Open Mind

What’s Up With That?

March 2, 2008 · 162 Comments

Anthony Watts has a post comparing global average temperature anomaly from 4 major sources. Two of them are surface temperature estimates: GISS (NASA’s Goddard Institute for Space Studies) and HadCRU (the Hadley Centre/Climate Research Unit in the U.K.). The other two are satellite estimates of lower-troposphere temperature, from UAH (University of Alabama at Huntsville) and RSS (Remote Sensing Systems). It’s an interesting thing to compare these data sets, but it’s well to bear in mind that they don’t measure the same thing. The surface data are estimates of actual surface temperature, while the satellite data are for the lower troposphere, i.e., a rather large segment of the lower part of earth’s atmosphere. Nonetheless, we expect these data sets to show strong correlation.


And so they do, as is well illustrated by Watts’ first graph (click the graph for a larger, clearer view):

So far so good. But now we see our first sign of serious trouble with Watts’ analysis of the situation:


I was particularly impressed with the agreement of the 4 metrics during the 1998 El Niño year as well as our current 2008 La Niña year.

Whoa! Doesn’t Watts realize that these data sets are on different scales?

All these data are temperature anomaly. Anomaly is the difference between temperature at a given time, and the average temperature for the same time of year during some reference period. So temperature anomaly doesn’t really tell you, in absolute terms, how hot or cold it is — it tells you how much hotter or colder it is, than it was (on average) during the reference period. And there’s the rub: these data sets use different reference periods. GISS uses the reference period 1951 to 1980, HadCRU used 1961 to 1990, and the satellite estimates use 1979 to 2000. The coldest of these reference periods is the 1951-1980 GISS reference, the warmest is the 1979-2000 satellite reference. That means that GISS anomaly is the difference between present temperature and a colder time period, satellite data are the difference between present temperature and a warmer time period.

We can’t directly compare the numbers in a meaningful way without compensating for the difference in reference. Otherwise, it’s just like measuring my height in inches above Shaquille O’Neill (which makes the number quite negative) while measuring a newborn child’s height in inches above the ground (which makes the number certainly positive), noting that the infant’s number is greater, and concluding that the newborn is taller than I am. If we fail to compensate for the different reference, then we expect that the GISS numbers will be highest, the HadCRU numbers next, and the satellite data lowest. And that’s exactly what we observe.

Pretty basic, right? Anybody who pontificates about trends in temperature metrics, and compares different sources of data, really should know this, right?

Yet it’s abundantly clear that Watts doesn’t know this — at least, he didn’t when he made this post. But it does give the opportunity to show that the GISS numbers are considerably higher than the others, and conclude (in ignorance) that it’s showing more heating than the others, complete with the suggestion that GISS has somehow made this happen. Watts doesn’t directly say so himself, he lets a reader do it for him:


One of the first comments from my last post on the 4 global temperature metrics came from Jeff in Seattle who said:

Seems like GISS is the odd man out and should be discarded as an “adjustment”.

Watts goes on to opine about some of the things he and others believe are wrong with the surface temperature record. It’s a lead-up to this:


So with all of this discussion, and with this newly collated data-set handed to me today, it gave me an idea. I had never seen a histogram comparison done on all four data-sets simultaneously.

Doing so would show how well the cool and warm anomalies are distributed within the data. If there is a good balance to the distribution, one would expect that the measurement system is doing a good job of capturing the natural variance. If the distribution of the histogram is skewed significantly in either the negative or positive, it would provide clues into what bias issues might remain in the data.

Of course since we have a rising temperature trend since 1979, I would expect all 4 metrics to be more distributed on the positive side of the histogram as a given. But the real test is how well they match.

Astounding.

Watts shows that for the satellite data sets, the anomalies are reasonably well “balanced” between positive and negative, but for the surface thermometer data sets they’re not. When he discusses the HadCRU anomalies he says:


Here, we see a much more lopsided distribution in the histogram. Part of this has to do with the positive trend, but other things like UHI, microsite issues with weather station placement, and adjustments to the temperature records all figure in.

Watts has shown that most of the HadCRU anomalies are positive. But what that really means is that the period under consideration (1979 to Jan. 2008) is almost entirely warmer than the reference period (1961-1990). In other words, the globe has warmed since then. But Watts concludes that he’s found evidence of UHI (urban heat island effect), problems with weather station placement, and adjustments to the temperature record.

Turning his attention to GISS, Watts says:


I was surprised to learn that only 5% of the GISS data-set was on the cool side of zero, while a whopping 95% was on the warm side. Even with a rising temperature trend, this seems excessive.

When the distribution of data is so lopsided, it suggests that there may be problems with it, especially since there appears to be a 50% greater distribution on the cooler side in the HadCRUT data-set.

Interestingly, like with the satellite data sets that use the same sensor on the spacecraft, both GISS and HadCRUT use many of the same temperature stations around the world. There is quite a bit of data source overlap between the two. But, to see such a difference suggests to me that in this case (unlike the satellite data) differences in preparation lead to significant differences in the final data-set.

It also suggests to me that satellite temperature data is a more representative global temperature metric than manually measured land-ocean temperature data-sets because there is a more unified and homogeneous measurement system, less potential bias, no urban heat island issues, no need of maintaining individual temperature stations, fewer final adjustments, and a much faster acquisition of the data.

Once again, Watts has used an entirely invalid comparison to conclude that there are problems with GISS or HadCRU data due to “differences in preparation.” He also concludes that the satellite data sets are more reliable, because they’re in better agreement. Of course they are; they’re using the same reference period.

The right way to compare these data sets is to use a common reference period for all of them. I did that here for the three major surface temperature records. For those who want to see what it looks like using the four sets chosen by Watts, here it is (click for a larger, clearer view):

Watts has been informed of the difference in reference period of the various data sets:


One of the things that has been pointed out to me by Joe D’Aleo of ICECAP is that GISS uses a different base period than the other data-sets, The next task is to plot these with data adjusted to the same base period. That should come in a day or two.

Yet surprisingly, he doesn’t put two and two together to get four.

I had decided not to post about this, in spite of its being discussed in reader commentary on this blog. This is nothing but an utter bonehead move by Watts, and we all make bonehead moves from time to time. One would hope that the folly of Watts’ analysis would be evident, so that his error would be quickly corrected and have no substantive impact on the global warming discussion. Besides, I’m working on an eagerly-awaited post on PCA and its use in proxy reconstructions, and even ‘though this is a “quickie” post it still takes time (I expect the final installment of the PCA series to be complete this week).

But I’ve noted over the last several days that Watts’ analysis is being reported on a number of anti-global warming websites, complete with links to the post and copies of some of his graphs. Apparently it’s important for the truth of this matter to be put into the blogosphere now, while the fur is flying, and not just as a comment on an unrelated thread. Yet although Watts has been informed of the problem, he hasn’t acknoweldged that it invalidates his analysis, he’s simply promised us more to come:


UPDATE: I’ve decided to make this a 3 part series, as additional interest has been generated by commenters in looking at the data in more ways. Stay tuned for parts 2 and 3 and we’ll examine this is more detail.

I’m very curious to hear what Watts will have to say on the matter.

Categories: Global Warming · climate change

162 responses so far ↓

  • TCO // March 3, 2008 at 12:17 am

    A. Just plot them all starting at “zero” like comparing stocks on a ticker. That’s a very appealing way of showing deviation and is what people are used to.

    B. Anthony Watts has a tendancy to confuse multiple hypotheses. He sees some big warming at a station with no movement of the sensor and says “looks like classic UHI”. Well. It MIGHT be UHI. I’m a skeptic. I want it to be UHI. I want to check for it being UHI. But I at least realize that it might be…GLOBAL WARMING.

    C. Anthony is a nice guy. But he’s really not that sharp. Take it easy on him.

  • Kevin // March 3, 2008 at 1:37 am

    re: TCO
    Anthony Watts may well be a nice guy. I don’t know much about him. I will, however, form some judgment about his character based on how he handles this once he achieves some clarity about the fact that anomaly data are relative to a particular reference period, and what the implications are for the type of analysis he’s trying to do. If I don’t see a strongly worded correction, I’m going to have to assume that he’s more interested in sowing doubt about good science than he is interested in pursuing truth. The fact that people are already echoing his mistake makes a strong correction from him a moral imperative, in my opinion.

  • TCO // March 3, 2008 at 2:02 am

    There’s a lot of moral imperatives floating around…

  • Brian D // March 3, 2008 at 2:06 am

    Even if he does issue a correction, the damage has been done — people will be quoting this one, or the places who quoted him, for a while. Which is why it’s so important to make sure that analysis like this is done skillfully *before* it becomes public. This is what peer-review is for, but the blogosphere lacks a peer-review process (or, at least, any process that can prevent poor reasoning from being published).

    Remember the whole fiasco a while back about the 15-year-old girl who plotted graphs on the wrong timescale to “disprove” Gore, who got quoted so passionately amongst the denialists? I keep thinking of that when I read about Watts.

  • Hansen's Bulldog // March 3, 2008 at 2:20 am

    Although it’s true that some of “the damage has been done,” that doesn’t mean that a clear rejection of the analysis, by Watts himself, won’t be valuable.

    It seems to me that it belongs not only on his own blog, but on the very post itself. After all, the references to it I’ve seen have links. Those who click the link, instead of seeing details of the analysis, will instead see Watts’ disclaimer.

    And Brian, you highlight another tremendous advantage of the peer-review process. Everyone has the capability of making a total bonehead move. Peer review catches most of those, and protects us from embarrassing *ourselves*.

  • TCO // March 3, 2008 at 2:31 am

    99% of the time, I avoid the tit-for-tat request thing, but let me indulge just once. Could we get a clear statement from Micheal Mann about the “acentric” centering in his PCA transform? I want to know if it was intentional or not (if so why…if not, then stating that people should not repeat it).

    The issue has become polluted by Steve McI confounding other aspects of the analysis with that one “flaw” (and many people even on Mike’s side believe it a tehcnical, minor flaw). Note I’m not trying to say that the problem was extreme…and while I have not crunched the math myself, my bet would be that the effect itself of that one factor is small (based on McI’s refusal to deconfound and measure impact).

    However, Mike’s truculence and refusal to address this is bad practice. It isn’t even excusable if he thinks (or even knows) that his enemies will exploit an admission with claims of wider problems. The real scientist always admits and corrects aNYTHING that is clearly wrong. Regardless of the meta-debate.

    At least Hansen admitted the Y2K mistake and fixed it (and it was minor as well…but was a flaw). Real scientists act this way. No bulldogggggging when you’re wrong.

    [Response: I came very close to deleting this, because I've stated explicitly that the issue of The Mann et al. reconstruction is precisely the topic of the imminent post on PCA, those who wish to argue the point can wait until the subject arrives, and won't have to wait very long. It's yet another issue that comes up all the time. I'm not shying away from it, but as I've said elsewhere, I am NOT letting yet another thread get hijacked by this issue. When the post arrives, everyone will have the opportunity to chime in, but until then that topic is off limits.

    Which applies to friend and foe (and neutral) alike.]

  • caerbannog // March 3, 2008 at 2:32 am

    This silly denier argument has already reached the on-line discussion-board of my hometown newspaper, the San Diego Union Tribune.

    I took a couple of minutes to smack it down with a link pointing right back here.

    And many thanks, Tamino, for taking the time to post this. I’m going to make sure that lurkers at the Union Tribune message boards have a chance to read your deconstruction of this latest silliness.

  • TCO // March 3, 2008 at 3:14 am

    I’ll go look in those posts. Was worried they would be theoretical and dry. and hard. BTW, I’m not neutral. I’m a skeptic. I mean a denialist. ;)

    [Response: The first three of the series (already posted) lay the foundation for addressing the issue. The next to come (some time this week) will specifically address the application of PCA by Mann et al. You might find the 2nd of the series theoretical and dry.]

  • Hank Roberts // March 3, 2008 at 3:54 am

    > Remember the whole fiasco a
    > while back

    The eponymous notion called something like ‘Plunder the Blunder’ has leaped directly from childish error to iconic item of faith without ever having met peer review. Too bad.

    A process rather like the one Oscar Wilde remarked on when he noted “America is the first country to have gone from barbarism to decadence without the usual intervening period of civilization.”

  • Martin // March 3, 2008 at 7:11 am

    TCO:

    A. Just plot them all starting at “zero” like comparing stocks on a ticker. That’s a very appealing way of showing deviation and is what people are used to.

    Not a good idea… the temperature record is very noisy, and this would contaminate every time series with (the negative of) its own noise (i.e., the accidental, random deviation from the trend line) at the starting point.

    The proper way is what (I suppose?) was done here: “un-anomalize” every time series by re-adding the subtracted-out average for its respective reference period.

  • Adam // March 3, 2008 at 10:22 am

    Watt’s blog post would probably have got published in E&E. He missed a trick there.

  • P. Lewis // March 3, 2008 at 10:43 am

    Was there an update on Anthony Watts’ post of Joe D’Aleo’s paper that Tamino (the blogger now known officially as Hansen’s Bulldog) wrote about?

    Still waiting I think. But perhaps Anthony Watts still believes 100% in Joe D’Aleo.

    Warning: holding your breath may lead to self-suffocation.

  • TCO // March 3, 2008 at 2:19 pm

    Martin: Good point. Hmm…you might say the same thing about stocks. Double hmm.

  • Lee // March 3, 2008 at 4:11 pm

    So I visited the “watts up with that” blog this morning, Mar 3, It sems to be up, adn workign fine, everything accessible, all comments readable - except that clicking on comments on that one thread, the one Tamino looked at here, yields this error message:
    —-
    Error 404

    Sorry, but you are looking for something that is not here
    —-

    Looks like the comments on that thread have been taken down. No correction or acknowledgment of his error that I could find, either.

    Hmmmm….

  • Lee // March 3, 2008 at 4:29 pm

    And now the thread itself is gone.

    Not corrected - gone without a trace.

  • Hansen's Bulldog // March 3, 2008 at 4:45 pm

    Lee, I have no reason to doubt your personally, but that’s an extraordinary claim. So I checked for myself.

    And you’re quite right. The whole thread has disappeared; I got the message “Sorry, but you are looking for something that is not here.”

  • P. Lewis // March 3, 2008 at 5:01 pm

    Watts’ site has since been updated to read

    UPDATE: I had mentioned that I’d be looking at this in more detail in parts 2, and 3. However it appears many have missed seeing that portion of the original post and are saying that I’ve done an incomplete job of presenting all the information. I would agree for part1, but that is what parts 2 and three were to be about.

    Since I’m currently unable to spend more time to put parts 2 and 3 together due to travel and other obligations, I’m putting the post back on the shelf to revisit again later when I can do more work on it, including show plots for ajdusted base periods. In the meantime, poster Basil has done some work on this of interest which you can see here

  • Hansen's Bulldog // March 3, 2008 at 5:05 pm

    Curiouser and curiouser. P. Lewis quotes the new version of the post (which has now reappeared at the original URL), but it should also be mentioned that the old version hasn’t just been updated — it’s been deleted, replaced with the text quoted by P. Lewis.

  • Eli Rabett // March 3, 2008 at 5:07 pm

    Hi, I looked at the four records about a month ago in a post which was titled

    Food for Tamino.

    If there was an outlier it was HadCRUT, which appears to have much less variability than the others. A major issue (such as it is) is the convergence of the two satellite records

  • Ken // March 3, 2008 at 5:11 pm

    TCO - It doesn’t matter if “Anthony is a nice guy” or not. Just by identifying himself as a meteorologist and posting such easily caught errors makes him a part of the problem (i.e. separating truth from error). The general public doesn’t distinguish between a meteorologist, who’s career was spent on TV, and research climate scientist with decades of numerical analysis experience. People like Anthony only cloud the issue by misleading the less informed. And retraction or not, the damage is done (i.e. more people are now more ignorant & confused than before reading Watts post) and will continue to spread on web-sites for years.

    Tamino, thanks for the post. I’ve been trying to explain Watts errors in other forums, and having the ability to link to your clear analysis will be great.

  • mmghosh // March 3, 2008 at 5:40 pm

    Mr Watts is at the Climate Change Conference organised by the Heartland Institute today, I believe.

    http://www.heartland.org/NewYork08/ConferenceTable.pdf

    He is scheduled to give a presentation in the afternoon today on Climatology.

  • P. Lewis // March 3, 2008 at 6:11 pm

    What I hear, I think, are the “wheels of disingenuity” grinding. What I hear, I think, are lacks of openness, fairness, honesty, and supposed scientific integrity. And where’s that audit path?

    Actions speak louder than words, they say, and I’d contend that Mr Watts’ actions in connection with Hansen’s Bulldog’s/Tamino’s posts on this matter and the earlier Joe D’Aleo/”Warming Trend: PDO And Solar Correlate Better Than CO2″ episode do seem to speak loudly. But I don’t suppose many of Anthony Watts’ acolytes would concur.

  • cce // March 3, 2008 at 6:13 pm

    Watts’ post was reported by Brit Hume during one of his “Political Grapevine” segments.

    http://www.foxnews.com/story/0,2933,333328,00.html

    Of note:
    “That is said to be a value large enough to erase nearly all the global warming recorded over the past 100 years. It is reportedly the single fastest temperature change ever recorded — up or down.”

    I forget the exect months, but for the satellite analyses, there were ~1 degree upward swings over 13 months from ‘97 to ‘98, and ~0.8 degree downward swings from ‘98 to ‘99. From ‘94 to ‘95 there was an upward swing in the GISS analysis equivalent to the recent downward swing, but that might be Pinatubo related. No equivalent swing in the HadCRUT analysis, although I only went back 30 years.

    The segment ends with this quote:
    “Some scientists contend the cooling is the result of reduced solar activity — which they say is a larger driver of climate change than man-made greenhouse gases.”

    January 2007 was the warmest month in years (warmest ever in the NASA analysis), but January 2008 was cold so it must be solar? Out with the 11 year solar cycle, in with the 2 year cycle!

    [Response: I think the Faux News report is not about the post referred to here, but about an earlier Watts post. They also repeat the "erased all the global warming so far" meme, which turn of phrase didn't come from Watts, and he himself has disavowed.]

  • Hank Roberts // March 3, 2008 at 6:24 pm

    Heartland:
    > S. Fred Singer, The NIPCC Report

    They don’t miss a PR trick.

  • George // March 3, 2008 at 7:24 pm

    Bulldog said: “I got the message “Sorry, but you are looking for something that is not here.”

    It strikes that this disclaimer should really accompany many (if not all) of Watts’ posts on the web.

    Looking for “Actual proof that El Nino drives global warming”?

    “Sorry, but you are looking for something that is not here.” (at “Watts up with that?”)

    Looking for “Actual proof of ‘contamination’ in the surface record”?

    “Sorry, but you are looking for something that is not here.” (at surfacestations.org)

  • Surly // March 3, 2008 at 7:40 pm

    http://209.85.207.104/search?q=cache:K-ak-qFlSYsJ:wattsupwiththat.wordpress.com/2008/02/27/a-look-at-temperature-anomalies-for-all-4-global-metrics/+A+look+at+temperature+anomalies+for+all+4+global+metrics:+Part+1&hl=en&ct=clnk&cd=1&gl=ca

    I believe that is the cached version — for comparison’s sake.

  • Lee // March 3, 2008 at 8:20 pm

    This morning, I posted a questin at WAtts’ blog, on the ”
    Day !… thread”, and he answered:

    Lee (08:35:25) :
    Anthony,
    What happened to your “histogram” analysis of the four temp records? The one that;s been getting play all over the internet?
    REPLY: it has been edited until I can work on it again.
    —-
    I just posted a followup. We’ll see how he treats this.
    —-
    No, Anthony, it has not been “edited.” It has been removed entirely, along with all comments. You pulled it down from your blog, and replaced it with a note that it was “incomplete”. Not, it was not incomplete. It was incorrect. You were comparing records that are on different scales, and treating them as if they were on the same scale.
    Removing the post and comments on the post is not he proper way to deal with an error.

    [Response: I just looked, and the post is still replaced with the "put on the shelf" note, but the comments seem to be intact.]

  • Marion Delgado // March 3, 2008 at 8:43 pm

    When evaluating climate denialists, a key error would be to assume every fresh offense against the truth is exceptional or happening in a vaccuum. Go back to when Anthony Watts first started posting on climate blogs - get a notebook. note down every time he says something provably false. circle everything he says that’s provably false AFTER it’s been shown to be provably false.

    If you wish, rank them 1-5 or 1-10 or something. Then look at your results. THEN decide what he’s up to.

  • cce // March 3, 2008 at 8:52 pm

    So it passed through Tech Central Station on the way to Fox News. “Grapevine” is right.

  • Hank Roberts // March 3, 2008 at 9:00 pm

    > Watts up
    “Due to extensive traffic today on my main website, which is being deluged and will be taken offline for awhile, ….”

    Must be time for the speech at the big hotel.

  • cthulhu // March 3, 2008 at 10:04 pm

    mwhaha the original post was hillarious enough but the update tops it.

    His justification for removing the post is not that it’s quite clearly incorrect but that it’s “incomplete” and he just doesn’t have time to , you know - complete it. Also it’s the readers fault for “missing” what he was actually saying.

    Apparently the whole post was incomplete at the time, up in the air, a light laugh, not really serious in the first place.

    I have to wonder when he says he will eventually “do more work on it” whether this means he’ll “fix it” or “continue screwing it up”

    The pie charts in part 2 would be quite amazing, we would have seen how the GISS pie chart is redder than the others, red being the color of bias of course. Also red being the color nearly always used to highlight the spot of interest in mineral promotions.

  • dhogaza // March 3, 2008 at 11:39 pm

    Also red being the color nearly always used to highlight the spot of interest in mineral promotions.

    ROTFL! I’d forgotten about that particular idiocy on the part of McIntyre …

  • Lee // March 4, 2008 at 12:55 am

    Tamino, you are correct. The comments thread is back up. It was still “does not exist” when I had checked.

    There is still at least one place in the comments where someone has posted a links to their own graphics of Anthony’s “analysis,” with Anthony’s approving comments about it.

    What’s really funny is the comments pointing out how the various distributions have nearly the same standard deviations - duh!.

  • cce // March 4, 2008 at 1:01 am

    Just caught the tail end of Glenn Beck (don’t worry, it will be repeated about 20 times in the next day). Got to see Bob “Global Warming Stopped in ‘98″ Carter talking about the weather and sea ice. Then it was on to Watts and his photos and graphs. Didn’t get to see JohnV’s graph, for some reason.

  • Ailuropoda Melanoleuca // March 4, 2008 at 2:27 am

    I saw that post but only skimmed through it. At the end, I was tempted to tell him that drawing a conclusion based off a trend on a one year time frame was the equivalent of winning a small jackpot on a slot machine after three tries and running around claiming that the casino is loosing money. I saw a lot of ups and downs in the chart Watt posted and concluded that longer time frames are needed to draw any meaningful conclusions on data trends. Though at the end, I did draw the meaningful conclusion that it wasn’t worth arguing about it with him.

  • fred // March 4, 2008 at 6:46 am

    I am going to defend Mr Watts. Not against the charge of having made an error when he compared the different time series. This he seems to have done, and its unfortunate, but we do all make mistakes.

    But his work on surface stations has been valuable and necessary and has stimulated valuable work by others. The pity is that this sort of work had not been done by the people who run the sites or who use the data they provide.

    The US surface station record and the stations needed scrutiny. ROW needs it still more. What we have seen so far, and its a valuable contribution, is that a majority of US stations surveyed have been out of spec. We have seen in another part of the wood that 74% of stations have had some part of their readings altered. From McIntyre, confirmed by Atmoz, we have also discovered that the algorithm by which the alterations were done has not had the effect it is supposed to have.

    Now, it may be that all this is immaterial to the validity of the trend reported by GISS. There is no way of knowing before the work is done, in spec stations used, and the algorithms cleaned up. Thanks to Watts and the volunteers, and thanks to Atmoz and McIntyre, we are beginning to get an idea of what has to be done to put the surface temperature record on a sound footing. Everyone should look forward to seeing this done, instead of abusing the project. And instead of delivering these wild emotional rants and personal attacks defending a series of poor administrations and problematic changes to the data. Everyone should want this, because it is interesting as well as important to get to the real extent of recent warming. And the warming that may or may not be to come.

    Obviously one year’s cooling data, the other thing Watts has posted on recently, while interesting, is not a trend. Something it would have been nice for everyone to have born in mind when acclaiming various summers and years as ‘the warmest since…’

    Too many here seem to take all this far too personally. There is nothing wrong with scepticism. There is nothing wrong even with obdurate persistence in looking for evidence against an hypothesis when all the evidence seems to be for it. It is not a mark of moral weakness. It is an essential element in the public debate, from which the truth will eventually emerge. Yes, I know most of you think it already has. Well no, it hasn’t. People are still arguing. But it will. Have a bit more faith in humanity’s ability to find its collective way through, and stop abusing people who you may differ from, but who are in their own way contributing to the process.

  • Hansen's Bulldog // March 4, 2008 at 12:13 pm

    There is nothing wrong with scepticism.

    Permit me to express skepticism at the suggestion that Watts’ and McIntyre’s efforts deserve the name “skepticism.”

  • Boris // March 4, 2008 at 1:27 pm

    Watts and McIntyre provide fuel for right wing conspiracy theorists:

    http://newsbusters.org/blogs/noel-sheppard/2007/09/17/nasa-s-hansen-playing-enron-accounting-games-climate-data

    ” h/t Anthony Watts):”

    “Of course, when Enron and other companies were found to be cooking their books at the expense of investors, this was front page, headline news for months.”

    I don’t know how they’d go about things differently if it was Watts and McIntyre’s intention to promulgate a conspiracy theory. Hmmmm…

  • Imran // March 4, 2008 at 2:06 pm

    Indeed it doesn’t matter if he’s a nice guy or not … and indeed he has made a boneheaded mistake …. but that shouldn’t detract from the fact that he has a point …. he is just articulating what an increasing number of people are observing. The world, in general, is becoming increasingly skeptical because its becoming increasingly clear that the warming predicted in various IPCC reports hasn’t happened. We can rant about the science, and talk about the difference between long term CO2 warming vs. ‘weather’ … but the truth is that we the predictions made have woefully underestimated the low side in the last decade - particularly the IPCC 2001 report -which has now got ~ 8 years of actuals. It doesn’t matter if its meant for the long term because it has been proven wrong in the short / mid term. That’s a fact. And any bonehead can pull out Hansens monthly data since 1998 and its very very hard to see any warming trend there.

    [Response: You really don't flatter Watts when you defend him with more mistakes, including (but hardly limited to) cherry-picking 1998.]

    We need to stop using terms like ‘enemies’ and ‘the other side’ and admit we all know a hell of a lot less than we think we know. Lets get back to science with an open mind - we all have to say ‘I might be wrong’. This debate is way to important to be so polarised with much of the science still in its infancy.

  • George // March 4, 2008 at 2:14 pm

    There is nothing wrong even with obdurate persistence in looking for evidence against an hypothesis when all the evidence seems to be for it. It is not a mark of moral weakness. It is an essential element in the public debate, from which the truth will eventually emerge.”

    Like it did in the case of smoking?

    After years of lies, obfuscation and other deception on the part of the tobacco industry?

    After millions of deaths due to smoking related illnesses worldwide?

    Besides, calling certain ideas stupid is not a moral judgment — not if they are based on faulty reasoning and faulty facts.

    There is an easy solution for those who do not wish to have their ideas subjected to rigorous criticism. They should simply stay out of the debate.

    In case no one noticed: science is not a picnic in the park with the relatives where you have to be on best behavior.

    The whole idea that people with no expertise in climate science specifically, or a relevant field like mathematics (as in the case of Bullydog) think they know more about this stuff than the scientists who have made it their life’s work to study it is just absurd.

    It’s like going to get brain surgery and having some guy off the street with no training in medicine sitting there in the operating room saying to the actual surgeon “This is what I think is wrong and this is what I think you should do. He has migraines, you say? Off with his head!”

  • Hank Roberts // March 4, 2008 at 2:15 pm

    Well, yeah, and the government went “oh, tsk tsk” when producers encouraged users to go grossly into over-leveraged energy debt, overshooting the environment’s ability to support the payments.

    Shocked, shocked ….

  • mmghosh // March 4, 2008 at 2:32 pm

    Fred

    I think what you are forgetting is that these stations have been set up primarily as weather stations, not climate stations. They need to supply local populations with weather data, i.e. I live in the city, so I need to know what the temperature in my city is, so that I can decide whether I need a sweater or not. This means that the weather data has to inform me about the temperature (including any UHI effect) if any. I do not really need to know the rural temperature 500 miles away, although that may reflect the “correct” state of warming of the earth.

    Now, as has been pointed out by the author of this blog, it is possible to take a trend only of selected isolated “rural” sites. If I am not mistaken, this has already been done, and shows that it matches the current warming trend identified by other techniques.

    However, climate scientists feel (in my opinion rightly) that all weather stations contribute to overall global warming data when overall climate considerations have to be made. If urban areas have to be taken into consideration, then certain corrections have to be made. As I am not an expert, I would tend to trust the general body of climate scientists to come up with a defensible technique of doing this where they are answerable to their peers in conferences and peer-reviewed literature.

    While Mr Watts may have legitimate concerns, the correct place to voice them is present his alternative data via the peer review process through journals and conferences. It may be slower, and less attention-grabbing, but the science that emerges is ultimately stronger.

  • caerbannog // March 4, 2008 at 3:10 pm


    Obviously one year’s cooling data, the other thing Watts has posted on recently, while interesting, is not a trend. Something it would have been nice for everyone to have born in mind when acclaiming various summers and years as ‘the warmest since…

    You need to learn the difference between a sample size of 1 and a sample size of N.

  • luminous beauty // March 4, 2008 at 5:03 pm

    “There is nothing wrong with skepticism.”

    Using the word as a semantic shield to defend what is at it’s root, cynical paranoia, is right?

  • Hansen's Bulldog // March 4, 2008 at 5:20 pm

    Well, I guess it got a little too “hot” for Anthony Watts. His post now looks like this:

    I recently plotted all four global temperature metrics (GISS, HadCRUT, UAH, RSS) to illustrate the magnitude of the global temperature drop we’ve seen in the last 12 months. At the end of that post, I mentioned that I’d like to get all 4 metrics plotted side-by-side for comparison, rather than individually.

    UPDATE: I had mentioned that I’d be looking at this in more detail in parts 2, and 3. However it appears many have missed seeing that portion of the original post and are saying that I’ve done an incomplete job of presenting all the information. I would agree for part1, but that is what parts 2 and 3 were to be about.

    Since I’m currently unable to spend more time to put parts 2 and 3 together due to travel and other obligations, I’m putting the post back on the shelf (archived) to revisit again later when I can do more work on it, including show plots for adjusted base periods.

    The post will be restored then along with the next part so that people have the benefit of seeing plots and histograms done on both ways. In part 3 I’ll summarize 1 and 2.

    In the meantime, poster Basil has done some work on this of interest which you can see here.

    The original post has disappeared from view. And even this version may soon change; he added the 2nd-to-last paragraph since the version reported by Lee (in reader comments here).

    I think his “update” is shameful. It’s a pathetic insult to imply that any part of the problem is that readers “missed seeing” his intention to do it “in more detail” in parts 2 and 3. The post didn’t need more detail; it needed retraction. It’s also highly dubious to say that it’s an “incomplete job of presenting all the information.” It wasn’t incomplete, it was just plain wrong. And it’s not just me (and a few other bloggers I’ve noticed) who have pointed this out, his own readers expressed disappointment (in commentary on his own blog) that it was so. Instead of setting the record straight, he’s claiming it’s incomplete, that readers missed his intention to do it in more detail, and that he’ll revisit it again later when time permits.

    Only then, apparently, will the post be restored. If he actually does so, it’ll be fascinating to compare what it looks like when it’s restored to what it looked like before it was removed.

    The real kicker is this: I got up this morning intending to comment here that it’s *still* not too late for Watts to do the right thing. But when I captured the present text on his website, I noticed there was an additional comment (from our own reader Lee, who first reported the post’s disappearence here), with a response from Watts:

    Lee (19:21:51) :

    Imran,

    it seems Anthony has now picked up that the reason for the ’skewed’ histograms is the differing baseline periods - so he removed the analysis rather than post a correction. However, thera re stil approving pointers from Anthony her ein this thread, pointing to others who have done the same thing, but using a different graphic - same error, being propagated here.

    Atmoz has done this properly, with the same baselines - and the histograms fall damn near on top of each other.

    Anthony, you made a foolish mistake. Own up to it, post a correction, and move on.

    REPLY: You are right I did make a mistake, and that mistake was in not getting part 2 finished and posted before I left on my trip. Part 2 would cover the GISS data adjusted to the same baseline, compared to all other with histograms, and part 3 would compare all the results and ask some new questions. Since some folks seemed to miss that part 2 and 3 were to cover those issues but focused only on what was up in front of them, I decided to put it back on the shelf until I could return home and finish the job. If you don’t like that method, I’m sorry, but that is the method I’ve chosen.

    The real question here though is: why does GISS uses a different baseline period than the other 3 metrics?

    Watts’ gall is amazing! Even after being hit over the head with his error by his own readers and his own allies (Joe D’Aleo and Atmoz), after being handed (on a silver platter, by Lee) the opportunity to do the right thing and MOVE ON, he *still* maintains that his “mistake” was “not getting part 2 finished and posted…” And he takes the opportunity to throw a (dud) grenade at GISS by suggesting that the GISS choice of baseline period is somehow questionable. It doesn’t make a damn bit of difference!

    This post shows that Watts made a bonehead mistake. Several defenders state quite quite correctly that everybody makes mistakes, and it has no moral implications. But the way he has handled it *does*, and has told us a lot more than the analysis of his erroneous post. All he had to do was say, “Oops — my bad!” and we could have gained respect for his character.

    His post wasn’t put “on the shelf.” Try “under the rug.”

  • John Lederer // March 4, 2008 at 6:01 pm

    I have no dog in this fight — I am an interested sideline observer of the global warming debate whose scientific training, such as it is, was forty years ago.

    But I do feel forced to make a comment.

    The predilection to turn debates about facts into ad hominem attacks is repulsive. It impedes advancing accepted knowledge just by throwing dust around, but it does more damage in a very human way.

    People make mistakes. Scientists make mistakes. Sometimes lollapaloozas of mistakes.

    If those mistakes are pointed out courteously, it is easy to correct them and move on. Everyone benefits.

    However, when egos become involved, and they do when errors are presented as “gotcha, you SOB”, each error becomes a matter of great controversy, much obfuscation, and the whole process grinds to a snail’s pace.

    I have to think that in some degree the vituperation heaped on Watts for what appears to be an “oops” is because in the past he has done a lot of pointing out of fundamental mistakes. As a result of his establishment of surfacestations.org the credibility I give the instrumental record has dropped . I think that would be the natural result for any disinterested observer. This thread has a bit of the taste of revenge about it.

    I think we would all benefit by seeing if we could mitigate “scientific heating”. Most mistakes are because we people are fallible, not because people seek to deceive.

    All sides are responsible. All sides would benefit from courtesy.

    [Response: Let's not ignore which side it was that turned what should be a scientific debate into a war against climate scientists. It's often said, even by well-meaning people (like yourself, I assume), that *we* should be blamed for firing back when it's the denialists who wage war against us. If you want to chastise, you should at least *start* with the ones who fired the opening salvo(s).

    Let's not ignore the fact that Anthony Watts purports to be part of a process to audit the surface temperature record, but clearly he doesn't even understand the meaning of temperature anomaly. If you can't run with the big dogs, stay on the porch.

    I submit that your doubts about the validity of the surface temperature record are based far more on propaganda than on any valid analysis from Watts.

    Let's not forget that Watts' response to a clear exposition of his *factual* error is to pretend it isn't, try to hide the post, and imply that the GISS baseline choice is questionable. I meant it when I said that if he had simply admitted "Oops -- my bad!" then I, and I'd guess most readers here and elsewhere, would have gained respect for his character.]

  • John Lederer // March 4, 2008 at 6:07 pm

    Looking at the “combined” graph you prepared suggests that it might be a valuable pointer.

    There are marked deviations between the satellite and instrumental records in a few places (e.g. 1990, 1995). Looking at the particulars for those places might reveal something.

    Second, one can see whwre the claim of “no global warming since 2002″ comes from. I appreciate teh ideas of noise, of the distinction between weather and climate, etc. Still, there seems to be a good possibility of something fundamental going on.

    [Response: As I pointed out in this post, episodes of stagnation such as we're witnessing now are NOT evidence that "something fundamental is going on." They're exactly the kind of fluctuation that is not just expected, but *inevitable.* If there were NOT such episodes (given the nature of the signal and noise), THEN I would suspect that either something fundamental is going on, or that the data were seriously in error.]

  • P. Lewis // March 4, 2008 at 6:07 pm

    Even Raven started pointing Watts in the right direction early on!

    His “trip” was to the Heartland bash (where, unlike Joe d’Aleo, he is not a contributor to the “Not IPCC” report). He was down to speak in Track 2, Climatology, in the Julliard Complex, 5th Floor, on the afternoon of the 3rd March.

    Says it all.

    Present and future credibility rating? Near zero (IMHO)!

  • Lee // March 4, 2008 at 6:14 pm

    I just posted this at his blog - I want to document it, as it is still awaiting moderation there, and one earlier comment of mine did not get through. It’s a bit blunt - but his actions, IMO, deserve some blunt language in response.

    —–
    Anthony:
    “The real question here though is: why does GISS uses a different baseline period than the other 3 metrics?”

    No, not only isn’t that the “real” question, it isn’t a relevant question at all. The choice of baseline period makes NO DIFFERENCE AT ALL to the climate analyses.

    I do some carpentry form time to time. If I use a tape measure to mark a board at 22 1/2 inches, I might hook at the end of the tape over the end of the board and mark at the ‘22 1/2′ on the tape. I might be concerned whether the tape hook is bent, and start at the “1″ and mark at ‘23 1/2′. Or I might start at the “10″ on the tape so I can simply subtract a ‘tens’ unit, and measure to “32 1/2.” In practice, I sometimes do any of these three things, whichever is easier at the time, and in each case, I get a board 22 1/2 inches long. The choice of zero point MAKES NO DIFFERENCE to the measurement.

    GISS uses a different zero point that other analyses. It makes no difference to the climate analysis. None at all. Zip, nada, zero, zilch. Your histogram strategy is simply irrelevant - all it shows is that when you use a different zero pint, different years are above and below the zero point. Choice of zero point MAKES NO DIFFERENCE TO THE CLIMATE ANALYSES. Further explanation, as you propose to do, will not change that.

    Your silly mistake, apparently, is not understanding what a temperature anomaly is. You said flat out that the different histograms show that there are problems with the GISS data set - it does not. Period. Your analysis is simply irrelevant to the claim you were making. The proper thing to do in that case is to withdraw the claim.

    Further, ATMOZ showed that when the curves are put on the same baseline, the same zero point, the histograms fall right smack on top of each other. He linked to that analysis from your blog. You commented on his analysis, at his blog. YOU SAW AND COMMENTED ON HIS ANALYSIS showing that when put on the same zero point, the curves fall smack on top of each other. You could simply link to that and say, oops, I was wrong - the analysis is already done for you.

    Why are you so insistent on hiding and failing to correct your error? It does not reflect well on you.

  • George // March 4, 2008 at 6:30 pm

    “Watts under the rug?”

    Catchy.

  • Hank Roberts // March 4, 2008 at 8:02 pm

    Climatoscopists. It’s a point of view.

  • caerbannog // March 4, 2008 at 8:07 pm


    Second, one can see whwre the claim of “no global warming since 2002″ comes from. I appreciate teh ideas of noise, of the distinction between weather and climate, etc. Still, there seems to be a good possibility of something fundamental going on.

    The “something fundamental” is the fact that the oceans contain far more heat energy than the atmosphere does.

    A La Nina may help cool the *atmosphere*, but it does not change overall amount of heat energy in the Earth’s environment (which includes the oceans). A La Nina cools the atmosphere by causing the ocean to absorb heat energy from the atmosphere. That heat isn’t “going away” by any means.

  • Gavin's Pussycat // March 4, 2008 at 8:13 pm

    If you want to chastise, you should at least *start* with
    the ones who fired the opening salvo(s).

    …and never stopped firing. Who are the ones that will lie anytime anywhere if they believe they won’t be caught out? Like when the audience is a John Lederer, rather than you or me?

    Leaning over backwards again HB. Doubleplusungood.

    [Response: At times I've considered it a mistake to go too far giving the benefit of the doubt. But I don't regret those mistakes.

    I like the psuedonym.]

  • John Lederer // March 4, 2008 at 8:15 pm

    “As I pointed out in this post, episodes of stagnation such as we’re witnessing now are NOT evidence that “something fundamental is going on.” ”

    That is precisely true — in both directions.
    ————————-
    I do apologize if my sermonette seemed aimed at the original post here in particular. It was triggered by the comments on this thread, but it could equally be aimed at threads I have seen on the other side.

    Regardless of its origins, the aspersions seem a spiral of increasing intensity with a “tit-for-tat” motivation both ways.

    It really is becoming intrusive. I do not think it serves the scientific debate or the ethos of anyone.

    Perhaps if we gave grants for good humor and courtesy, science would be well served

    [Response: Alas, I suspect that good humor and courtesy will always have to be their own reward.]

  • JCH // March 4, 2008 at 8:45 pm

    “A La Nina may help cool the *atmosphere*, but it does not change overall amount of heat energy in the Earth’s environment (which includes the oceans). A La Nina cools the atmosphere by causing the ocean to absorb heat energy from the atmosphere. That heat isn’t “going away” by any means. …”

    This is the way it seems to come out when I read about it. Nice to see a confirmation.

  • Ian Forrester // March 4, 2008 at 8:57 pm

    John Lederer said: “All sides are responsible.”

    No John, one side has integrity, honesty and scientific ability, the other side has none of these. Do you know which side is which?

  • mmghosh // March 4, 2008 at 10:20 pm

    Mr Lederer

    There are mistakes and mistakes.

    I think that before taking on scientists who have spent lifetimes working on data in difficult fields, it is necessary to spend at least some time in the particular discipline acquiring the understanding of the nature of publication.

    As the writer of this blog has suggested, there is well known, tried and tested method to get one’s observations published. The first step is to discuss one’s observations in departmental reviews, where they get assessed and criticised by one’s fellows and departmental staff.

    The next step is a presentation at a local or regional meeting, where a similar process of criticism follows.

    The next step is sending the observation to an accredited journal, where the error would probably have been picked up by an alert reviewer, normally there are 2 or 3. Finally, the article, when published, can attract the attention of the readership of the journal who may criticise it in letters to the Editor.

    This lengthy and rigorous process has been evolved through centuries of scientific work. The reason is that we all make mistakes, and everyone is anxious not to publish work that is incorrect - it affects future possibilities of getting grants and attracting students. This process ensures maximum exposure of one’s work to the largest possible community of one’s peer before entering the scientific literature.

    Even after this process, mistakes do creep in, and there are investigations done by journals and article deletions after publications, or retractions published by authors.

    But, as in this case, this process generally prevents in propagation of mistakes.

    Occasionally, a researcher makes a brilliant original observation, or a paradigm-shifting discovery when some of these steps might be bypassed. In this situation, Mr Watts does not, perhaps, make a convincing case for either the brilliance or the originality, or in the last resort, the correctness of his analysis.

  • John Mashey // March 4, 2008 at 10:34 pm

    re: ocean heat content
    blogger formerly known as tamino covered this jan 18 2007.
    However, at that point,
    there was believed to be a few years of cooling.
    That turned out to be spurious, as was discussed on RealClimate 18 April 2007 as Ocean Cooling - Not.

    This heat balance thing is in some sense way more important than surface oscillations, since first law of thermodynamics applies to whole Earth.

    Maybe HB will revisit, but in any case, people should go back and study the Hansen, Nazarenko st al 2005 paper. Good stuff.

  • henry // March 5, 2008 at 12:11 am

    To paraphrase George:

    “The whole idea that people with no expertise in climate science specifically, or any relevant field (as in the case of Al Gore) think they know more about this stuff than the scientists who have made it their life’s work to study it is just absurd.”

  • TCO // March 5, 2008 at 12:15 am

    I think there is a common human problem of failure to admit fault. In addition, there is a common problem of disaggregation of issues. Disaggregate, disaggregate, disaggregate.

    For instance:

    Issue 1: Was Watts analysis wrong?

    Issue 2: Is Watts a bad analyst?

    It’s POSSIBLE that the answer to both questions is YES. But they are separate issues and it’s POSSIBLE that the answer to the first is yes and the second no. In any case, we should DISAGGREGATE the questions. So that we can examine and understand each separately. Also, so that men of good faith can agree to those points that are not contested…thereby moving the debate to questions that are in debate. So, one could imagine Watts agreeing to point one, but debating point two.

    This is not just a Watts thing. It’s a common problem. I struggle with it myself. But what we as scientists need to do is concede ANY POINT where we are in agreement, REGARDLESS of the danger that an “opponent” will exploit the admission. This is the way of the scientist. It is not the way of the lawyer.

  • henry // March 5, 2008 at 12:26 am

    Tamino:

    From your post:

    “The right way to compare these data sets is to use a common reference period for all of them. I did that here for the three major surface temperature records. For those who want to see what it looks like using the four sets chosen by Watts, here it is (click for a larger, clearer view)::

    My question is, for the graph you made, since choice of reference made no difference to trend, why use GISS? Why use the oldest, coldest period selected from all four original charts?

    WMO suggests two methods: end of the current decade (2000) or use of another period for data handling purposes (GISS’s 51-80).

    The question can be worded two ways:

    1) Why does GISS use the oldest, or

    2) Why does everbody else use newer periods?

    [Response: For the graph in this post, all four data sets are set to the reference 1979 to Jan. 2008 (the entire time span of the satellite data). I don't know about HadCRU. The satellite data sets use a later period because the data don't start until 1979, and to use a given base period you have to have data during that period. So they can't use 1951-1980 or 1961-1990; we just don't know what the numbers were for 1950 through 1978.

    GISS uses 1951 to 1980 because that's what they've always done. It really makes no difference, but one has to be aware that it's not the absolute value that matters, it's the changes. Altering the reference period simply moves the entire graph up or down, but the changes from one moment to another are unaffected.]

  • John Lederer // March 5, 2008 at 12:45 am

    “This lengthy and rigorous process has been evolved through centuries of scientific work. ”

    Do you think the taradition academic publication process is still that most suitable in an Internet age?

    I think there still may be merit in final, formal “published” work– though publishing will not be in print.

    But it seems to me that your early steps, particularly the departmental and informal review might better be accomplished by a blog type posting.

    This would both allow the researcher remote in place and perhaps time to obtain its benefits and allow many more to contribute.

    In my scanty knowledge of scientific history I am struck by how many of the “great” discoveries were made by scientists in closely related but distinct fields, or by dedicated, knowledgeable, amateurs.

    Might it be possible that by broadening the mix of people able to criticize and expand, the end result would be improved?

  • Douglas Watts // March 5, 2008 at 12:54 am

    “There is nothing wrong even with obdurate persistence in looking for evidence against an hypothesis when all the evidence seems to be for it. It is not a mark of moral weakness. It is an essential element in the public debate, from which the truth will eventually emerge.”

    Well , no. None of that is true. Not when someone’s obdurate persistence actually delays and prevents action on a phenomena that will cause real harm to real people.

    The Precautionary Principle was devised precisely as the antonym of the above quote, particularly in the matter of endangered species, where federal agencies study forever if they are actually endangered, and take no action to save them, until they actually are extinct.

  • EliRabett // March 5, 2008 at 12:57 am

    TCO in principle the first question is a matter of fact (or interpretation) while the second is a matter of experience.

  • papertiger // March 5, 2008 at 1:08 am

    So after all is said we are left with a 0.2 degree rise on average, since 1950.

    Un huh.

  • TCO // March 5, 2008 at 1:15 am

    Agreed.

  • papertiger // March 5, 2008 at 1:17 am

    Glad I came over here. A point 2 degree temp shift is startling.
    I hope that’s in Celsius, because I would hate to think all those homes were torched in Seattle because of 0.2 degrees Fahrenheit.

  • Hansen's Bulldog // March 5, 2008 at 1:22 am

    papertiger:

    on another thread someone basically said that Steve McIntyre was completely ignorant of statistical analysis. Clearly I’m no friend of McIntyre, but I commented that I disagree with claims that he’s statistically clueless.

    You, on the other hand, ARE statistically clueless.

  • Hank Roberts // March 5, 2008 at 1:34 am

    > clueless

    Busy, though.

    Results … about 2,030 English pages for +papertiger +climate

  • papertiger // March 5, 2008 at 1:37 am

    Hansen’s bulldog:

    You might be forgiven for fixating on a tree, you are a dog after all.
    I’m looking at the forest. That graph there (an amalgam of the four main world thermometers - which I suppose isn’t controversial) says there has been a 0.2 Celsius temperature rise since 1950. Let’s see, 0.2 C equals 0.26 F.

    I take that back, bulldog. I am appalled that homes in Seattle were torched over this trivia no matter which temperature scale is used.

    [Response: As I said in the post, for that graph all four series were shifted to a common reference period. As I said in a response to an earlier comment, for that graph the chosen reference period is 1979 through Jan. 2008. So that graph indicates present temperatures are around 0.2 deg.C above the 1979-2008 average.]

  • TCO // March 5, 2008 at 1:40 am

    I am the uber troll.

  • P. Lewis // March 5, 2008 at 2:43 am

    papertiger said

    Let’s see, 0.2 C equals 0.26 F.

    Ahem! Were you involved in the Mars (crash) lander by any chance?

    Add another 0.1°F.

  • chriscolose // March 5, 2008 at 2:56 am

    I have not read through all the posts here or there, so I’m not sure if this has been discussed much but the new thing over there seems to be why GISS uses a 1951-80 baseline when the others use a different one. Just another part of the GISS conspiracy. I gave a quick reason, but the most likey one is that Hansen and co. are all being funded by the eco-nazis to manipulate the surface stations to get big time warming and so to carbon tax everyone and kill the economy so that the eco-nazis can rule the world

    [Response: Sigh... If ever there was a non-issue, the choice of baseline period is it. It really makes no difference at all.]

  • caerbannog // March 5, 2008 at 4:05 am


    I gave a quick reason, but the most likey one is that Hansen and co. are all being funded by the eco-nazis to manipulate the surface stations to get big time warming and so to carbon tax everyone and kill the economy so that the eco-nazis can rule the world

    Creationism and global-warming denial: two cases where parodies are often indistinguishable from the genuine article.

  • jl // March 5, 2008 at 4:16 am

    The back and forth bickering is enough to make me cry. As far As I can tell changing base periods is easy. At least compared to the other kind of adjustments used in setting up Gridded Temperature Estimates.
    I made a couple of charts I thought you might want to look at
    stations
    I used subtraction not a trend, but for a beginner what do you think???

    [Response: You're on safe ground subtracting 1975 from 2005; changing baseline won't affect that at all. Subtraction gives results that are noisier than (and probably not as meaningful as) trend, but that doesn't mean it's not a useful statistic (and it sure is easy to calculate). Are you using raw or adjusted data?

    I'm not sure what "t" is in your graphs -- is it tenths of a degree? For comparing one set of stations against another, I'd recommend making all the x-axes the same scale. Also, some summary statistics (mean, standard deviation, and total number) would help. You could also experiment with using a narrower interval width in your histograms; it'll reduce the precision of any single interval, but increase the detail over the entire range.

    You must have put a lot of work into this; there's a great deal of data to process! If you get more results, please keep us posted.]

  • John Cross // March 5, 2008 at 4:41 am

    P. Lewis: I came across papertiger over at skeptical science where we were discussing the sun being the cause of recent warming because you know that Jupiter is warming. I was having a hard time understanding his point until I realized that he couldn’t convert w/cm^2 to w/m^2. He even posted over on Lubos’s site trying to get help. It was fairly amusing, but I think that Tamino is right except I will go further and say mathematically clueless.

    John

  • jl // March 5, 2008 at 5:24 am

    thanks for the reply
    T is the name of the data stream though it is in tenths of degrees. as to raw or adjusted, whatever v2.mean from the NCDC is I think raw definetly
    before any corrections by Hansens GISS code
    actually I used the output from STEP1.
    in total I found 1885 stations with a mean of 7.31 thenths of a degree, With a standard deviation of 8 according to open office.
    The rural stations had much less only 169 with a mean of 9.04 with a standard deviation of 7.87.
    The greater that 50 had 385 stations and a mean of 6.6 and a standard deviation of 7.83.
    I’ll see if I can upload the raw files I used.
    thanks for the advise. After spending many hours looking a temperature data I’m under the impression that the biggest reason that all of the urban stations haven’t been censored is because there are very few rural stations with long histories, though after reading your blog I think there is a need for a temperature estimate made only with “good stations” even if it only go’s back to 1950.
    once again thankyou jacob L

  • mmghosh // March 5, 2008 at 5:40 am

    Mr Lederer

    You would be surprised by how many “great discoveries” are made after years of basic observational work.

    Why do you think that participation of laypersons in research programs would enhance the programs? Are you suggesting that numbers necessarily enhance content? This is a very curious statement.

    Researchers have been collaborating on usenet groups, bulletin boards and listservs well before the worldwideweb. In fact, the worldwideweb was developed primarily for researchers to collaborate at CERN. Every scientist in the world today communicates constantly with his/her colleagues via email.

    There is a place for anyone to participate in science. Its called graduate school. And it involves hard work - study, taking exams and often giving up opportunities in business and industry.

    While it is true that one can work on one’s own, it obviously requires extra diligence and training. You also have the possibility of making mistakes if you are working on your own outside the academic establishment - mistakes which have been identified and corrected in the past.

    Why do you not trust standard scientific procedure in this case - non-scientists over scientists? It is the same procedure as research in pharmaceuticals, medicine, physics, chemistry etc.

  • dhogaza // March 5, 2008 at 5:56 am

    I am appalled that homes in Seattle were torched over this trivia no matter which temperature scale is used.

    I see one of the denialist memes here: the science must be wrong because someone torches some homes.

    As scientifically valid as “climate science must be wrong because some of them are mean”.

    Or the upcoming creationist movie “Expelled”.

  • fred // March 5, 2008 at 7:44 am

    “Not when someone’s obdurate persistence actually delays and prevents action on a phenomena that will cause real harm to real people.”

    Some of us cannot get it through our heads that

    1) what convinces us does not convince others

    2) the others have a perfect right to speak their minds wherever they want, on the net, in the press, on TV, in journals, in Union Square for that matter.

    It is from the combination of everyone speaking their minds that we will get to the truth.

    Much of the indignation here about scepticism on some aspects of AGW is basically a cry for censorship in the name of the common good. We know one thing about censorship: it never delivers the common good. The right place for Watts to publish (and McIntyre too) is whever they feel like publishing. It will all come out in the wash, whether they are right or wrong, and the more widely they and everyone else publishes, the quicker it will come out. Stop getting so indignant they are saying what they think, if you really don’t like people saying what they think there are plenty of places they can’t. One doubts you’d like living in any of them.

  • Heretic // March 5, 2008 at 8:28 am

    Dunno if this will be allowed by Tamino, but it speaks volume on our feline friend’s abilities in numbers. Seems papertiger hasn’t worked on his maths very much since this thread:
    http://www.skepticalscience.com/global-warming-on-jupiter.htm

    [Response: I think the point's been made. No more on this please, it's piling on.]

  • Dano // March 5, 2008 at 10:03 am

    Hansen and co. are all being funded by the eco-nazis… carbon tax everyone … kill the economy … eco-nazis can rule the world…

    Creationism and global-warming denial: two cases where parodies are often indistinguishable from the genuine article.

    I often wonder whether comedy writers use The Internets to practice their parody routines.

    Nonetheless, I like the comment, as the only thing denialists have is conspiracy theory. They have no climate theory to test, so why not resort to conspiracy?

    Best,

    D

  • guthrie // March 5, 2008 at 10:13 am

    John Lederer-
    Academic publishing is not ideal in the internet age, but the problem is, who will pay for the system? even with online journals, you have to run a staff doing the various publishing things you have to do, not to mention sending stuff out for review etc. Just now the subscribers to the journals pay for all that. I personally would like all academic journals to be freely available, however that would require subsidy from central gvt or suchlike.

    As for scientific history, you have to be aware that the modern scientific setup has evolved in step with increases in knowledge, increased specialisation and capital required. No amateur or outsider is likelyto make a breakthrough in particle physics, for example, because you need supercomputers and synchrotrons to do the experiments and calculations, as well as quite a number of years learning the basics in order to be able to actually do the work required in the first place. Same goes for chemistry, biology, etc etc. There have been one or two non-academic contributors to the climate change science, but not that many at all. I cnanot recall all the details, but there was one guy who was able to make a contribution precisely because he spent several years educating himself up to the necessary standard, and had the money from his business to be able to then carry out some experiments.

    Needless to say, this is not what is available to the average person.
    The problem with broadening the mix of people who can comment is that it makes the good stuff harder to find amongst all the sewage. You can see this online every day.

    [Response: I agree that everyone publishing everything makes it harder to find the good stuff. But I also agree with earlier commentary that everyone has the right to make his opinions and analysis public. I guess the peer-reviewed literature is the place to look for very significant hard science, the net (or elsewhere) for other stuff.

    That helps communication among scientists. But the peer-reviewed literature doesn't help the lay reader because it's usually too technical.]

  • CuriousTom // March 5, 2008 at 12:29 pm

    Since I’m not a professional scientist, even though I have degrees in science, I have some very basic broad questions to pose. I’ve read hour upon hours of comments, stories, and abstracts but I’ve never seen these broad questions specifically addressed(outside of a textbook back in college which I can’t recall the precision of it’s basic answers right now) and thought since this is the most current article I’d ask them here.

    1) Am I correct in assuming that there are primarily 5 factors in driving weather and climate on earth?

    Here’s the factors as clearly as I can tell. a)solar activity, b)earth’s atmospheric makeup, C)earth’s rotation, gravitational pull of the moon and other celestial bodies, d)the physical properties of earth’s surface, ie. snow and ice, water, land, forest, prairie, exposed rock, etc., and finally earth’s own internal heat.

    One possible thing that I’m not sure about is the impact of other extraterrestrial events or influences. Like how being exposed to much higher cosmic radiation levels might or might not affect the climate, etc.

    Am I correct or am I missing something or including something not important? I’m just talking in broad generalizations and not being specific, like I understand these things interact and it all quickly becomes much more complex.

    Also, it’s easy to look up CO2 atmospheric levels over time, either from modern actual measurements or from ice core and sea floor coring data etc. It’s also relatively easy to find estimates of human CO2 emissions per year, but I constantly see bandied about claims about human input to the modern measured levels of CO2. The numbers I often see range from 1/30 of one percent of the total atmospheric CO2 to as high as 2 to 5 percent of total atmospheric CO2 being directly attributable to human activities. It seems from other papers indirectly that 100% of the increase in CO2 levels over the past 50 years or longer are directly attributable to human emissions. I have no idea of knowing if any of these are even remotely accurate from what I’ve read. I’m sure it’s just a matter of me not reading enough to find the proper material. Is there a paper with a published and widely scientifically accepted estimate of the percentage of current atmospheric CO2 directly attributed to human emissions? I can’t seem to find a reference for one.

    I don’t want to be presumptuous, but please don’t confuse my simple questions for a lack of understanding of the details and the complexity of the issues being discussed.

  • Imran // March 5, 2008 at 1:22 pm

    Its hardly ‘cherry-picking’ when I’m referring to something from the front page of Hansens own GISS website :
    ‘http://data.giss.nasa.gov/gistemp/graphs/Fig.C.lrg.gif

    All I’m trying to say is that in order to avoid getting marginalised from whta is increasingly becoming mainstream public opinion you have to admit that you might be wrong.

  • Barton Paul Levenson // March 5, 2008 at 1:43 pm

    henry writes:

    [[To paraphrase George:

    “The whole idea that people with no expertise in climate science specifically, or any relevant field (as in the case of Al Gore) think they know more about this stuff than the scientists who have made it their life’s work to study it is just absurd.”]]

    You do know that Gore was Revelle’s student, right?

    In any case, I’m a little tired of the denialists assuming that if they can find a way to knock Al Gore, global warming falls apart.

    Global warming theory doesn’t depend on Al Gore. It was originated by Svante Arrhenius in 1896.

    Global warming doesn’t depend on Al Gore. It started around 1750 and would have preceded the same way if Al Gore had never been born — except that, partially through his efforts, we may soon begin to mitigate it.

    There’s also the question of the ad hominem fallacy. Proving that Al Gore is a bad man doesn’t prove that what he says is wrong. If Hitler said that the French Revolution was in 1789, he’d be right, even though he’s Hitler. If it turns out that Al Gore tortures small children in his basement, what he says about global warming A) happening, B) being due to technology, and C) constituting a big, serious problem, will still be true.

  • J // March 5, 2008 at 2:39 pm

    Curious Tom wrote:

    I constantly see bandied about claims about human input to the modern measured levels of CO2. The numbers I often see range from 1/30 of one percent of the total atmospheric CO2 to as high as 2 to 5 percent of total atmospheric CO2 being directly attributable to human activities. It seems from other papers indirectly that 100% of the increase in CO2 levels over the past 50 years or longer are directly attributable to human emissions.

    As of last month, atmospheric CO2 concentration as measured at Mauna Loa was 386 parts per million (by volume). When monitoring started in 1958, it was 316 ppmv, and the generally accepted “preindustrial” concentration was 280 ppmv.

    We know that this increase is due to human activities for several reasons. There are some subtle arguments involving isotopes, which I won’t go into here, but the main point is that we have a pretty good idea of how much we’re emitting, and it’s an amount that’s actually larger than the atmospheric increase. Thus, the oceans are soaking up part of the CO2 from fossil fuels (something that is confirmed by very painstaking measurements of increasing CO2 in the world’s oceans over the past couple of decades, by e.g. Takahashi et al.)

    If you ignore the fact that the oceans are actually soaking up part of our CO2, and only consider what’s remaining in the atmosphere, human activities (fossil fuels + land use) have increased the CO2 concentration by 38% above the preindustrial level.

    This is a very big change. From the Vostok and Epica ice cores, we can be fairly confident that CO2 never exceeded 300 ppmv at any time in at least the past half-million years. We are well on our way to an atmosphere last seen during the Oligocene, approx. 20 million years ago. FYI, the Earth looked rather different at the time.

    ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_mm_mlo.txt

  • J // March 5, 2008 at 2:58 pm

    Imran wrote:

    And any bonehead can pull out Hansens monthly data since 1998 and its very very hard to see any warming trend there.

    Sure, but why listen to a bonehead’s opinion, when it’s easy to ascertain the facts?

    Using your own stated source (GISSTEMP monthly data), since Jan. 1998, the global land+ocean temp index has increased by 0.15 degrees C/decade, plus or minus 0.08 degrees C/decade.

    So even starting with the exceptionally strong El Nino year of 1998, there’s a positive and statistically significant warming trend.

  • George // March 5, 2008 at 5:11 pm

    If it turns out that Al Gore tortures small children in his basement…

    He does?

    I can’t believe it….Well, maybe I can. It’s Al Gore, after all.

    Word has it he has also used the internet tubes in ways that I can not repeat here — that and boring prisoners to near death with his Global Warming slideshow.

    I hear the latter was what finally made Khalid Sheikh Mohammed (the 9/11 mastermind) crack — even though most people mistakenly believe it was water-boarding.

    [The CA Rumor Weed has gone wild in these parts. You'd better be careful, Paul]

  • John Mashey // March 5, 2008 at 5:30 pm

    Several years ago, I attended a talk, along with another 30 people, at our local community center.

    Al Gore could have used the first half of the slides, and his talk wouldn’t really have changed, unsurprisingly, since many slides in both cases were taken from standard scientific publications. I didn’t hear much I didn’t already know about climate, but it’s always nice to see how it’s presented. The speaker even talked about denialists some, with obvious disgust, perhaps because he knew many of the early ones personally.

    Who was speaking?
    Burton Richter, Nobel Physics winner and very senior scientist:
    http://www-group.slac.stanford.edu/do/people/richter.html
    Here’s the full talk Burton gave at Stanford at an Applied Physics Colloquium. He used a subset for the more general audience, i.e., he did all the climate slides and some of the energy slides. Our town is peculiar [50% of adults have advanced degrees, 10% PhDs] so he didn’t water it down much.
    http://www-group.slac.stanford.edu/do/brichter/presentations/2004_10_05.htm

    Now, sometimes, even Nobel physics/chemistry winners go off the rails towards the ends of their careers, but usually they don’t, based on the limited sample of 3 Nobel winners I’ve met, and Richter is still sharp. Now, he is not a climate scientist, but a very broad, very senior physicist, and Stanford is not only one of the world’s top universities, but it is exceptionally strong in energy&environment research.

    I’d guess Richter talks to the same climate scientists Gore does, which is why they tell the same story.

  • Lee // March 5, 2008 at 6:03 pm

    So, Anthony allowed and answered my post of yesterday. Apparently, he is closing comments on what is left of that thread. And he is not withdrawing his claim that his analysis shows problems with the GISS data set. He is going to do more work - he wants to ‘learn’ about this stuff.
    He doesnt have permalinks to comments - it is in here.
    http://wattsupwiththat.wordpress.com/2008/02/27/a-look-at-temperature-anomalies-for-all-4-global-metrics/

    Here is his answer:

    REPLY: No “hiding”, it is quite simple. Try 4 days our of the office, 16 hours of traveling yesterday and a conference presentation. I think perhaps you are not reading the main page or the replies, since you seem unaware. This will be covered in parts 2 and three, as I originally stated. If you are unhappy with the schedule of updates thus far, there is little that I can do for you.

    However since I don’t wish to continue in an obviously cicrular argument right now, but rather want to focus on getting caught up, your post is a clear reason for me to close comments until I get caught up and can finish parts 2 and 3. “You could simply link to that and say, oops, I was wrong - the analysis is already done for you.” Yes I could, but I’m going to do it myself, because I want to see and learn for myself. If I wanted to rely on what others say, there would be no need to learn anything.

  • J // March 5, 2008 at 6:59 pm

    I submitted a comment over there, before Mr Watts announced he was shutting down the comments.

    It hasn’t appeared — maybe he’s still moderating the last few in the queue, or maybe he decided he didn’t want to post it for some reason. His business.

    Speaking of Watts’s blog … a while back, in the open thread, fred linked to a post by Roy Spencer over there:

    http://wattsupwiththat.wordpress.com/2008/01/28/spencer-pt2-more-co2-peculiarities-the-c13c12-isotope-ratio/#more-619

    is Spencer’s post. Be interested to have a reaction to it. Is it just nuts?

    Our esteemed host aka Tamino aka HB replied:

    [Response: It’s fine *until the final step*, which is just nuts.

    Perhaps I’ll post on the topic soon.]

    Are you still thinking of following up on that?

  • steven mosher // March 5, 2008 at 7:16 pm

    Lee. Now you understand the appeal of auditing.
    Now you understand why we have continually asked Dr. Mann to correct the simple geographical errors he has made over and over again.

    http://www.climateaudit.org/?p=2406

  • Lee // March 5, 2008 at 7:48 pm

    Mosher, did you catch the part in that CA thread where McIntyre admits:
    “#45. The location errors don’t make a speck of difference to the final NH reconstruction,”

    Anthony, on the other hand, has made a clear charge - that his analysis casts doubt on the GISS analysis. However, his analysis isn’t even wrong - it is irrelevant to that question. Yet, he wont retract it, or even admit there is any problem.

    Auditing as Stevie Mac practices it - nitpicking every niggling issue even if it “don’t make a speck of difference” has no appeal. And when I see Stevie Mac editing what he has said there without documentation, and even editing MY posts there without documentation, I find myself even less impressed by it.

  • George // March 5, 2008 at 8:16 pm

    Steven Mosher:

    Now you understand the appeal of auditing.
    Now you understand why we have continually asked Dr. Mann to correct the simple geographical errors he has made over and over again.

    I thought the appeal of “auditing” was so that you could charge people with fraud with only the slimmest bit of “evidence” and then disclaim any and all responsibility for doing so.

    You know, equate their scientific work with the “Piltdown Man” fraud based on the existence of ’simple geographical errors’.

    You are right. Auditing is great — and very appealing indeed!

  • Hank Roberts // March 5, 2008 at 8:22 pm

    > I’m going to do it myself, because I
    > want to see and learn for myself. If I
    > wanted to rely on what others say,
    > there would be no need to learn …

    Lee, now you understand why scientists replicate — redo research, collect data, build models — rather than auditing one others’ work. Science comes not from turning out other people’s pockets but from investigating interesting questions.

    If he addresses the question independently, does his own work and publishes (or even blogs) and comes to different conclusions, then compare the work and look for the difference.

    Point being — four climate time series when compared come to the same conclusion within the known error range, and the differences are attributable to known sources of error.

    If someone were to, say, create a time series of temperature by, oh, imagine getting all the oil companies’ drilling logs and accumulating the borehold temperature measurements at various depths in different places — that would be an independent source of climate change information. There’s good science done on borehole temperature work to cite.

    Someone in, say, the mining or petrochemical industry could do that.

    Likely the petroleum companies did it already, in their proprietary climate models. But someone could do it in the public view and be contributing.

    OR of course you can just copy and paste.

  • Svante's Poodle // March 5, 2008 at 9:57 pm

    CuriousTom, you could do a lot worse than start out with the IPCC WG1 report, found at http://www.ipcc.ch. Dig in gradually. Lots of references there.

  • guthrie // March 5, 2008 at 11:14 pm

    Hank- the petroleum companies have proprietary climate models? What do they predict?

  • caerbannog // March 6, 2008 at 12:46 am


    Hank- the petroleum companies have proprietary climate models? What do they predict?

    This might help — copied and pasted from a post of mine in an earlier thread.

    A little googling turned up this: http://tinyurl.com/2pa2r5


    Resource Relation AAPG Bulletin (American Association of Petroleum Geologists) ; Vol/Issue: 74:5; Annual convention and exposition of the American Association of Petroleum Geologists; 3-6 Jun 1990; San Francisco, CA (USA)

    Description/Abstract:

    Numerous investigators have examined the potential use of numeric climate models and paleogeographic reconstructions to predict the deposition and preservation of organic-rich sediments, which may ultimately develop into hydrocarbon source rocks. These studies have concentrated on the Mesozoic and Cenozoic eras.

    Although geologic conditions during these periods were different than that of today, they do have many similarities. In contrast, the boundary conditions associated with the Paleozoic are dramatically different. For example, no significant land plant cover is assumed in pre-Devonian simulations. In addition, for many of the simulations the bulk of the land mass was situated in the southern hemisphere at high latitudes. This compares with the Mesozoic and Cenozoic distributions that exhibit nearly coequal land-sea distributions in the two hemispheres.

    An examination of the results of paleoclimate simulations for time slices in the Paleozoic reveal significant changes in spatial distribution of marine conditions that would favor high levels of organic productivity and organic preservation through time. The authors study of the stratigraphic record, though incomplete, has revealed a favorable correlation between organic-rich black shales, capable of acting as hydrocarbon source rocks, and those regions that had both high preservation efficiencies and elevated levels of organic productivity. These results suggest that numeric climate models can be effectively used to predict source rock distribution throughout the Phanerozoic.

    As you can see, the oil companies have been in the climate-modeling business for quite a long time.

  • Marion Delgado // March 6, 2008 at 2:51 am

    “fred’s” utterly mendacious defense of Anthony Watt’s well-coordinated obfuscation and harrassment program is longer than a typical Anthony Watt AEI reprint.

    I hope that annoys the lurkers.

  • Hank Roberts // March 6, 2008 at 3:35 am

    guthrie, this will begin an overview:
    http://www.google.com/search?q=%22petroleum+exploration%22+%2B%22climate+model%22

  • EliRabett // March 6, 2008 at 4:04 am

    Curious Tom asked an important question:

    ” Am I correct in assuming that there are primarily 5 factors in driving weather and climate on earth?

    Here’s the factors as clearly as I can tell. a)solar activity, b)earth’s atmospheric makeup, C)earth’s rotation, gravitational pull of the moon and other celestial bodies, d)the physical properties of earth’s surface, ie. snow and ice, water, land, forest, prairie, exposed rock, etc., and finally earth’s own internal heat.”

    But he left out a really major factor - the biosphere, plant, animal life and increasingly humans.

  • CuriousTom // March 6, 2008 at 6:00 am

    Thanks for the replies. I was asking to see if most of the informed people here generally agree in that these are the main basic factors which drive the climate on earth.

    EliRabett, I agree that the biosphere has a big impact, but I left that out because the biosphere largely falls into 2 of the factors I already mentioned, being the makeup of the surface of the earth and the makeup of the atmosphere of the earth. Of course the biosphere can change and affect both or are even perhaps the most influential factor in shaping both to a large degree.

  • steven mosher // March 6, 2008 at 1:36 pm

    Lee, of course.

    If you read the thread, I suggested that the error
    could be something rather simple, unrelated to the analysis, and St.Mac responded as you note.

    That’s not the point. I’ll say it again, this has nothing to do with the truth of AGW.

  • Svante's Poodle // March 6, 2008 at 2:09 pm

    Curious Tom: leave off the dot from

    http://www.ipcc.ch

    damn lack of preview :-(

  • Barton Paul Levenson // March 6, 2008 at 3:09 pm

    CuriousTom writes:

    [[The numbers I often see range from 1/30 of one percent of the total atmospheric CO2 to as high as 2 to 5 percent of total atmospheric CO2 being directly attributable to human activities.]]

    Try 27% of the CO2 currently in the air is directly attributable to human activities. The preindustrial level is about 280 ppmv. The current level is 385 ppmv. 100 - 100 (280 / 385) = 27%.

  • Barton Paul Levenson // March 6, 2008 at 3:12 pm

    fred writes:

    [[Much of the indignation here about scepticism on some aspects of AGW is basically a cry for censorship in the name of the common good.]]

    Garbage. Nobody is saying Watts or McIntyre can’t publish their crap on the internet, or in pamphlets, or newspapers. But scientists are not going to pay attention to it until it is published in peer-reviewed science journals. That’s not censorship.

  • Hank Roberts // March 6, 2008 at 4:09 pm

    Fred, there’s extensive work done and published on the weather network. For a century or more. If there were something to publish using M’n'M’s issues someone would do so.

    It’s less than a tempest in a teapot.

  • George // March 6, 2008 at 4:15 pm

    [Much of the indignation here about scepticism on some aspects of AGW is basically a cry for censorship in the name of the common good.]]

    Some people seem to believe that “skepticism” is about believing and saying whatever the hell they wish to believe/say. That’s not skepticism and it is certainly not science.

    Skepticism (in its original scientific sense) is not the same as “contrarianism”, though some seem to believe as much.

    Also, as the Supreme court has ruled time and again, not all speech is protected by the first amendment.

    for example, when there is no fire, you have no right under our Constitution to yell “fire” in a crowded theater just because you feel like it.

    And you have no right under our Constitution to tell lies about “How safe smoking is” when you know damned well that it causes cancer. (particularly for the purpose of misleading the public into thinking its just fine to keep buying and smoking cigarettes) .

    To prevent that kind of rubbish is just good common sense.

  • dhogaza // March 6, 2008 at 4:54 pm

    Hank says …

    It’s less than a tempest in a teapot.

    Mosher makes it clear above

    If you read the thread, I suggested that the error
    could be something rather simple, unrelated to the analysis, and St.Mac responded as you note.

    That’s not the point.

    Exactly. The point is to discredit the GISS surface temp record and the scientists behind it.

  • John Lederer // March 6, 2008 at 5:13 pm

    “Why do you think that participation of laypersons in research programs would enhance the programs? Are you suggesting that numbers necessarily enhance content? This is a very curious statement.”

    Though not precisely a research program, I think you have a very good answer in this thread:

    The surfacestations.org effort to document possible microsite issues in recording temperature records would be an example of an effort by many amateurs to examine the accuracy of the instrumental surface temperature data which provides a key base for much work.

    The effort requires much work over broad geographyt but does not require a high degree of knowledge or skill (map reading, GPS knowledge, and elementary digital photography skills seem key).

    The result — a database of station photographs–does not critically depend on the reliability of the observer.

    However, the result, at least to my mind, raises unsettling questions about the reliability of the surface temperature record.

  • caerbannog // March 6, 2008 at 5:35 pm


    However, the result, at least to my mind, raises unsettling questions about the reliability of the surface temperature record.

    It’s a darned good thing that we have the UAH and RSS satellite data sets with which the surface temperature records can be cross-checked.

  • Lee // March 6, 2008 at 5:47 pm

    John Lederer says:
    “However, the result, at least to my mind, raises unsettling questions about the reliability of the surface temperature record.”

    John, why? What exactly in that set of photographs raises questions about the reliability of the determination of the change over time of the temperature anomaly, after heterogeneity and UHI adjustments, et al, are in the algorithm and being made?

    Please be specific.

  • mmghosh // March 6, 2008 at 6:53 pm

    Mr Lederer,

    Microsite issues have a small effect on the trend of change of the temperature anomaly.

    They are quite important for assessing weather, i.e. absolute values of temperature. This is important for Mr Watts who is a professional weather analyst.

    They are much less important for assessing climate which look at changes in anomaly trends, which depend much less on microsite issues. You can get a reliable anomaly trend even if you use a wrongly calibrated thermometer which gives faulty absolute readings, provided the fault remains constant.

  • Hank Roberts // March 6, 2008 at 7:24 pm

    Lee, it’s all third hand copying and pasting by now. There’s no evidence, just a long trail of people repeating other people’s suspicions that the bad news can’t be true.

  • dhogaza // March 6, 2008 at 7:25 pm

    AFAICT, the basic issue is one of personal incredulity. “I don’t understand how scientists can correct for errors in the data, therefore I the results must be bogus”.

    I’m not trying to pick on John personally, I am speaking of the denialist meme in general, and not just in regard to AGW. It’s common in anti-evolution circles, among those who insist a 767 couldn’t possibly start a fire that would cause a WTC tower to collapse, etc.

  • John Lederer // March 6, 2008 at 8:45 pm

    “They are much less important for assessing climate which look at changes in anomaly trends, which depend much less on microsite issues. You can get a reliable anomaly trend even if you use a wrongly calibrated thermometer which gives faulty absolute readings, provided the fault remains constant.”

    The problem with characterizing the result of what I suppose would now be a “known unknown” is that you don’t know whether your proviso is true or not.

    Looking at the pictures at surfacestations.org, one category of microsite influence seems to be the result of increased wealth — the small town fire department gets a window air conditioner for the office, the parking lot gets paved, a storage shed gets built.

    Cumulatively these factors might add up, and do so over a time scale somewhat akin to post 1970 climate warming.

    Do they? We don’t know, but the pictures indicate that we are measuring an extraneous effect, and one that seems to occur at nominally “rural” sites. Quite likely it has increased over the last few decades.

    Maybe I am unduly influenced from my own experience. I have lived in rural Wisconsin for 35 years.

    The old town hall, now the municipal garage, had the parking lot paved and window air conditioning installed in my memory. A new metal workshop/garage building is under construction. It isn’t a formal weather site, but has long had one of those little “box in a window” weather stations to aid in making plowing/salting decisions. I suspect the changes have likely affected it.

    I know no good way to separate trends in the instrument readings caused by local environment changes over time from those caused by climate warming.

    Do you?

    What the pictures illustrate is that we do not have good control of the data gathering process. That is serious. Much subsequent builds on the assumption that the temperature trends are measuring climate and not tax base growth in rural communities.

    Perhaps when surfacestations.org has completed its census we may be able to separate out stations that are free from such influences and do some sort of comparison with the rest . That might give a handle on the “known unknown”. Until then, or some other method of verification, I think everything built on the instrumental surface record has a shakier foundation than we thought.

  • P. Lewis // March 6, 2008 at 8:56 pm

    However, the result, at least to my mind, raises unsettling questions about the reliability of the surface temperature record.

    And so begins again the life cycle of the Greater Spotted Databad (Gaviiformes confusidus), forever doomed to repeat the same mistakes because the species is incapable of change.

    Along soon, if not already lurking, will be the Often Spotted Putyouright (Piciformes repeatimus), forever doomed to repeat the same mistakes, because the Greater Spotted Databad is incapable of change.

  • Hansen's Bulldog // March 6, 2008 at 9:28 pm

    Anthony Watts has restored the post as part 1 of his 3-part series. He’s also posted part 2 of the series.

  • Deech56 // March 6, 2008 at 9:30 pm

    This may be a silly question, but did all the stations suddenly improve over the last year to come up with the humongous drop in temperature that some have been touting as the end of global warming. Period. Of course, the fact that the temperature record has suddenly become useful has not escaped my notice.

  • Mike B. // March 6, 2008 at 9:39 pm

    Watts is still talking about Unadjuested data vs Adjusted data like somehow a different base period is “adjusting the data”.

    He is also describing the histogram results as having warming or cooling biases which again makes no sense.

    Doesn’t seem like he has really learned anything.

  • dhogaza // March 6, 2008 at 9:53 pm

    And so begins again the life cycle of the Greater Spotted Databad (Gaviiformes confusidus), forever doomed to repeat the same mistakes because the species is incapable of change.

    Sigh, so true. On the other hand, loons, confused or not, have been around for a long, long time so perhaps being a loony wingnut has survival value …

  • Hank Roberts // March 6, 2008 at 10:02 pm

    John writes, repeating stuff frequently reposted by people who’ve read it at CA or other sites second or third hand (sigh)

    > Do they? We don’t know, but the
    > pictures indicate ….
    > ..
    > … I know no good way to separate
    > trends in the instrument readings …

    And have you talked to the weather service people who maintain the site closest to you?

    What time of day is the temperature taken? Daytime? Nighttime? Both?

    If the A/C were running, would it be running when the temperature’s recorded?

    Do they leave the A/C running on holidays and weekends?

    Do they take the temperature on holidays and weekends?

    Ask yourself, is everyone else but the denial crowd really so stupid as not to have thought of these things?

    And why do you think it is that this is brought up over and over where the public can see it, but never published along with the dozens of papers that actually assess the performance of the weather instruments, compare how accurate they are, and discuss how and why the data is analyzed?

    Do you even know where to start looking for a primary source on this issue?

    Sheesh. Okay, I’m tired.
    I’ll give it up for a while and calm down.

  • John Lederer // March 6, 2008 at 11:19 pm

    Hank,

    I would be pleased and interested if you could point to a source that shows that the types of extraneous local environment factors indicated in the pictures are (1) known and (2) compensated for.

    I am not satisfied that the general algorithmic adjustments applied to the data does it. You should not be either.

  • Johan i Kanada // March 6, 2008 at 11:53 pm

    Anthony has done good job of documenting the apparent sloppy standards applied to surface station locations in the US. For that effort laudable he is somewhat known in the blogosphere.

    He does not know statistics, however, which is also something he admits. Consequently, in this case he made a pretty basic mistake (and/or didn’t communicate his intentions very well). But is this a reason to crucify him? Give the guy a break.

    And, if it is an apology for presenting inaccuracies you want (from a non-scientist/amateur), I recommend you start with Gore. He even made a film to promote his.

    [Response: In my opinion, Watts has done more than simply document the condition of surface stations in the U.S. He has misrepresented the impact of station conditions on temperature records.

    He did indeed make a pretty basic mistake. But he *still* insists that his mistake was simply failure to have part 2 ready! He hasn't admitted that his analysis was wrong -- he just hid it under the rug when it came under fire, and has now restored it but with the errors intact and neither disclaimer nor admission of error.

    I'm the one who said that if he'd simply admitted "Oops -- my bad," then we would have gained respect for his character. The way he handled this, however, discredits him far more than I ever could.

    Your opinion of Gore is as wrong as your opinion of Watts.]

  • John Lederer // March 7, 2008 at 12:55 am

    “He has misrepresented the impact of station conditions on temperature records.”

    Can you explain? Just a matter of tone, or has he calculated something wrong?

    What disturbs me about the surface station record is that it is easy to say “well that is not by the book” and see how readings for a particular station might be wrong, but very difficult to get a handle on the scope of the problem.

    When surfacestations.org has a completed global census that might help some. It would at least allow comparison between stations with no discernible problems and those with.

    However, it would seem that the database would have to be continually updated unless governments tightened up on enforcing their standards.

    There really is no good substitute that I can see for good data gathering.

    More and more we miss satellites expressly designed to measure radiation to and from the earth.

    [Response: I posted about surfacestations here.]

  • Hank Roberts // March 7, 2008 at 12:57 am

    John, temperatures are measured once or twice a day, check each station’s record.
    If there’s an air conditioner running, you know how hot the exhaust is. Do you know how many standard deviations it takes to kick out a temp. as anomalous?
    Do you know how an anomaly that shows up on work days but not weekends and holidays will be noticed?

    Do the basic research yourself and then ask for help. If you don’t know how, ask a reference librarian. If you don’t have a public library, say where you are located and I’ll see what I can do to help you.

    This is a good general guide:
    http://catb.org/~esr/faqs/smart-questions.html

  • Hank Roberts // March 7, 2008 at 1:07 am

    And a few places to start:

    http://eetd.lbl.gov/EA/Reports/41973/41973-2A.html
    Existing Climate Data Sources and Their Use in Heat Island Research,Chapter IIA. Methods of Historical Analysis

    http://cdiac.ornl.gov/r3d/ushcn/ushcn.html

    http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B7XNX-4KKNJ94-2&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=9db3ae1116ed04fbd3d9d62a60e4d6a2
    “… Utility of weather information from on-farm and weather stations was evaluated for the application in studies on the genetics of heat stress. Daily milk yield of 31 primiparous Holstein cows was collected at Tifton, GA, from April 28 to July 19, 1993. Weather information was recorded on-farm and was available from weather stations in Georgia. Analyses used daily average of temperature–humidity index (THI). Effects of threshold of heat stress and the rate of decline in milk after the threshold were estimated. …. At the Tifton weather station, 3 km away from the farm, the threshold was THI = 20 and the rate was the same. … Subsequent analysis included 2260 test day records from the same farm from 1993 to 2003 and weather data from Tifton station….”

    Above as an example of uses of local data.

    Below an example of using and vetting raw data for analysis:

    https://notus2.afccc.af.mil/SCISPublic/services/databases.asp

    “… The surface database contains worldwide surface weather observations (synoptic, airways, METAR, synoptic ship) for approximately 10,000 sites. All weather elements transmitted are retained; in some cases computed/derived values are incorporated into the record. Also available are “station files”-individual station datasets for selected stations that have received more quality control….”

    John, these took five minutes; a good reference librarian can do better by you in the same amount of time.

    Learn how to look up answers and work out who you think will help you find your own answers. Don’t be spoon-fed.
    Not by anybody. All that teaches you is trust and distrust, neither helpful here.

  • Jim Arndt // March 7, 2008 at 1:32 am

    Hi,

    Tamino I am the first to admit “my bad”. I jumped and didn’t bother to do the adjustment or look to see if it was done. Thought it was already done. I did the “assume” thing. But as for Mr Watts I still have a great respect for him and his work. Mistakes happen. As for the surface stations please see his presentation. Fairly well done and explains it more simply.
    http://gallery.surfacestations.org/watts-NYC-2008/page1.html

  • J // March 7, 2008 at 2:32 am

    One commenter on Watts’s most recent post writes: “Man you are smart. I only understood about 8 words. Keep up the good work!!”

    That, fundamentally, is the problem. Watts himself, as we’ve seen via this fiasco, has only a shaky understanding of basic statistics (and a shakier understanding of the field he’s trying to “audit”). Nonetheless, a fair portion of his audience is even less competent. It’s like the blind leading the blind over there.

    And while I’m sure many of Watts’s commenters proudly call themselves “skeptics”, their skepticism only runs in one direction — when it comes to examining claims against their fixed point of view, they’re remarkably credulous and naive.

    Watts’s initial blog posting involved some blunders, as Tamino pointed out. In the comments that followed, way too many people used Watts’s misleading histograms as confirmatory evidence to support their pre-existing beliefs that there are problems with GISTEMP (or, in some commenters’ suggestions, outright accusations of fraud). It’s not just one or two commenters, it’s the majority of the 86 comments.

    Somebody calling himself “Obsessive Ponderer” wrote “If you do a histogram of the satellite data from the same dates (I did it from Jan 1979 to Dec 1997) the data is symmetrically centered on 0 with the the 0 mark’s cumulative %age = 54%. If I am not mistaken, this is about as perfect as you can get with real data. No warming, no cooling.”

    That goes way beyond wrong. Mr/Ms OP doesn’t understand what an anomaly means, nor the difference between a positive anomaly and a positive trend. Seriously, he/she is suggesting that if the median value of a time-series is near 0, the time-series must have no trend (”No warming, no cooling.”)

    Of course, nobody bothers to correct this nonsense. Instead, it just fuels even more paranoia about NASA, Hadley, and climate science in general. Just to pick another example from the comment thread: “I think you already proved, no chiseled in concrete, that the GISS data, and some other data, is wrong and even deliberately falsified (though you would never make such an accusation!) It may already be too late to mitigate the humanitarian disaster which could arise if temperature is really going down!”

    Normally I would have some sympathy for Mr Watts, who seems to be a bit in over his head. But take the combination of his postings plus the comments (that virtually all go unchallenged), and what you end up with is a vast quantity of misinformation propagated outward across the blogosphere.

    Sorry for the long post here, Tamino, but Mr Watts seems to be disinclined to post my comment in that thread. (It was submitted before he closed comments, but never showed up….)

  • Heretic // March 7, 2008 at 3:26 am

    And of course, Watts’s effort to correct the “misperceptions” of his readers and posters is also limited by his own competence and bias, whether or not he denies having one (bias). I am not nearly as mathematically conversant as our host or some readers here but some stuff is shocking even to me. This is nonsense.

    At least Eli Rabett was able to acknowledge when and how he did a big mistake. I am ever more skeptical of the skeptics. I don’t believe that Watts deserves that title in its true, better meaning. Rushing to accuse GISTEMP of all evil without even a pause to make sure you’re not completely off is revealing of a certain mindset.

  • KH // March 7, 2008 at 3:33 am

    Hank, Guthrie, and Caerbannog: re petroleum companies climate modeling.

    I work in the petroleum industry (geophysicist) and any climate modeling done probably isn’t what most people think of as climate modeling. Generally geologists are interested in reconstruction of paleoclimates - millions to several hundred million years ago. Paleoclimates affected sea levels and thus sedimentary deposits world wide. The geologists don’t really work with recent climate.

    In studying paleoclimates, geologists are well aware of how climate has changed over geologic time. I think this is part of the reaseon many of them incorrectly attribute recent climate change to natural causes - they see it as just one more historic change.

    Hank, your idea of getting temperature logs from wells to invert for the past temperature record is interesting. As an older, but still keen geophysicist, I took a university course on geophysical inverse theory last fall. For my course project I thought of inverting temperature data as you mentioned. However, I discovered there were/are a couple of problems with that. While temperature data is recorded, it is often not requested, so oil companies often don’t get temperature data. And if a temperature log is received, the depth interval that the data is recorded is usually fairly deep - the data isn’t recorded right to surface. As was pointed out on an earlier post on this blog (September?) it is the shallower data that will give the information on more recent climate. There may be a few wells where such data is recorded to surface, but that would require an extensive search. Anyway, I picked another project.

  • Hank Roberts // March 7, 2008 at 4:08 am

    KH, here’s where I read that:

    “… As an indication of the extent to which the petroleum industry accepts the very same climate physics used in understanding global warming, at least one major oil company is using climate simulations of the Cretaceous (based on CO2 induced warming on top of geography changes) to spot promising oil formations. In fact, one of Paul Valdes’ industry-funded Cretaceous simulations was more or less embargoed from publication for five years because of its potential value to exploration. I hasten to add that Paul is one of the most respected climate modellers in the business, and his case shows that it is entirely possible to get some funding from the fossil fuel industry without compromising one’s research. The paleogeographic atlas project at U. of Chicago has also gotten funding from the oil industry from time to time….”

    and in the following response, an inline reply:

    “[... have sent Paul some email to see if there is any work in the petroleum exploration literature which explains how this is done — for all I know, the details may be proprietary. I am also checking up on the publication status of the specific Cretaceous simulations I referred to, and will post the references once I hear from Paul. –raypierrre]

    There’s a more recent (and currently open) thread related to the Cretaceous:

    http://www.realclimate.org/index.php/archives/2008/01/the-debate-is-just-beginning-on-the-cretaceous/

    Of course it’s possible the information is proprietary, but perhaps there will be more available.

  • guthrie // March 7, 2008 at 12:44 pm

    John Lederer- you are correct insofar as it is worth looking at what happens at stations that affects their temperature readings. However the proper scientific way of doing this is not to set up a website, publicise your photos and continually crow about how bad the results from the stations must be, judging by the photos.

    The scientific way is to study the temperature record of stations with known anomalies, whilst gathering data on lots of stations and possible things which can affect their temperatures, and withholding judgement on the issue. Then you analyse the data on what changes affect station data in what way, and write it up.

    Watts seems to prefer the first approach of crowing about possible errors first, without actually doing the scientific work to see what effect the surroundings of the instruments might have in the first place. Furthermore, he is as has already been mentioned, woefully unaware of what goes on with these stations in the first place.

  • John Lederer // March 7, 2008 at 1:57 pm

    Thank you gentlemen for your on point references about why it does not matter if a temperature recorder is near an air conditioner exhaust, above pavement, or located at the top of a telephone pole.

  • caerbannog // March 7, 2008 at 4:07 pm


    Thank you gentlemen for your on point references about why it does not matter if a temperature recorder is near an air conditioner exhaust, above pavement, or located at the top of a telephone pole.

    If the surface temperature record is so heavily contaminated by air-conditioner exhaust vents, etc… then why is the surface temperature record correlate so well with the satellite-based tropospheric temperature records? Do air-conditioning vents have an impact on microwave radiation emitted by the Earth’s atmosphere? “Enquiring minds” want to know!

  • caerbannog // March 7, 2008 at 4:09 pm

    arrrgghhh.

    That should be:

    “then why does the surface temperature record correlate so well…”

  • Hank Roberts // March 7, 2008 at 5:04 pm

    John, of course it matters. That’s why they throw out any extreme aberrant rating. That’s why they flag any station that reports exactly the same number over and over. That’s why they take four or five years of data from surrounding areas and look back at any questionable station, using the accumulated evidence.

    That’s why they warn that errors in the data for an individual station may persist for five years before they’re corrected.

    I’m sick of looking this stuff up over and over and tired of retyping even the short summaries.

    People get this nonsense in their heads and quit thinking for themselves and show up at one climate forum after another pasting in the stuff long since discredited that they find in all its original shining glory because the denial sites don’t correct their false claims.

    Sick of it. Look it up.

  • J // March 7, 2008 at 5:17 pm

    caerbannog:

    Believe it or not, what you intended as humor is actually playing out right now over on Watts’s blog.

    When Watts posted his original, misleading histograms, the difference between the satellites & GISTEMP was taken as proof that GISTEMP was doctored. Now that Watts has corrected his post, the close agreement between the satellites & GISTEMP … is being taken as evidence that the satellites must be wrong, too.

    They even mention, portentiously, that the beginning of the satellite temp record (1979) coincides with “Just when the Air Conditioning revolution took off.”

    As best I can tell, the new skeptic line is as follows:

    (1) Some type of in-situ data (radiosondes, whatever) must have been used to develop the original algorithms for inverting the satellite microwave radiometer data;

    (2) A different type of in-situ data (weather stations) has been shown to be crap, because all the stations are located next to air conditioners in parking lots.

    (3) Therefore, the MSU/AMSU satellite temperature trend is irrevocably contaminated by UHI/microsite problems.

    One interesting variant on this line, mentioned by another commenter over there, adds an extra step (to get around the fact that the surface data aren’t actually used to calibrate MSU per se), by pointing out that Mears et al used a GCM (!!!) to model the diurnal temperature cycle, as one small part of their 2002 reanalysis of the MSU record.

    Anyway, yes, caerbannog, you thought you were joking but within the last 24 hours we really do have people literally claiming that the satellite temperature record is invalid because it coincides with “the Air Conditioning revolution”.

  • mmghosh // March 7, 2008 at 7:14 pm

    Mr Lederer

    What I find particularly curious is your contention that professionals in the field of temperature reporting would not consider temperature anomalies in their own stations.

    Do you seriously believe that they do not check with their regional stations, the other stations in their geographical area, hold regular audits and instrument calibrations?

    Do you fly on aircraft or travel on ships? Is your contention that weather stations do not know their job and that a group of non-professionals can possibly be doing a better job than trained professionals at recognising the source of possible errors in temperature data?

    What is your basis for the supposition that amateurs do work better than trained professionals at gathering data?

  • John Lederer // March 7, 2008 at 7:47 pm

    “If the surface temperature record is so heavily contaminated by air-conditioner exhaust vents, etc… then why is the surface temperature record correlate so well with the satellite-based tropospheric temperature records? Do air-conditioning vents have an impact on microwave radiation emitted by the Earth’s atmosphere? “Enquiring minds” want to know!”

    I think that is the strongest point in confirming the trend reliability of both record sets.

  • John Lederer // March 7, 2008 at 8:11 pm

    Here is one portrayal of the trends of the various data sets, breaking the trend into periods when they seem by eyeball to have different slopes.

    I don’t know enough about the development of this graph to vouch for its accuracy, but it does show a very high correlation among the data sets.

    http://i29.tinypic.com/302oxli.jpg

    Assuming that there is no cross-referential development of the surface and satellite records, I would regard it as a strong indication of the overall trend accuracy of the global values.

    It would be interesting to break the data down by land and sea. I do not know any reason why the types of microsite environment extraneous factors we have been talking about would be a factor for ocean surface measurements.

  • Mike B. // March 7, 2008 at 8:24 pm

    (3) Therefore, the MSU/AMSU satellite temperature trend is irrevocably contaminated by UHI/microsite problems.

    Ahhh, remember the good old denier days when the UAH satellite values were the gold standard? Back when they showed a cooling trend before they fixed the bug…

  • steven mosher // March 7, 2008 at 8:44 pm

    Guthrie. Noaa have already done the study you suggest. When they set up the CRN they did a side by side study to judge the effects of siting.
    The Pristine site ( CRN1) was compared to a ( CRN2 ) site. An ASOS at an airport. The site was close to asphalt but not planted above it like you see in many of of Anythony’s surveys. The difference in mean temp was on the order of .25C. This difference was modulated by clouds and winds. This study confirmed work done by climate scientists like gieger and oke.
    Microclimate matters.

    We have good climate science that shows that micro climate matters. I don’t know why
    people persist in denying that science? Shrugs.

    We have no climate science that SHOWS these microsite effects have bias = 0. In fact, we have climate science that shows the opposite.
    USHCN SHAP adjustment is one such demosntration.

    So, we have NOAA, the agency that supplies GISS with data. Telling us this.

    1. Microsite matters.
    2. We need to improve the network.
    3. Our studies show that historical stations have
    some bias.
    4. In the interim we are applying new statistical techniques to find some of these undocumented issues. ( see Menne’s work)

    This is consensus climate science.

    Further. Noaa have determined ( CRN studies )
    that only 135 stations in the US are REQUIRED to capture climate trend signals. Again. This is not me. This is NOAA. this is the agency charged with maintaining data quality. This is the agency that GISS rely on. I accept their climate science at face value.

    My point. Very simple. very focused. Unrelated to GLOBAL WARMING.

    Climate science tells us that we can sample the Air surface tempature with 60 Optimal stations
    ( shen 98) NOAA, using a different approach,
    argues that CONUS requires 135 stations to
    track long term climate trends.

    Lets focus on NOAAs study. If we want to track the CLIMATE ( not the weather) then we need 135 good stations ( these are called CRN)

    Today USHCN has 1221 stations. Some of these stations are in violation of standards ( on rooftops for example, using outdated equipement etc etc)

    Since climate science tells us that we only need 135. and since climate science tells us that microsite can influence the signal, and since CONUS is such a tiny portion of the globe,
    I think it would be good practice to follow quality standards.

    Here is ANOTHER way to look at it. If the
    adjustment mechanisms of GISS adjust all the bad sites, then what is the point of the adjustment? Just elimate the bad sites. If you can identify them to adjust them, then just eliminate them. They are not needed to characterize the long term climate trends. Heck,
    we can do that with tree rings.

  • L Miller // March 7, 2008 at 8:46 pm

    John imagine you have a clock in your house. It keeps good time but is not set properly. Say, it’s out by 3 hours and 20 min, but you are not 100% just how much it’s out by. If you need to know exactly what time it is this can be a problem. If you are trying to time how long your roast has been in the oven the fact that they are out doesn’t matter at all because the start and end times are out by equal amounts.

    Now imagine you are trying to time how long your roast has been in the oven and someone comes along behind you and sets the clock to what they believe to be the correct time. Even if they are right you have still lost track of when your roast is supposed to come out of the oven.

    Similarly, while a weatherman reporting current conditions may be interested in the accuracy of the individual sites, what matters to someone studying clime change is consistency. As long as the site is consistent any error in absolute values will vanish when you calculate warming/cooling just as the error in your clock did in my example above. If someone comes along and wants to “fix” the stations they actually create an anomaly that makes the data more difficult to use for evaluating climate change.

    The type of errors climatologists need to look for in the data are therefore very different then what a weatherman needs to look for. The climatologist needs to look for anomalous changes in the values being returned from the weather station, something that is not going to be apparent from physically examining the station.

    Errors that effect the climatologists work can, however, be spotted by comparing it to other similar stations nearby. If they are all doing roughly the same thing, but one suddenly starts doing something else, or suddenly jumps up/down in a way not reflected by the others then that is a clear indication of an error that needs to be accounted for.

    Counter-intuitively, you can actually get this type of an error from *fixing* a previous error in the site, just as you did in the clock example above. If you have just one site that you fix and others nearby to use as a reference, you can still correct the error, but if you start “fixing” a whole bunch at once all bets may be off.

  • dhogaza // March 7, 2008 at 9:21 pm

    The difference in mean temp was on the order of .25C. This difference was modulated by clouds and winds. This study confirmed work done by climate scientists like gieger and oke.
    Microclimate matters.

    We have good climate science that shows that micro climate matters. I don’t know why
    people persist in denying that science? Shrugs.

    And I don’t know why people persist in conflating temperature measurements with trends. No one argues that microclimate won’t impact recorded temperatures. The point is that it’s up to YOU people to show that it is affecting the computed trends in a material way.

  • Hank Roberts // March 7, 2008 at 10:08 pm

    > steven mosher // March 7, 2008
    > at 8:44 pm

    Cite please.

    If you’re reporting second hand, please ask your source for a cite.

    And tell us why you consider your source believable if you haven’t read the original source yourself.

    Let’s have a look at the original source and follow the trend of the older and newer stations and see if there’s any difference in the trend.

    Not the temperature, the trend.

    Cite, please?

  • John Lederer // March 7, 2008 at 10:53 pm

    “The point is that it’s up to YOU people to show that it is affecting the computed trends in a material way.”

    I disagree.

    If you went into a lab that measured weight and noticed a smear of peanut butter and jelly on the scale, or went into a bacteriology lab and noticed flies buzzing from bacteria culture to bacteria culture, you would say “this lab does not meet standards. It’s work cannot be relied upon.”

    There are standards for weather stations. Many of the stations appear to have drifted from them over time. Without evidence one cannot say “Well maybe we violated the standards, but prove the results are wrong”. The purpose of the standards is to ensure reliability. No adherence to standards, no presumption of reliability.

    As to whether we are dealing with a mere temperature inaccuracy or a temperature trend inaccuracy, note my statement above. In many case it appears that the station drifted from standards over time — a tree grew, an air conditioner was installed, the parking lot was extended, the sensors were moved to put in a sidewalk, whatever.

    Those don’t just affect temperature. They affect the trend. They are changes, not a clock consistently wrong.

    How much of an effect? We don’t know. The strongest indication we have that it is not material is the fairly good correlation with the satellite record.

  • mmghosh // March 7, 2008 at 11:46 pm

    Mr Lederer

    I’m not claiming that there are no issues with weather stations. Of course there are issues with them as there are with every human activity, which need constant upgradation and investment.

    The points are that:
    1. This is the work of professionals who proceed at a certain speed.
    2. If Mr Watts can provide investment into these problems this approach might help.
    3. Bringing up these issues is separate from the issues of warming itself - which is indicated by the trend which even you, by now, must realise shows warming. Or are you suggesting that there is no real warming trend?
    4. This issue is of minor importance to the real story of the global warming and what is causing it. In fact, it is really a non-issue.

    Anyone can criticise any data collection mechanism in any public or private sector program. All programs have problems can be improved by funding - increasing manpower or by purchasing better and more accurate intrumentation.

    What does that have to do with whether the data from the mechanism is useful and can show worthwhile results?

    For example, the work done by Sanger in sequencing the insulin protein did not follow modern lab procedure. If your logic was to work in other scientific disciplines, then all previous work is “invalid” because it does not meet “current standards”. You don’t apply this logic in other disciplines, so why are you applying it here?

    If weather station data had showed a significant difference in temperature trends gathered from other sources, then you might have had a point. But they don’t. So you do not need to continue to labour it.

  • Gavin's Pussycat // March 8, 2008 at 6:43 pm

    > The point is that it’s up to YOU people to
    > show that it is affecting the computed trends
    > in a material way.

    Dhogaza, not even that: it is to show that research demonstrating that it is not doing so (e.g., Peterson et al, 1999, GRL), is invalid.

  • Gavin's Pussycat // March 8, 2008 at 7:31 pm

    Steven Mosher, about using 135 stations in the US to
    monitor climate, I beg to differ. It’s about redundancy.
    Having so many stations at short distances from one
    another is an asset, not a liability.
    It allows you t cross-validate, as GISS (and everybody
    else) are doing: every station’s trend is compared with
    a weighted average of those of its neighbours. Erroneous
    station trends stand out like a sore thumb. And the
    redundancy allows you to throw out the erroneous trends
    (note: trends, not stations, though you could do that
    too) as again GISS and everybody else is doing.
    With only 135 stations this wouldn’t work nearly as
    well. Nearby stations’ temperature trends correlate
    beautifully, but the correlation gets poorer in
    proportion to inter-station distance.
    I have a little secret for you, Steve :-) Even
    stations that are well designed, well operated and
    100% within specs, will occasionally produce wrong
    data, display long term drifts, etc. A century is
    a long long time. Cross-validation allows you to
    eliminate the station trends that actually are wrong,
    even from stations nicely within specs.
    Wouldn’t you sign for that? I know I would.

  • Lee // March 8, 2008 at 9:23 pm

    Anthony does it again. He has a post up now examining the trends over the last ten years of the 4 datasets - and he demonstrates once again that he has no clue about what an anomaly is or how it ‘impacts’ the data. Nor of statistical significance, nor of what cherry picking is and how it impacts analyses. Nor of the sources of the data he is ‘analyzing.’
    Here is his “conclusion”


    And finally we have the NASA GISS land-ocean anomaly data showing a ten year trend of 0.151°C, which is about 5 times larger than the largest of the three metrics above, which is UAH at 0.028°C /ten years.

    From my previous posts dealing with the subject of global temperature anomalies, this difference with NASA GISS likely has to do with the fact that they use a different base period than the other three metrics, but the question is why do they deviate from the other three with that choice? Given some of the recent issues Steve McIntyre has brought up with missing data at NASA GISS, it also makes me wonder if the GISS dataset is as globally representative as the other three.

    Treating the NASA GISS data as being an outlier due to that difference, and by taking a “3 out of 4 approach”, one could conclude though that there has not been much of a trend in global temperature anomalies in the past ten years.

  • mmghosh // March 8, 2008 at 11:11 pm

    Mr Lee,

    Although Mr Watts is confused about this issue, his confusions raise some valid points.

    One of them is using different baselines. Obviously, this concept is difficult people such as Mr Watts, or Mr Fred or some others who visit this blog.

    As many posters have demonstrated, putting all the graphs on the same baseline does eliminate the confusion so that even people such as Mr Watts can follow the trendline.

    Why do the different agencies not present data in this common format, in that case? I recently visited the NCDC site which shows graphs occasionally using the 1901 to 2000 baseline, occasionally using the 1979-2000 baseline. I understand when scientists say that they do this to demonstrate different aspects of the temperature story. But to non-scientists, this can be really confusing (as has been shown by the responses to Mr Watts posting).

    It would also help if all temperatures were reported in Celsius rather than Fahrenheit, rainfall in cm rather than inches etc. Some people from the anti AGW group seem to have difficulty with the differences between deg F and deg C.

  • Hank Roberts // March 8, 2008 at 11:21 pm

    Well, he’s established that he turns only one way. Next candidate for worthy opposition, please come forward.

  • Gavin's Pussycat // March 9, 2008 at 10:04 am

    > Why do the different agencies not present
    > data in this common format, in that case?

    Hmm yes, why not? Perhaps because the kind of analysis Mr Watts is trying to do ought to be done by people minimally versed in quantitative methods (and as a corrollary, not dumb)?

    Someone long ago remarked on Kuiper’s seminal “The Atmospheres of the Earth and Planets” that the equations it contained “look like barbedwire, and share the same function: keeping trespassers out”. Perhaps these different baselines serve as a useful tripwire, to trip up ignoramuses before they proceed to more easily confoundable non-issues :-)

    (Yes I know I am being mean; no I don’t feel apologetic.)

  • Hank Roberts // March 9, 2008 at 4:03 pm

    Yep. What gets me is the almost complete absence of discussion of contemporary climate science papers, both by the scientists online, the serious readers, and the curious bystanders.

    Every single discussion gets piled on by people all fired up by CA and WUWT and the Isdos and other websites.

    They keep loading these people with blanks and sending them in to make noise.

    I sure miss threaded newsreaders with killfiles.

    Come to think of it, by now, assuming sci.environment has stayed about the same, the information content there is likely to be better than what any blogger can do online.

    RC deals with the major papers, but the threads do spend most of their time whacking airhead questions.

    Sites like this, sad to say, are just made unreadable. It sank Prometheus because RPJr was fond of the moles, but I fear it’s sinking this site just because the host tolerates having the majority of the text written by moles.

    Look at what happened to dot.earth.
    It got so popular that nobody goes there anymore.

    It’s time I went back and did my reading in the libraries for a few weeks.

    Good luck.

  • mmghosh // March 9, 2008 at 4:19 pm

    GP

    The really significant point, as it appears to me is that the whole of 2007 showed a warming trend, inspite of the most significant La Nina effects in the past century, and being at the lowest point od the 11 year solar irradiance cycle.

    When the next El Nino arrives, it will arrive on the rising solar irradiance curve. Then the GHG-induced warming will become much more significant - from the NASA GISS site this would appear to start in 2009-2010.

    http://data.giss.nasa.gov/gistemp/2007/

    What natural cooling processes could be operative in 2010? There might be another simmering volcano like Pinatubo that will lower the anomaly - I looked for data from current vulcanology on Google but was unable to find any.

    China, India, Brazil might be planning to lay down more coal firing stations, I believe. Will they produce enough aerosols as they did in the 1940-1970 period? The problem with that would be that CO2 would rise together with the aerosols.

  • L Miller // March 9, 2008 at 4:55 pm

    Changing the baseline would create unnecessary confusion and is mathematically worthless because it doesn’t actually change anything important.

    Imagine trying to read a climate paper and verity it’s results only to find the data set that climate paper used has changed it’s baseline and now shows completely different values. Sure you could adjust it if you knew what baseline they happened to be using on that day, but this forces you to make a whole series of corrections just for the simple act of reading a paper. Worse, every person who wants to read that paper needs to make the same adjustments.

    In the rare case it’s required it’s so much simpler to have the researcher make any adjustments to the baseline when they write the paper.

  • Hansen's Bulldog // March 10, 2008 at 3:15 am

    I’m posting this to multiple threads. Things have gotten a bit out of hand. The level of hostility is … well … over the top.

    So I urge everyone to try an experiment. For you next comment, make the exact point you wanted to make, but leave the hostility at the door. I really don’t want to censor viewpoints. But as has been pointed out, too much insult not only turns people off, it interferes with communication. Meanwhile, it’s time for me to reconsider my moderation policy … because “moderation” doesn’t seem to apply.

    And please don’t send a comment saying it’s really my fault. I already know.

  • JimR // March 10, 2008 at 10:03 pm

    mmghosh - keep in mind that 2007 started in a warm El Nino phase that peaked in Dec. 2006. Jan. 2007 was the warmest Jan. in the GISS record and the D-J-F temperature was also the warmest on record showing that 2007 started off very warm. Without this very warm start 2007 would have been a fairly cool year.

    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt

    http://www.ncdc.noaa.gov/oa/climate/research/2007/ann/enso-monitoring.html

    I really don’t understand why Hansen in the GISS 2007 Summation only mentioned the La Nina underway as 2007 ended but not the El Nino underway as 2007 started.

  • Lee // March 11, 2008 at 3:01 pm

    I’m posting this here to preserve it - Anthony has taken to cutting my posts over there, especially the posts about anomalies.
    Criticism of my explanation - either here or there - is welcome.

    Sheaks3,

    You asked:

    “But the baseline is the temperature they consider normal and the anomaly is the difference between the normal temperature and the measured temperature i.e. anomaly. How did they determine the normal temperature?”

    Thank you for asking that, because I think this is the basis of the misunderstanding.
    The zero point is NOT claimed to be a “normal” temperature in the sense of “what the temperature should be.”. It is simply a zero point, a baseline - a ‘norm’ meaning an agreed baseline, not a ‘normal’ meaning ‘what it should be.’ An anomaly of 0.2C does not mean it is 0.2C warmer than some perfect “normal” temperature - it means it is 0.2C warmer than the mean for the 30 year baseline period.

    Why is this done? It falls out of the algorithms that allow comparisons of individual stations or grid cells. Lets say I want to compare, here in California, Redding and Walnut Creek temperatures for July 15. A hot day in WC is 95F, while a hot day in Redding is 108F - so if I want to know if its a comparatively hot or cool day, I cant just compare raw temperatures. I could just pick a single month for a baseline, say, July 15, 1980, and look at deviations from that one month. But that makes the comparison vulnerable to weather in that one month. Perhaps July 15, 1980 was foggy in Walnut Creek, so the temps that day were cold for July. That would mean that the comparison was always to a cool WC date, and it would tend to make WC look like it got a lot hotter since 1980 than it really did. Redding may have been hot on that day, so the comparison would make it look like Redding is getting colder, adn WC is getting hotter - but the difference is being driven by weather on that one day in July 15, not any actual change in temperatures at those sites.

    To avoid this, I can take the average of the temp on July 15 for a 30 year period. Empirically, 30 years is long enough to average out the weather, the hot and cold days, and get a typical (not normal, not ‘perfect’, but typical for that 30 year period) July 15 temperature. And I can compare to that 30 year mean value to see if this year’s July 15 is warmer or cooler than the mean of that 30 year period. And I can compare Redding and WC to see if they are showing the same trends, against the same base line period, with the baseline long enough to have averaged out weather differences in the baseline period.

    Again, not warmer or colder than “normal” but warmer or colder than the mean of the July 15 temperatures for the period 1960 - 1989, for example.

    Anomaly values ONLY make sense in comparison to the baseline, or in comparison to another set of anomaly values determined from the same baseline.

  • VG // March 13, 2008 at 9:44 am

    From Lucia’s blog quoted ad verbatim (I thought she was on your side of the debate?)
    3. What about falsification of IPCC’s prediction?

    No matter which method we choose, the IPCC estimate is outside the uncertainty bands. Moreover if we examine the IPCC’s estimate for their uncertainty bands (which are only provided in graphical form), the entire range of uncertainty the IPCC applied to their estimate falls outside the current data set.

    So, looked at both ways, IPCC near term prediction of 2C/century for the trend appears provisionally falsified.

Leave a Comment