998 was a normal year.
Take a look at the a picture is worth a thousand words post on this same blog. It contains a really nice graph of temperature averages and shows why your argument is worthless.
Posted by: Who Cares | December 14, 2007 3:43 PM
Hello. I came here as someone who is interested in learning more about climate change from an unbiased source. I've seen Tim Lambert's name appear on a lot of blogs so I had to check this place out. I am sadly dissappointed. Is this the best we can do, sophomoric name called and politcal rhetoric? Take the topic at hand for example. From what I can gather, I think it's clear this Bob Carter guy simply states that he is playing with the baseline and purports that the familiar graph we always see does the same. I don't know who he is but I think he's being somewhat glib and I also think he has a point. By saying "It's been cooling since 1998" is designed to get people's attention about things like this graph and how it's constructed. Does the shape of the curve change with a new starting point? If so, how is the ideal starting point determined? Can someone tell me? To the other point regarding Mr. Wegman, again I'm late to the game here but from what I can gather through research he testified before Congress regarding whether or not sound mathematical practices were applied by research scientists, correct? Seems reasonable to me. It therefore also follows that he should stand behind his work and his findings going forward. Anyone with a college degree knows you have to show your work and defend it to get a passing grade. What's wrong with that?
Posted by: Charles the Hammer | December 14, 2007 4:23 PM
Charles Martel, yes indeedy, if you pick a small interval, the shape of a fit to a curve does depend strongly on the starting point which is why this is an evil thing to do with a small interval, or a starting point chosen after you have seen the data. See what happens to the estimate if you picked 1982 or 1992 right after El Chichon or Pinatubo blew off.
"sound" mathematical practices are again one of those things that "sound" good, but often are only cudgels. You may be somewhat disillusioned by the answer. Scientists regard mathematics as a means to an end, the end being learning about the system. Mathematicians regard mathematics as the end and science as a peversion. While the best mathematics is the best tool, scientists feel little compunction about using rough and ready math that happens to be lying about, and on occasion will invent (pull things out of their asses, see for example Feynman integrals, fluxions, Dirac delta fns, etc) things that are not quite mathematically rigorous. Of course, mathematicians want to have perfect answers that are valid for all cases, including the most pernicious, and the cases the scientists look at are almost never in the pernicious category.
Posted by: Eli Rabett | December 14, 2007 4:59 PM
But that isn't what Carter said. You need his definition of "current" before you can level your criticism.
I'm not defending Carter. He seems to have left out his definitions for "net global warming", and "current". If he defined these, then I would know better what he is talking about. In the meantime, Tim's graphic is clearly not addressing the parts of Carter's statements that ARE understandable.
No it doesn't. The trendline since 1850 in Tim's grapnic includes data prior to 1998 to make up the trendline datapoint for 1998 (and probably later). That means that the trendline is not entirely composed of data "since 1998", so it does not refute Carter's contention.
Posted by: nanny_govt_sucks | December 14, 2007 5:38 PM
Peter Bickle:
This is another example of cherry-picking a specific subset of a graph to rationalize a conclusion that you want. It is quite obvious from the graph that the year-to-year variance is such that it is immediately obvious that any trend (or absence of trend) over such a short time period is most likely to be illusory.
Jack Lackton:
I suppose that might be the case, if the "climate faithful" were arguing that CO2 is the only--or even the primary--source of year-to-year variation, rather than a factor influencing long term trends. But of course, they aren't. So this is just another case of deceptive cherry-picking
Charles the Hammer:
The ideal starting point must be determined in an unbiased manner--i.e. without knowing what the data looks like in advance. Unless there is a strong justification for doing otherwise, the best approach is to simply use the entire dataset.
Alternatively, you could pick several starting dates out of a hat at random, and see what the predominant trend is.
Or you could randomly pick a block of 5 years or so. Choose the date in that range with the median temperature as your starting point, and measure the trend from that point for a minimum of 10 years onward.
Posted by: trrll | December 14, 2007 5:46 PM
NOAA announced yesterday: "Including 2007, seven of the eight warmest years on record have occurred since 2001 and the 10 warmest years have all occurred since 1997. The global average surface temperature has risen between 0.6°C and 0.7°C since the start of the twentieth century, and the rate of increase since 1976 has been approximately three times faster than the century-scale trend."
Conclusion: Nanny_govT and Jack Lacton can't read.
Posted by: Thom | December 14, 2007 6:13 PM
"Does the shape of the curve change with a new starting point? If so, how is the ideal starting point determined?"
If one is concerned about the "shape" of the curve and "ideal" starting points (and ending points), there is a very good chance that one is up to no good.
Most scientists (honest ones, at least) prefer trends over "net" changes between endpoints.
The curve may be oscillating up and down (due to El Nino, for example) and each of the endpoints may fall on a different part of the oscillatory signal (eg, peak and valley), but if the oscillatory signal is sitting on an upwardly trending "ramp", linear regression will find it -- and, importantly, the endpoints will not assume greater weight than any other points in the data series.
Posted by: JB | December 14, 2007 6:20 PM
Speaking of graphing, I'm going to start graphing the frequency of na_gs' mendacicizations and their type.
That is: is the frequency related to his presumed length of time it takes us to forget his last foray into FUD?
Best,
D
Posted by: Dano | December 14, 2007 7:16 PM
To my comment "There is no plateau in temperature since 1998. "
nags said: "But that isn't what Carter said. You need his definition of "current" before you can level your criticism."
Why does your statement remind me so much of Bill Clinton's "it depends on what the meaning of is is"?
You and I both know (from Tamino's blog) that the temperature has trended upwardly whether one starts in 1998 or 2000.
(Of course, you would never admit that because you don't admit being wrong even when Tamino spanks you hard)
If Carter's "plateau" comment refers to a period starting in 2001 or later, then he is really playing games because the shorter the period, the less confidence one can have in the trend.
Posted by: JB | December 14, 2007 9:34 PM
Peter Bickle:
Actually the satellite measurements have gone down substantially over 2007. Bickle and co are a bit behind the times. They should be saying we are now experiencing global cooling.
Posted by: Chris O'Neill | December 14, 2007 10:00 PM
Speaking of graphing:
1) "Eyeballing" a time-series chart is not a good idea. In particular, human eyes are drawn to local minima/maxima.
2) Picking any extreme point and drawing a straight line is also not a good idea.
3) At the very least, one needs to do a linear regression, or if one is serious, one can do many more things. I recommend tamino's fine Analyze This tutorial.
4) People who are mathematically literate knows this stuff. If one is not, it's easy enough to learn. However, some people who are not mathematically illiterate do these things, in which case, keep an eye on your wallet, and read "How to Lie with Statistics", "How to Lie with Charts", "How to Lie with Maps", "Innumeracy", "Damned Lies and Statistics" in self-defense against against clever cherry-pickers and frauds.
===== For this particular case, I suggest a simple exercise that anyone can do using Excel or equivalent.
A) Start with NASA GISS Global Annual Mean Surface Air Temperature Change, used to create the graph. [I use GISS rather than Tim's Hadley chart because the data is simpler for this exercise.
B) Select a set of {year, temperature anomaly, 5-year-smoothed anomaly}, ending at 2006. I picked 1977 as an example, but as we'll see, the start date doesn't matter for this exercise.
C) Paste the data into the spreadsheet, starting at A1, i.e., A1 = " 1977 .13 .00"
Select A1:A30. Data>Text to Columns Choose Delimited, >Next Choose Space and "Treat Consecutive delimiters as one", >Finish Select Column A and Delete (This puts the 3 columns into A, B, and C.) Select columns B-C, Format Cells, Number, 2 decimal places. Select columns D-E, Format Cells, Number 3 decimal places
D) Set Cell D1: =(B$30-B1)/(A$30-A1) Then Fill Down (Ctrl-D) to cell D29 (2005). NOTE: the $ are important.
D1:D29 gives the temperature slope from each year to 2006, totally dependent only on 2006 and the choice of start year, and ignoring everything else.
E) Set cell E1: =SLOPE(B1:$B30, A1:$A30) Then Fill Down (Ctrl-D) to E29.
E1:E29 gives the slope fo the linear regression from each year to 2006. Naturally, D29 = E29 (in this case, -.08), since the regression for 2005 to 2006 is the same as the simple slope.
QUIZ0: Any difference between columns B (yearly) and C (5-year smoothed)?
QUIZ1: Any difference between columns D and E?
QUIZ2: If you're a cherry-picker, which of column D or E would you choose if you wanted to prove warming had stopped? Which start years would you pick? [Although, most wouldn't try to pick 2005, since most people understand that year-to-year gyrations happen.] Column D of course is the analog of eyeballing the chart, and if you happen to pick a local minimum or maximum, you can get what you want.
F) Now, let's do a graph.
Select A1:A29, then CTRL-select D1:D29, so you have 3 columns, 1977-2005.
You now have a chart in which: Series1 gives the simple slop from that year to 2006. Series2 gives the regression slope to 2006. Expand the chart for visibility.
Series1 jiggles all over the place, and even manages to get below zero 4 times. Series2 is much more stable, and only fluctuates much in the last few years, unsurprisingly.
QUIZ3: What will happen when 2007 results are in? A: Series1 may look different, because each point only depends on the values of start and end years, whereas Series2 won't look much different, except at the very end, because it depends on all the intermediate data as well.
Series2 hovers around .02 degrees/year. Anyway, this all takes 15 minutes or less,
Posted by: John Mashey | December 14, 2007 10:44 PM
Well, no, the letter is meant to try to convince the world that CO2 doesn't cause warming and that therefore we should do nothing about it.
You can read it for yourself, you know? Tim conveniently provided a link.
Posted by: dhogaza | December 14, 2007 11:47 PM
That's beside the point. The graphic Tim presented does not support a refutation of Carter. This is not an issue about warming or cooling since a certain time. It's about whether a trendline is the same from two different starting points.
Take it up with Carter.
Posted by: nanny_govt_sucks | December 15, 2007 12:47 AM
Wrong. It is about whether it is valid to cherry-pick 1998 as the starting point in order to alter the trendline. The answer is one known to anybody who has studied elementary statistics: it is not. The graphic proves that 1998 is cherry-picked as a statistical fluctuation, much higher than either the years before or the years after, proving that 1998 is an invalid choice of starting point.
Posted by: trrll | December 15, 2007 1:16 AM
The world has a never-ending supply of dishonest nutcases. You don't "take it up with them" because by definition you don't expect a rational response. You warn people about them.
Posted by: Chris O'Neill | December 15, 2007 1:40 AM
Trrll:
If we can say with confidence it is warmer now than any time even in the past 1000 years, does that reliably entail anything about our knowledge of global climate other than the fact being asserted? Measuring from 1998 or 1558 doesn't seem to actually entail anything sans knowledge of the necessary and sufficient conditions of the temperatures at the various points in time.
Posted by: Kevin | December 15, 2007 2:03 AM
"'WEGMAN: Carbon dioxide is heavier than air. Where it sits in the atmospheric profile, I don't know. I'm not an atmospheric scientist...' Perhaps the blue on the graph above is supposed to represent CO2 settling into the vall