Scroll To Top

Wednesday, January 28, 2015

HadCRUT4 - equalling (slightly besting?) the hottest year with a lie from Anthony Watts at WUWT

Sou | 8:53 AM Go to the first of 28 comments. Add a comment

HadCRUT4 is out. It's another hottest year ever on record result at 0.563C above the 1961-1990 mean. That just tops the next hottest, 2010, which was 0.555C above the mean if you go by straight numbers. It makes it equal hottest once you allow for confidence ranges.

Some people were wondering if HadCRUT4 would come out on top. It could be argued that it did. Just. The Met Office is being cautious, saying it is too close to call, which is fair enough. From their press release - my emphasis:
26 January 2015 - Provisional full-year global mean temperature figures show 2014 was one of the warmest years in a record dating back to 1850.
The  HadCRUT4 dataset (compiled by the Met Office and the University of East Anglia's Climatic Research Unit) shows last year was 0.56C (±0.1C*) above the long-term (1961-1990) average.
Nominally this ranks 2014 as the joint warmest year in the record, tied with 2010, but the uncertainty ranges mean it's not possible to definitively say which of several recent years was the warmest.
Colin Morice, a climate monitoring scientist at the Met Office, said: "Uncertainties in the estimates of global temperature are larger than the differences between the warmest years. This limits what we can say about rankings of individual years.
"We can say with confidence that 2014 is one of ten warmest years in the series and that it adds to the set of near-record temperatures we have seen over the last two decades."

WUWT and GWPF - disinformers lie through their teeth!

There is no such hesitation or caution among the climate disinformers though.  Professional liar denier Anthony Watts claims, wrongly, that:
"UK Met Office says 2014 was NOT the hottest year ever due to ‘uncertainty ranges’ of the data". 

That's wrong. The UK Met Office did NOT say 2014 was not the hottest year ever. It's one thing to say it's too close to call even though the numbers show it the hottest. It's quite another to say that it was NOT the hottest year ever. Just how desperate must he be feeling. Honestly, you've got to wonder if Anthony Watts can lie straight in bed!

His headline was above a "press release" by the GWPF, which is an anti-mitigation lobby group from the UK.  In the body of the release, David Whitehouse, who I guess is paid to tell fibs, wrote:
Quoting the temperature to one hundredth of a degree and the error on that measurement to a tenth of a degree is not normal scientific practice. It is against normal scientific practice to have an error of the measurement larger than the precision of that measurement. This means that most scientists would have rounded the data so that it was 0.6 +/- 0.1 °C. If this is done to the HadCRUT4 dataset it is even more obvious that there has been a warming “pause” for the past 18 years.

That's not right or correct or accurate. The Met Office provides data to three decimal places and reported it to two decimal places. 0.563C was the reported median and 0.529  and  0.600 were the lower and upper bounds at the 95% confidence interval.

I guess when the disinformers can't claim that an ice age is imminent, muddying the hottest year with a pause is the next best thing.

I'm very pressed for time this week but will just add two more things. First a chart of HADCRUT4 - click to enlarge it:

Data Source: UK Met Office Hadley Centre

No time for comments - you can read the WUWT archive here.

Two Million Views

Secondly, yesterday HotWhopper had it's two millionth page view. Thank you to everyone!


  1. The stupid, it burns as you say. Really, he quotes the Met Office phrase of 'uncertainty ranges', but then uses that to make a certain statement about 2014's place as the hottest (or not) year. Even when he lies, he does so stupidly.

  2. "....yesterday HotWhopper had it's two millionth page view. Thank you to everyone!...."

    It was me. Your welcome!

  3. So that's 2014. How is January 2015 looking?

  4. I don't know about elsewhere, but it's been a relatively cool January here in south eastern Australia. At the Australian Open this year they are wearing jumpers - unlike last year when the players were collapsing in the heat.

    1. ice age! ice age! iiiiiice aaaaaaaaage!

      [anti-Poe: ;-) ]

  5. It's odd that the denial-o-sphere had the opposite reaction back when NASA made a tiny adjustment to the U.S. annual temperature record which resulted in 1934 becoming a fraction of a degree warmer than 1998.

    The results of the adjustment were:
    Before: 1934 and 1998 in a statistical tie, with 1998 edging out 1934.
    After: 1934 and 1998 in a statistical tie, with 1934 edging out 1998.

    No objections to fractions of a degree back then -- the cry was "NASA admits that 1998 wasn't the warmest year!"

  6. Now that the deniers have accepted the concept of confidence limits, I look forward to their acceptance that their beloved "pause" is not statistically significant either

  7. "How is January 2015 looking?"
    I keep a NCEP/NCAR reanalysis tally here. The first ten days were very globally cool, then very warm, easing off lately. On average so far, much like December.

    1. thanks Nick. Does "just like December" mean hot? I can't tell from the +0.225 so far.

    2. David,
      With GISS, for example, December was a mid-ranking month for 2014.

  8. When multiple datasets have the same year as the hottest, doesn't that increase the odds that 2014 was the hottest?

  9. Much like December Nick? With GISTemp, December was the second hottest December ever :)

    (Fourth hottest in HadCRUT4)

  10. I wish HadCRUT4 would stop presenting global averages to 3 significant figures. It's not very convincing.... and why can they give 3 significant figures, yet GISTEMP and NOAA only 2, when they are working from the same raw data?

  11. Their model is quite different to GISS's, David. Nick or Victor might be able to shed more light. You'll notice that their report acknowledges the confidence limits. I'd go by that if you're uncomfortable using the data.

  12. DA,
    "why can they give 3 significant figures, yet GISTEMP and NOAA only 2, when they are working from the same raw data"

    Well, I predicted it (at short range) to 3 figs :)

    Here is NOAA giving 6 figs.

    It really doesn't matter. People say you shouldn't give more sig figs than the accuracy for a measurement. But this is not a measurement; it is a calculated average of more than a million measurements. Measurement error (or at least its random part) is much reduced. Their uncertainty is based on a model of what would happen if you measured differently.

    It's not splitting hairs. The advice re measurement is on the basis that you don't have any basis for the third fig. But here you do - the 0.563 is a well characterised number (which is why I mentioned my prediction). They acknowledge that it should be seen as a sample of a variable population.

  13. Nick, I understand how the measurement error is much reduced -- it goes like 1/sqrt(number_of_measurements).

    I admit I don't know the particulars of their data model. But if HadCRUT calculates to 3 significant figures, why don't GISTEMP and NOAA? Surely they want to squeeze all the sign digits they can get out of the data....

    And, is there any point to calculating to 0.001 C? I can't see any reason why the third digit matters. It's not like this data point can be compared to some theoretical prediction....

  14. On Nick's site you can track the month as it unfolds. It's a cool feature. Well, except for Jan 2015, which to me looks hot.

  15. This site also gives a daily reanalysis:

  16. If you write a publication, it is good practice not to give more significant digits as the errors allow. In this case the error is mainly because the stations only sample the surface temperature and interpolation is necessary. This error is about 0.1°C.

    Thus all groups give more significant digits than one would in a publication. That is no problem what soever, because this is data, not a publication. People perform computations on this data and it would be a pity if there would be any reductions in accuracy in those computations due to premature rounding of the data by the data providers.

    For many computations this is not important, but for some it might be. It is thus a compromise between accuracy for as much applications as possible versus the size of the dataset. Given that for small stations datasets, data storage and transfer is almost for free nowadays, people may tend to add more digits nowadays.

    A good example of a computation that needs many digits would be the application mentioned in the post, the computation of rankings. 2014 is the warmest year in the HadCRUT dataset and if you are interested in the statistics of rankings it would be a pity not to have unique rankings, that would make the analysis harder.

    There is probably still data coming in and data that was transferred in near real time is still being updated by data that is quality controlled by the weather services. Thus the values can still change. If in summer the average of 2014 is still the highest, I have no problem calling it a record. The temperature record over 2014 is the only case I know where all of a sudden people claim a need to take uncertainties into account before ranking. Normally the record is simply the highest value, like I wrote in my last blog post.

    Excerpt: Last September a new marathon world record was set in Berlin. Dennis Kimetto set the world record with a time of 2:02:57, while the number two of the same race, Emmanuel Mutai, set the world second best time with 2:03:13. Two records in one race! Clearly the conditions were ideal (the temperature, the wind, the flat track profile). Had other good runners participated in this race, they may well have been faster.

    Should we call it a record? According to the traditional definition, Kimetto run fastest and has a record.

    According to the new definition, we cannot be sure that Kimetto is really the fastest marathon runner on the world and we do not know what the world record is. Still newspapers around the world simply wrote about the record as if it were a fact.

  17. I think the deniers are tring to invoke memories of John Clarke:

    John Clarke: I'm a humble man, Mr Copperfield, but I refuse to be remembered as he who ran the 100 metres at the Sydney Olympics over 94 metres

    To non Aussies look up "The Games" on ABC television.

    R the anon.

  18. Oh yes and of course this on the record...
    John Clarke: Mr Wilson. Do you know who is the current 100 metres all-comers Australian record holder?Mr Wilson: Can I guess?John Clarke: There's not much point in guessing.Mr Wilson: Is he an African American?John Clarke: He's not an African American, no.Mr Wilson: Is he that Canadian from Jamaica?John Clarke: No, he's not a Canadian from Jamaica.Mr Wilson: I give up.John Clarke: The 100-metre record in this country, Mr Wilson, is currently held by Bryan.Mr Wilson: Bryan?Bryan Dawe: Yes.Mr Wilson: Congratulations.Bryan Dawe: Thank you.John Clarke: A new mark, Mr Wilson, set at a blistering session last Wednesday. I wish you'd been there. We were down there and we had a bet.Mr Wilson: Was this wind assisted?John Clarke: No, and we'd had a couple, and in my view Bryan is not in quite the nick he was in at the same stage of last season.


  19. Have I missed something? On this blog I used to be able to reply to comments, or to just follow all comments without commenting myself.

    1. Try it now, Harry. I changed it when Numerobis was having problems commenting, but they haven't been here lately and, anyway, it caused more problems than it solved.

    2. I've been lurking! But life has gotten very busy of late (my startup is starting to actually start up).

    3. Good for you. Best of luck. (Care to share?)

    4. I feel funny using your blog to pimp my stuff, but since you asked:

      I'm working on Imaginary Spaces (link above) -- an app to let anyone design architecture for work or play, and 3d print them. We're hoping to be on ipad/android soon (just takes jumping through hoops), there's a free demo available on mac/windows now.

      And we're doing workshops at makerspaces for kids using our app, to teach them 3d modeling ideas, as per our facebook post here:
      If anyone wants to try to replicate this in their town, drop us a line (

      Nothing to do directly with climate, though I do wonder whether the advent of 3d printing might reduce the number of widget-laden ships crossing the Pacific.

    5. Brilliant - looks superb :)


Instead of commenting as "Anonymous", please comment using "Name/URL" and your name, initials or pseudonym or whatever. You can leave the "URL" box blank. This isn't mandatory. You can also sign in using your Google ID, Wordpress ID etc as indicated. NOTE: Some Wordpress users are having trouble signing in. If that's you, try signing in using Name/URL or OpenID. Details here.

Click here to read the HotWhopper comment policy.