A New Leaderboard at the U.S. Open

There has been some turmoil yesterday on the leaderboard of the U.S. (Temperature) Open and there is a new leader.

A little unexpectedly, 1998 had a late bogey and 1934 had a late birdie. (I thought that they were both in the clubhouse since the turmoil seemed to be in the 2000s.) In any event, the new leader atop the U.S. Open is 1934.

2006 had a couple of late bogeys and fell to 4th place, behind even 1921. I think that there’s a little air in the 2006 numbers even within GISS procedures as the other post-2000 lost about 0.15 strokes through late bogeys, while it lost only 0.10 strokes. It is faltering and it might yet fall behind 1931 into 5th place.

Four of the top 10 are now from the 1930s: 1934, 1931, 1938 and 1939, while only 3 of the top 10 are from the last 10 years (1998, 2006, 1999). Several years (2000, 2002, 2003, 2004) fell well down the leaderboard, behind even 1900. (World rankings are calculated separately.) Note: For the new leaderboard see http://data.giss.nasa.gov/gistemp/graphs/Fig.D.txt. The old data has been erased; by sheer chance, I had the old data active in my R-session but I can’t give a link to it.)

GISS U.S. Temperatures (deg C) in New Order

Year Old New
1934 1.23 1.25
1998 1.24 1.23
1921 1.12 1.15
2006 1.23 1.13
1931 1.08 1.08
1999 0.94 0.93
1953 0.91 0.90
1990 0.88 0.87
1938 0.85 0.86
1939 0.84 0.85

Here’s the old leaderboard.

Year Old New
1998 1.24 1.23
1934 1.23 1.25
2006 1.23 1.13
1921 1.12 1.15
1931 1.08 1.08
1999 0.94 0.93
1953 0.91 0.90
2001 0.90 0.76
1990 0.88 0.87
1938 0.85 0.86

157 Comments

  1. pk
    Posted Aug 8, 2007 at 12:54 PM | Permalink | Reply

    No doubt whispered with a British accent.

  2. Ron Cram
    Posted Aug 8, 2007 at 1:10 PM | Permalink | Reply

    I think this answers my question regarding what years were affected by the error. Apparently every year since 2000. Although I still do not understand how 1934 got the late birdie to take the lead.

  3. John Goetz
    Posted Aug 8, 2007 at 1:19 PM | Permalink | Reply

    pk, I was thinking the same thing as I read the posting :-)

    Steve, does GISS recognize / acknowledge the above or is this your analysis only thus far? When do the world rankings come out?

  4. Steve McIntyre
    Posted Aug 8, 2007 at 1:29 PM | Permalink | Reply

    This is based entirely on GISS numbers as presently online. The results are pretty much in accordance with what I ball parked (and BTW I don’t necessarily agree that their processing is written in stone.) 2006 wasn’t adjusted as much as the other years – why: who knows?

    They’ve said that this does not make any real difference to the world rankings. The problems with Chinese and Indonesian SST temperatures are each distinct. I’ve observed many times that the US results, where there are rural stations, differ remarkably from the ROW results and the ROW results deserve even more analysis than the US results.

  5. Stan Palmer
    Posted Aug 8, 2007 at 1:29 PM | Permalink | Reply

    Would it be possible to put up the former leader board for comparison?

  6. bernie
    Posted Aug 8, 2007 at 1:30 PM | Permalink | Reply

    Steve:From whence?

  7. John Goetz
    Posted Aug 8, 2007 at 1:35 PM | Permalink | Reply

    Steve, thanks for the explanation. I had gone to the GISS website looking for a press release and, when I found none, thought I would ask :-)

  8. J Edwards
    Posted Aug 8, 2007 at 1:45 PM | Permalink | Reply

    Steve, Have you done a new linear regression on this data? I’m getting some very interesting results (trend is .04 deg C/decade).

  9. J Edwards
    Posted Aug 8, 2007 at 1:46 PM | Permalink | Reply

    Should be .046 deg C/decade.

  10. DeWitt Payne
    Posted Aug 8, 2007 at 2:01 PM | Permalink | Reply

    Steve, could you put up a plot of the lower 48 US states temperature anomaly from 1979 through 2006 using the new GISS, the latest Hadley and the UAH and RSS MSU data? I was only able to find the data from version 2 of HADCRU up to 2005 in easy-to-use form. The correlation between yearly averages from Hadley and UAH seemed to be quite good to me, 0.92 using Excel CORREL function. Interestingly enough, the linear trend through the yearly averages was slightly higher for the UAH satellite lower troposphere data. In fact it was 0.29 C/decade compared to 0.24/decade for the surface temperature. This is, IIRC, about the amount the models predict.

    Yes, we’re talking about less than 2% of the earth’s surface, but all the systematic errors in the satellite data interpretation found so far have been global, so one good calibration point may be enough.

  11. Geoff Olynyk
    Posted Aug 8, 2007 at 2:10 PM | Permalink | Reply

    Let me get this straight, because I want to be sure of the significance of this: with the adjustments due to the fixing of the “Hansen Y2K error”, the ranking of the hottest years in America has been largely re-ordered? If that’s the case, that’s pretty big news. I hope the University of Guelph (that’s where you work, right?) has their publicist ready to handle the calls!

  12. Steve McIntyre
    Posted Aug 8, 2007 at 2:34 PM | Permalink | Reply

    #11. No, it’s my friend and coauthor McKitrick at U of Guelph. No calls for me yet. They’ve been “re-ordered” – have they been “largely re-ordered” – that’s a stronger claim.

  13. Jack
    Posted Aug 8, 2007 at 2:41 PM | Permalink | Reply

    Does this affect the global rankings substantively?

  14. Mark T.
    Posted Aug 8, 2007 at 2:42 PM | Permalink | Reply

    2006 had a couple of late bogeys and fell to 5th place, behind even 1921.

    Shouldn’t that read “fell to 4th place”, not 5th?

    Mark

  15. Jack
    Posted Aug 8, 2007 at 2:42 PM | Permalink | Reply

    Sorry, now I’ve read post #4. Nevermind.

  16. Sam Urbinto
    Posted Aug 8, 2007 at 3:42 PM | Permalink | Reply

    I wonder now what everyone thinks of the surfacestations work over at the other blogs. Or even the openness and worth of the adjustments done, and the “trust us we’re climate scientists, no need to examine our work” mantra.

  17. Posted Aug 8, 2007 at 4:15 PM | Permalink | Reply

    Is it just me or have they changed their numbers again? I’d swear this morning (Europe time), when I posted in the Hansen-Y2K thread, 2006 was still ahead of 1921.

  18. Posted Aug 8, 2007 at 5:01 PM | Permalink | Reply

    Don’t suppose you might link to the actual page within the GISS website?

  19. BarryW
    Posted Aug 8, 2007 at 5:13 PM | Permalink | Reply

    My guess is that the AGW True Believers will say that it means nothing and we’re just nitpicking.

  20. Earle Williams
    Posted Aug 8, 2007 at 5:18 PM | Permalink | Reply

    Re #19

    bigcitylib,

    Maybe not as obvious as gravel on asphalt, but the link is in the last paragraph.

  21. JS
    Posted Aug 8, 2007 at 5:24 PM | Permalink | Reply

    Well, that is the first of the statistical mistakes… I wonder what it will look like after the asphault, wind and bbq adjustment.

  22. tetris
    Posted Aug 8, 2007 at 5:26 PM | Permalink | Reply

    Re: 19
    BCL
    Bit of a cold shower? Must be.
    Assuming GISS adjusts 2006 down some more [as they may well have to] we’ll be left with a bladeless hockey stick.
    Talk about putting the puck in the net. Well done, Steve.

  23. tetris
    Posted Aug 8, 2007 at 5:42 PM | Permalink | Reply

    Re:5
    Stan
    The old leaderboard has been erased, but Steve appears to have archived that data. It certainly makes one wonder why the need to cover the tracks?

  24. Posted Aug 8, 2007 at 5:44 PM | Permalink | Reply

    Even a stopped clock is right twice a day. Congrats Steve. Did a quickie post on this at:

    http://bigcitylib.blogspot.com/2007/08/give-denier-his-due.html#links

  25. Mark T.
    Posted Aug 8, 2007 at 5:52 PM | Permalink | Reply

    It certainly makes one wonder why the need to cover the tracks?

    I don’t think they’re covering their tracks, it’s simply outdated information now and should be removed.

    Mark

  26. JerryB
    Posted Aug 8, 2007 at 6:09 PM | Permalink | Reply

    Climate change in action.

    Several renditions of GISS USA historical temperature average ‘anomalies’
    may be viewed at the following links.

    as of June 2000
    as of March 2001
    as of February 2007
    as of this morning

  27. Arnost Khun
    Posted Aug 8, 2007 at 6:13 PM | Permalink | Reply

    A saved version of the data (excludes 2006) is available via the Wayback Machine here:

    http://web.archive.org/web/*/http://data.giss.nasa.gov/gistemp/graphs/Fig.D.txt

  28. tetris
    Posted Aug 8, 2007 at 6:20 PM | Permalink | Reply

    Re: 26
    Mark T
    One does not discard old data sets just because there are now new “adjusted” ones. We have refence libraries full of “old” data, invaluable for comparative analysis. We didn’t erase Newton’s calculations after Einstein demonstrated that they were partially flawed and “updated” our understanding of the universe. [Steve M., my apologies for the somewhat off-topic analogy].

  29. Posted Aug 8, 2007 at 6:25 PM | Permalink | Reply

    Congratulations Steve!

    On your own you’ve managed to reduce temperatures more than the Kyoto Protocol ever could!

    –Chris

  30. John Goetz
    Posted Aug 8, 2007 at 6:25 PM | Permalink | Reply

    #28…Google cached a version on Jul 23, 2007 15:04:08 GMT. Here it is:
    Contiguous 48 U.S. Surface Air Temperature Anomaly (C)
    ——————————————————
    year Annual_Mean 5-year_Mean
    ———————————
    1880 -.27 *
    1881 .28 *
    1882 .08 -.24
    1883 -.67 -.29
    1884 -.63 -.41
    1885 -.52 -.46
    1886 -.29 -.39
    1887 -.20 -.21
    1888 -.33 -.07
    1889 .28 -.05
    1890 .18 -.11
    1891 -.20 -.19
    1892 -.50 -.22
    1893 -.73 -.39
    1894 .16 -.31
    1895 -.65 -.23
    1896 .17 -.11
    1897 -.08 -.22
    1898 -.14 .02
    1899 -.41 .00
    1900 .57 -.01
    1901 .05 -.11
    1902 -.13 -.13
    1903 -.64 -.33
    1904 -.48 -.35
    1905 -.46 -.37
    1906 -.01 -.21
    1907 -.25 -.17
    1908 .14 -.03
    1909 -.28 .01
    1910 .27 -.11
    1911 .15 -.15
    1912 -.87 -.08
    1913 -.04 -.16
    1914 .08 -.30
    1915 -.13 -.33
    1916 -.52 -.31
    1917 -1.06 -.35
    1918 .06 -.41
    1919 -.11 -.08
    1920 -.41 .17
    1921 1.12 .14
    1922 .17 .02
    1923 -.07 .17
    1924 -.73 -.05
    1925 .36 -.05
    1926 .04 -.02
    1927 .14 .01
    1928 .08 -.03
    1929 -.57 .18
    1930 .16 .15
    1931 1.08 .27
    1932 .00 .63
    1933 .66 .60
    1934 1.23 .42
    1935 .03 .39
    1936 .18 .43
    1937 -.13 .35
    1938 .85 .36
    1939 .84 .44
    1940 .03 .49
    1941 .62 .35
    1942 .09 .21
    1943 .16 .20
    1944 .14 .21
    1945 -.03 .21
    1946 .70 .17
    1947 .09 .18
    1948 -.08 .12
    1949 .20 -.10
    1950 -.30 -.05
    1951 -.42 .14
    1952 .32 .27
    1953 .91 .32
    1954 .82 .45
    1955 -.05 .42
    1956 .27 .25
    1957 .14 .12
    1958 .07 .09
    1959 .17 .03
    1960 -.23 .00
    1961 .00 .02
    1962 -.02 -.03
    1963 .19 -.01
    1964 -.08 -.05
    1965 -.12 -.07
    1966 -.24 -.16
    1967 -.10 -.19
    1968 -.28 -.19
    1969 -.23 -.17
    1970 -.12 -.22
    1971 -.10 -.11
    1972 -.36 -.04
    1973 .25 -.05
    1974 .16 -.08
    1975 -.19 .07
    1976 -.23 -.08
    1977 .37 -.23
    1978 -.51 -.15
    1979 -.58 .03
    1980 .22 -.12
    1981 .65 -.01
    1982 -.36 .11
    1983 .00 -.02
    1984 .02 -.01
    1985 -.42 .24
    1986 .73 .30
    1987 .85 .26
    1988 .34 .52
    1989 -.18 .52
    1990 .88 .41
    1991 .70 .26
    1992 .31 .39
    1993 -.43 .28
    1994 .47 .11
    1995 .35 .06
    1996 -.17 .39
    1997 .05 .48
    1998 1.24 .54
    1999 .94 .76
    2000 .65 .88
    2001 .90 .76
    2002 .68 .69
    2003 .65 .73
    2004 .60 .80
    2005 .85 *
    2006 1.23 *
    ———————————

  31. Paul S
    Posted Aug 8, 2007 at 6:30 PM | Permalink | Reply

    With the correction for the period 2000-2007, what do the records now indicate for total warming in the US in that time period?

  32. Mike
    Posted Aug 8, 2007 at 6:31 PM | Permalink | Reply

    We’re clearly experiencing wild climate change across the board!

  33. gdn
    Posted Aug 8, 2007 at 6:38 PM | Permalink | Reply

    Several renditions of GISS USA historical temperature average anomalies’
    may be viewed at the following links.

    So the average of the 1930s was +.500 from the defined mean, and the average of the 1990s was +.424, and the 2000s so far is .493?

  34. Mark T
    Posted Aug 8, 2007 at 6:43 PM | Permalink | Reply

    One does not discard old data sets just because there are now new “adjusted” ones.

    Given that these results are from calculations that were in error, yes, we do. Their presence serves no point.

    We have refence libraries full of “old” data, invaluable for comparative analysis.

    There’s a difference between old data and data that was created in error. The error in this case created meaningless data, with no value for comparison. Certainly studies that used such erroneous data should now be reassessed, however.

    We didn’t erase Newton’s calculations after Einstein demonstrated that they were partially flawed and “updated” our understanding of the universe.

    Newton’s calculations still hold for less than relativistic velocities, and hence, serve as a good approximation to what really happens in a lab. The analogy here is thin, at best.

    Mark

  35. Geoff Olynyk
    Posted Aug 8, 2007 at 6:48 PM | Permalink | Reply

    I’m still pretty amazed by this. This is going to be huge in the media. Think of how many studies are invalidated by this! Can someone who has access to WebOfScience, etc. do a citation search for one of the Hansen/Karl GISS temperature studies to see how many dependencies there are?

  36. gdn
    Posted Aug 8, 2007 at 6:55 PM | Permalink | Reply

    …and from the 19-teens to the 1930s, the trend was +.356/decade, but from 1970s average to the average of the 2000s is +.158/decade?

    o.k., the 2000s have fewer years, and this mucks with the numbers…but still…

  37. gdn
    Posted Aug 8, 2007 at 6:57 PM | Permalink | Reply

    …and from the 19-teens to the 1930s, the trend was +.356/decade, but from 1970s average to the average of the 2000s is +.158/decade?

    o.k., the 2000s have fewer years, and this mucks with the numbers…but still

    Correction: …and from the 19-teens to the 1930s, the trend was +.356/decade, but from 1980s average to the average of the 2000s is +.158/decade?

    o.k., the 2000s have fewer years, and this mucks with the numbers…but still

  38. aurbo
    Posted Aug 8, 2007 at 7:07 PM | Permalink | Reply

    Re #19:

    The link you requested showing the revised GISS Annual Mean Temperature Graph is here.

  39. JerryB
    Posted Aug 8, 2007 at 7:10 PM | Permalink | Reply

    I would like to encourage caution regarding the recent adjustments to the adjustments.

    They may be replaced by other adjustments at any time that the folks at GISS my decide to do so.

    The recent changes to the adjustments seem to me to be a rather dubious attempt
    to reconcile incompatibilities between using some USHCN adjustments for pre-2000
    data, and using raw USHCN data for subsequent years.

    Meanwhile, we can enjoy watching the show while avoiding taking it too seriously.

  40. tetris
    Posted Aug 8, 2007 at 7:20 PM | Permalink | Reply

    Re: 35
    Mark T
    The question I have, as I know a good number of others do too, is: was this data which you argue was “created in error” in fact created for a purpose? Therein lies the adder, because if so, the legal implications and political ramifiactions are of consequence. Your other points all well taken.

  41. Willis Eschenbach
    Posted Aug 8, 2007 at 8:14 PM | Permalink | Reply

    JerryB, thanks for your post and your links to old GISS versions. Here’s a graph of the post-1970 changes in the four versions you linked to:

    Note the difference between the trend in the June 2000 “data” and the current “data” …

    w.

  42. bernie
    Posted Aug 8, 2007 at 8:27 PM | Permalink | Reply

    Steve:
    My guess is that your main point remains that the data is being adjusted by unspecified and unpublicized means leaving them still unreplicable. What I admire most about your work on this site is your commitment to the fundamentals of good science. Hopefully, this latest set of adjustments will trigger a greater degree of openness than has hitherto been evident.

  43. TCO
    Posted Aug 8, 2007 at 8:41 PM | Permalink | Reply

    35: While, the old data should not be on the front page and should be clearly labelled as in error, it is desirable not to delete it as mixed compilations or papers that have used the data set may be examined and it is useful to have what they worked off of.

  44. Tom T
    Posted Aug 8, 2007 at 8:44 PM | Permalink | Reply

    bigcitylib: You keep this up your going to have to call yourself a denier.

  45. STAFFAN LINDSTRÖM
    Posted Aug 8, 2007 at 8:46 PM | Permalink | Reply

    ALL…
    I⳶e been browsing through Nasa-Giss for some years and
    I would say that the rest of the world is even worse,
    perhaps much worse…If we get some more Pinatubos
    or Mt Helens … the port of Lisbon will once more freeze over
    It snowed there though Jan 29 2006 BUT did not settle
    like it did last time Feb 2, 1954…Cold in Europe
    warm in US at least annually…1987 warm US very cold
    year in all of Sweden…the latest one …Yes and I want
    to mention when I started browsing Nasa Giss in August
    of 2005 they had already the annual mean for 2005!!! Anybody else
    remembering seen/saved that?? Just that and you get a
    little suspicious regarding methods …Just tops
    of icebergs melting yet…

  46. Ron Cram
    Posted Aug 8, 2007 at 8:46 PM | Permalink | Reply

    Steve,
    Congrats again on this discovery. It seems you have your pick of peer reviewed journals to publish this story in now. I hope you pick one with a large readership.

  47. TCO
    Posted Aug 8, 2007 at 8:51 PM | Permalink | Reply

    It should just be a simple note. May even be best handled in the NASA paper already being worked on. None of the speculation, Cram-cackling, gloating, etc. Mosh-pit hits the right note.

  48. tetris
    Posted Aug 8, 2007 at 8:52 PM | Permalink | Reply

    Re: 44
    TCO
    Speaks to my point about “reference libraries”.

  49. TCO
    Posted Aug 8, 2007 at 8:56 PM | Permalink | Reply

    4/Steve: I think this post “problems with” should be more neutral. You don’t know if they have problems or not. Perhaps they’re underestimating warming even. I think when you are careful, you keep more caveats in, but every now and then the mask drops…

  50. Warwick
    Posted Aug 8, 2007 at 8:56 PM | Permalink | Reply

    Why have you gone all cutesy and presented this really interesting and important data as a golf game?

    For a start that guarantees those that neither know nor care about golf can’t follow the article with any sense of understanding.

    This has always been a great website with well presented factual information, okay often a bit too scientific for most, so why have you suddenly dropped into this golf analogy style?

    Have you fallen for the trap of trying to hide something or are you so proud of it that it deserved a better staus than normal (with the glee of overconfidence).

    Work of this significance needs a better standard of reporting, not some golf mumbo jumbo – leave that to a commentator to do in their blog for their audience that might understand it.

  51. Ron Cram
    Posted Aug 8, 2007 at 9:31 PM | Permalink | Reply

    Steve,
    I like your writing style. I’m sorry Warwick does not, but I hope you continue to write exactly as you do.

    Also, may I suggest you consent to an interview on Hannity and Colmes? I think Hannity would like to discuss the implications of your discovery and what other subjects you are investigating.

  52. Jonathan Baxter
    Posted Aug 8, 2007 at 9:38 PM | Permalink | Reply

    It should just be a simple note. May even be best handled in the NASA paper already being worked on. None of the speculation, Cram-cackling, gloating, etc.

    Oh sure. After years of condescension from the team at realclimate, I think we’re entitled to a little payback.

    1998: the warmest year on record
    1998: the warmest year on record
    1998: the warmest year on record
    1998: the warmest year on record
    1998: the warmest year on record

    OOPS!

    1934: the warmest year on record

    Doesn’t quite have the same scare-value, does it?

  53. Steve McIntyre
    Posted Aug 8, 2007 at 9:46 PM | Permalink | Reply

    One of the amusing things about this is why the rankings of 1934 and 1998 should be reversed after they are both long in the clubhouse? Why should the correction of an erroneous change from USHCN adjusted data up to 2000 to GHCN raw data after 2000 affect their rankings? Doncha think that such software needs a little cross-examination?

  54. Tom T
    Posted Aug 8, 2007 at 9:50 PM | Permalink | Reply

    50, What should he say? If the china and India sites are not accurate that is a problem no matter which way the bias is. 51, golf has no place if this is to be publish in a peer reviewed article. But this is a blog the same standards don’t apply.

  55. TCO
    Posted Aug 8, 2007 at 9:54 PM | Permalink | Reply

    Steve, Hannity is a moron. Everyone who knew him in Atlanta is amazed he is rich now. I guess he hits the average idiot level well.

  56. Jeff C.
    Posted Aug 8, 2007 at 9:54 PM | Permalink | Reply

    #51

    Just a hunch (could be wrong) but “Warwick” is not the true Warwick who has done so much admirable work in this area.

  57. Ron Cram
    Posted Aug 8, 2007 at 9:56 PM | Permalink | Reply

    re: 54
    Steve,
    This is exactly the same issue I was commenting on in #2 above. It looks like GISS knew they had other errors and decided to correct them at the same time they corrected the error you found. Perhaps you will have an opportunity to ask Jim Hansen about it sometime soon.

  58. Harold Vance
    Posted Aug 8, 2007 at 9:57 PM | Permalink | Reply

    When do we get to see the source code that GISS is using? If their work is publicly funded and having a major impact on public policy, we as taxpayers should have a right to access and review their code.

    This black box stuff has got to go the way of the dinosaur.

    I’ve been an application developer for ages and understand how easy it is to make mistakes, so I’m not interested in pointing fingers. However, the code they are using needs to be subject to a public audit and review process.

    Kudos to Steve and Anthony and to all of the other inquiring minds on this blog. Fascinating reading material.

  59. Ron Cram
    Posted Aug 8, 2007 at 9:58 PM | Permalink | Reply

    re: 56
    TCO,

    Are you suggesting Steve turn down an invitation from Hannity if one is forthcoming?

  60. Barclay E. MacDonald
    Posted Aug 8, 2007 at 10:01 PM | Permalink | Reply

    Wow! So Al Gore was right when he said “All of our economies will be hurt if the Year 2000 Problem is not solved in time.” http://bogota.usembassy.gov/wwwsagni.shtml. Thanks to Steve M. we are finally making some progress.

    Anthony Watts’ data collection may in fact be a positive endeavor leading to valuable information. It’s lookin’ pretty good today.

  61. Al
    Posted Aug 8, 2007 at 10:01 PM | Permalink | Reply

    Every single adjustment after 1972 is flat or decreasing. Even ignoring the post-2000 data that’s odd.

  62. TCO
    Posted Aug 8, 2007 at 10:11 PM | Permalink | Reply

    Steve needs to get his head straight, write the published papers, hand me the R code that he is holding out on, do an honest assessment of overall HS impact (not NA PC1) from decentering alone, etc. Those all first.

  63. Evan Jones
    Posted Aug 8, 2007 at 10:20 PM | Permalink | Reply

    “When do we get to see the source code that GISS is using? If their work is publicly funded and having a major impact on public policy, we as taxpayers should have a right to access and review their code.”

    To say nothing of scientific method!

  64. Dave Dardinger
    Posted Aug 8, 2007 at 10:24 PM | Permalink | Reply

    re: #51 Warwick,

    Why have you gone all cutesy and presented this really interesting and important data as a golf game?

    Well, for one thing, in open play, it’s always possible for someone to come back at the end. Particularly when there’s someone close to the leader. You mentioned “trap”. Well, landing in a trap can easily blow a lead. Not that I’m a golfer but both my brothers are. And coming from Jack Nicklaus’ hometown, I followed it. Right now I’d say the warmers are a stroke or two up but hit their tee shot in the deep rough on #15.

  65. J. Connor
    Posted Aug 8, 2007 at 10:26 PM | Permalink | Reply

    Interesting times. I wonder how this new adjustment will square with the observed solar activity. Wasn’t that discounted due to recent decreasing solar output not matching a continued rise in observed temperatures?

    In any event I am looking forward to some interesting conversations in light of this, congratulations.

  66. tetris
    Posted Aug 8, 2007 at 10:34 PM | Permalink | Reply

    Re:63
    TCO
    Out of curiosity, ever asked Hanssen and Co or anyone else that side of the fence for the same type of standards, by way of “those all first”? Assuming you did, ever get them to play? If so, I’m sure many of us would like to see. Straight head, cricket, the goose and the gander, and all of that, you know?

  67. Ron Cram
    Posted Aug 8, 2007 at 10:47 PM | Permalink | Reply

    re:63

    TCO,
    What is this nonsense about Steve withholding code? I do not buy it for a minute. And you want an honest assessment of the HS? You’ve got to be kidding, right? Steve has written more than enough on that subject. The Hockey Stick is broken. It cannot be taped back together.

    Not only are the 90s not the warmest decade of the last thousand years, they are not even the warmest decade in the 20th century… at least in North America. And as Pielke and Steve say, the warming bias outside the U.S. is probably greater than inside.

    TCO, why do you want to derail Steve’s efforts so much? What you are afraid of?

  68. Gerald Machnee
    Posted Aug 8, 2007 at 11:31 PM | Permalink | Reply

    It now seems “plausible” that there may have been a miscalculation. We can blame it on the computer. I saw a new model of a hockey stick in a sports store – it has a bend in the middle like a shovel handle or those curling brooms so us old timers do not have to bend over.

  69. paul graham
    Posted Aug 8, 2007 at 11:55 PM | Permalink | Reply

    Fantastic work, i’ve been reading the blog for months now, and i’m always impressed by the quality of you work. I just hope that the media won’t take too long to digest this information.

  70. Harry Eagar
    Posted Aug 9, 2007 at 12:15 AM | Permalink | Reply

    As somebody who doesn’t do computer code but does read history, I gotta say that learning that 1934 is a contendah is not a surprise.

  71. Mark T
    Posted Aug 9, 2007 at 12:26 AM | Permalink | Reply

    Yeah, Grapes of Wrath and all that… there was a reason, ya know.

    Mark

  72. Warwick
    Posted Aug 9, 2007 at 1:00 AM | Permalink | Reply

    #55
    Yes, I know about blog standards or lack of but this article is way out of context with everything else here that reeks of credibility

    #57
    Never done any work in the area, just a very interested observer who gets lost in the science here so usually rely on someone else’s interpretation. Just disappointed by lack of scientific approach – more like a style adopted by warmists.

    #65
    I think they are setting about building their own course and trying to get others to play on it. Me, I think the entry fee is way too excessive!

  73. Alan Woods
    Posted Aug 9, 2007 at 1:16 AM | Permalink | Reply

    #80

    Ron Cram, TCO is ‘trying to derail’ Steve’s efforts. He just has different ideas about what Steve’s priorities should be. Sure, he’s opinionated about it (and annoying), but viva la difference I say.

  74. Posted Aug 9, 2007 at 1:21 AM | Permalink | Reply

    Does the Giss set use the Folland and Parker bucket correction?

    JF

  75. PaddikJ
    Posted Aug 9, 2007 at 1:46 AM | Permalink | Reply

    Just posted the following at Pielke Sr’s site:

    Steve,

    First, congratulations. I’ve been a lurker at your site for about 6 months now and am very impressed at how you and other contributors are willing to wade in and get your hands dirty with raw data (if you’ll pardon the slightly mangled metaphor). Your patience and hard work have paid off handsomely for everyone.

    Now my question, regarding data and source code: NASA/GISS is a publicly funded entity, yes? Isn’t there some law(s) that states that everything obtained by publicly funded research belongs to the public? Does anyone out there know? Maybe the uber-wonk, Peilke Jr.? If the tradition of scientific glasnost is decaying, this could be the needed lever.

    Keep up the good work; you must be doing something right to have the AGW evangelicals so aroused.

    PS: I notice that many posters put a pithy saying under their closing, and as I’m starting to post comments now and again, I thought I should have one too; so I naturally turned to my hero, Richard Feynman. I finally managed to glean just 2, but since I can’t bear to give one up, I’ll post both:

    “Science: Belief in the ignorance of experts.”

    “What is not surrounded by uncertainty cannot be the truth.”

    I will add only that I think it’s been more like 9 mos since I started lurking here, I made a small contribution early on, and now I’m going to make another.

    Thanks for all your good work, all of you.

  76. PaddikJ
    Posted Aug 9, 2007 at 1:48 AM | Permalink | Reply

    Umm . . . that didn’t come out the way I expected. Could someone point me to a quick reference guide to using html tags?

  77. Nate
    Posted Aug 9, 2007 at 1:50 AM | Permalink | Reply

    Congrats Steve and Anthony on this initial finding. Having read many of your posts this does resemble one of the stylistic approaches you take from time to time.
    I would suspect once you have had more time to look at the data we will see a more serious post with more quantitative analysis. While changing which year was the warmest changes little for me on a statistical point of view, sadly many have been scared into believing in AGW based on the populist claims made by Gore (and others) that are proving to be ill informed (at least considering Gore seemed so confident that he’d bet his entire possessions that the claims are objectively true). What does catch my eye is the warming trend since 1970 has been cut in half. This will surely throw off the computer models quite a bit.
    I just hope you are able to set a good example out of this. Get a paper published, while providing full disclosure of the data and methodologies. Teach those guys over at realclimate how to follow the scientific method.

  78. rafa
    Posted Aug 9, 2007 at 2:04 AM | Permalink | Reply

    Dear Steve, very good job. Congratulations. If I understand well this is US data only. Maybe people promoting the CO2->T causal relationship will have to do some homework since I’ve been told (I ignore if is true) CO2 emission levels did fall in the US immediately after the 1929 crash. Maybe Ross can check it quickly.

    best

  79. John A
    Posted Aug 9, 2007 at 2:13 AM | Permalink | Reply

    I can confirm that the Warwick in #57 amd #74 is not Warwick Hughes, but another Warwick. This has caused confusion amongst some old timers but I’m sure its not done deliberately by anyone. WH doesn’t have a trademark on the name.

  80. Andrew Dodds
    Posted Aug 9, 2007 at 2:50 AM | Permalink | Reply

    Steve -

    Do we have any error bars on those temperatures?

    Rafa -

    CO2 is well mixed in the atmosphere, so any change in US emissions in the 1930s would not be noticed. In any case, the current understanding is that we don’t have much of an AGW signal prior to the 1970s.

  81. Warwick
    Posted Aug 9, 2007 at 2:53 AM | Permalink | Reply

    #81
    Cheers John A, I shall breathe again and return to lurking…

  82. Lurker
    Posted Aug 9, 2007 at 3:06 AM | Permalink | Reply

    GISS source code

    http://www.giss.nasa.gov/tools/modelE/

  83. Posted Aug 9, 2007 at 3:06 AM | Permalink | Reply

    If you continue to force the powers that be to revise surface temp trends, the case will eventually be made that you are validating the GCM’s. For the first time the ratio of tropospheric trends to surface trends will match the model predictions.

  84. rudiratlos
    Posted Aug 9, 2007 at 3:47 AM | Permalink | Reply

    i dont understand, is this official, i mean will this be the official data, that for instance the NOAA will use

    because that would mean a BIG victory for the sceptics,

    Great news

  85. Willis Eschenbach
    Posted Aug 9, 2007 at 4:14 AM | Permalink | Reply

    Lurker, you say:

    GISS source code

    http://www.giss.nasa.gov/tools/modelE/

    That is the source code for the GISS GCM computer model. What Steve M. is trying to obtain is the source code for the GISS temperature “data”, which is a very different thing.

    All the best,

    w.

  86. EW
    Posted Aug 9, 2007 at 4:27 AM | Permalink | Reply

    From Web of Science – citations of GISS papers 10 yrs back:

    Hansen J, Sato M, Ruedy R, et al.
    Global temperature change
    PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 103 (39): 14288-14293 SEP 26 2006
    Times Cited: 11

    Schmidt GA, Ruedy R, Hansen JE, et al.
    Present-day atmospheric simulations using GISS ModelE: Comparison to in situ, satellite, and reanalysis data
    JOURNAL OF CLIMATE 19 (2): 153-192 JAN 2006
    Times Cited: 49

    Hansen J, Sato M, Ruedy R, et al.
    Efficacy of climate forcings
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 110 (D18): Art. No. D18104 SEP 28 2005
    Times Cited: 38

    5. Santer BD, Wigley TML, Mears C, et al.
    Amplification of surface temperature trends and variability in the tropical atmosphere
    SCIENCE 309 (5740): 1551-1556 SEP 2 2005
    Times Cited: 35

    Oman L, Robock A, Stenchikov G, et al.
    Climatic response to high-latitude volcanic eruptions
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 110 (D13): Art. No. D13103 JUL 1 2005
    Times Cited: 6

    Hansen J, Nazarenko L, Ruedy R, et al.
    Earth’s energy imbalance: Confirmation and implications
    SCIENCE 308 (5727): 1431-1435 JUN 3 2005
    Times Cited: 64

    Santer BD, Sausen R, Wigley TML, et al.
    Behavior of tropopause height and atmospheric temperature in models, reanalyses, and observations: Decadal changes
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 108 (D1): Art. No. 4002 JAN 3 2003
    Times Cited: 22

    Hansen J, Sato M, Nazarenko L, et al.
    Climate forcings in Goddard Institute for Space Studies SI2000 simulations
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 107 (D18): Art. No. 4347 SEP 2002
    Times Cited: 101

    Hansen J, Ruedy R, Sato M, et al.
    A closer look at United States and global surface temperature change
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 106 (D20): 23947-23963 OCT 27 2001
    Times Cited: 102

    Russell GL, Miller JR, Rind D, et al.
    Comparison of model and observed regional temperature changes during the past 40 years
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 105 (D11): 14891-14898 JUN 16 2000
    Times Cited: 34

    Hansen J, Ruedy R, Glascoe J, et al.
    GISS analysis of surface temperature change
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 104 (D24): 30997-31022 DEC 27 1999
    Times Cited: 182

    Hansen JE, Sato M, Lacis A, et al.
    Climate forcings in the Industrial era
    PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 95 (22): 12753-12758 OCT 27 1998
    Times Cited: 150

    Hansen J, Sato M, Ruedy R, et al.
    Forcings and chaos in interannual to decadal climate change
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 102 (D22): 25679-25720 NOV 27 1997
    Times Cited: 108

    Hansen J, Sato M, Ruedy R
    Radiative forcing and climate response
    JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES 102 (D6): 6831-6864 MAR 27 1997
    Times Cited: 382

  87. Hans Erren
    Posted Aug 9, 2007 at 4:29 AM | Permalink | Reply

    re 80:
    http://cdiac.esd.ornl.gov/trends/emis/tre_usa.htm

  88. MarkW
    Posted Aug 9, 2007 at 4:46 AM | Permalink | Reply

    That was quite a temporal anomaly.

  89. Posted Aug 9, 2007 at 4:48 AM | Permalink | Reply

    Wasn’t quite clear from the earlier post: did the error simply take place in 2000, or is Steve speculating that it is actually a result of the “Y2K” bug?

  90. bernie
    Posted Aug 9, 2007 at 5:13 AM | Permalink | Reply

    The silence on RealClimate is deafening!

  91. bernie
    Posted Aug 9, 2007 at 5:15 AM | Permalink | Reply

    Steve:
    Perhaps you can repeat your most recent Pielke post here, adding who and where to write. Strike while the iron is hot and all that.

  92. bernie
    Posted Aug 9, 2007 at 5:21 AM | Permalink | Reply

    #42
    Willis
    Nobody has commented on your graph. Are you saying that despite these adjustments, the GISS data is showing an increase in the decade over decade temperature increase?

  93. JerryB
    Posted Aug 9, 2007 at 5:51 AM | Permalink | Reply

    Lets distinguish between what Steve has done from GISS’ response.

    Steve found a problem with how GISS was handling pre 2000 USHCN data
    relative to post 1999 USHCN data.

    The reponse by GISS, in addition to acknowledging, and describing the
    problem further, was to back out a portion of USHCN adjustments from
    pre 2000 data. GISS could have, and still could, decide instead to
    add post 1999 USHCN adjustments. If they had, this thread would not
    exist. If they do, this thread will seem odd at best.

  94. Bill Tarver
    Posted Aug 9, 2007 at 6:47 AM | Permalink | Reply

    [snip - religious words not allowed]

  95. Posted Aug 9, 2007 at 6:48 AM | Permalink | Reply

    Starting at comment #110, Gavin at RC is commenting on Steve’s corrections to GISS.

  96. Posted Aug 9, 2007 at 7:07 AM | Permalink | Reply

    Re: Bill’s comment

    Bill,

    Cannot you understand that it’s a big deal that 1930s were officially the hottest decade in the US on record? I mean, nearly 80 years later after pumping CO2 into the atmosphere like crazy, the 30s still top all? How can that possibly be given the precision with which climate scientists predict calamity?

    We’ve been hearing ad nauseum about 0.6 degrees C of warming over the last century. Now that we’re finding out just how dodgy the data is, maybe climate scientists should stop asserting so confidently that temps have rising a fraction of a degree over a hundred years.

  97. BarryW
    Posted Aug 9, 2007 at 7:27 AM | Permalink | Reply

    Re 98

    Oh, only a small change? Look at the graph in comment #42. If that’s correct Steve has managed to cut the trend in the US by almost a third with just one “minor” correction.

    If there are problems with the US data adjustments do you really think that other parts of the world are even close to ours in terms of reliability of measurements? What calculations are being used to correct those sites? I see many more True Believers in the AGW crowd than those of us who follow this site. For example, your post.

  98. Posted Aug 9, 2007 at 7:34 AM | Permalink | Reply

    Folland and Parker’s bucket correction adds .3 dec C to the SSTs in the ’30s. If this correction is incorporated into GISS then it is easy to re-establish the heat primacy of the last decade. Remove the ‘correction’. However, the correction is there for the purpose of making the models work, is it not?

    I know I have a vested interest in removing the bucket correction — it confuses the data from 39 to 45 — but I’m not sure that trying to re-establish a sensible correlation with CO2 levels is a valid reason for tinkering with the past willy-nilly.

    JF

  99. Steve McIntyre
    Posted Aug 9, 2007 at 7:38 AM | Permalink | Reply

    Gavin says that this has “nothing” to do with station quality problems. That’s not true. Defenders of abysmal quality sites argued that Hansen’s software could sort out bad data (sort of like Mann’s principal components, I guess.) Hansen’s software remains undisclosed, but it was obviously unequal to the challenge of identifying inconsistent splices. So I remain unconvinced that HAnsen or USHCN software can sort out bad data.

    Also as far as the global data goes – the big question is why the US data with its high proportion of rural data – has a negligible trend, while the rest of the world has a very strong trend. Has the growth of Chinese and Indonesian cities been properly allowed for? (Actually GISS makes no allowance for it.)

  100. Kenneth Fritsch
    Posted Aug 9, 2007 at 7:42 AM | Permalink | Reply

    The reponse by GISS, in addition to acknowledging, and describing the
    problem further, was to back out a portion of USHCN adjustments from
    pre 2000 data. GISS could have, and still could, decide instead to
    add post 1999 USHCN adjustments. If they had, this thread would not
    exist. If they do, this thread will seem odd at best.

    JerryB, it no doubt is me, but I sometimes have a difficult time following your points. I think many of the posters here might not have the details to fill in what you are saying. More details in these posts would be great for those of us here to learn — recognizing that I might be slower in that department than others.

  101. Posted Aug 9, 2007 at 7:46 AM | Permalink | Reply

    Steve, use some of the money I just put in the tip jar to feed Anthony Watts a nice dinner next time you see him, eh?

  102. JerryB
    Posted Aug 9, 2007 at 8:06 AM | Permalink | Reply

    Kenneth,

    See the thread “Quantifying the Hansen Y2K Error” at
    http://www.climateaudit.org/?p=1868
    including Reto Ruedy’s note to Steve, and my comments
    56 and 57 in that thread.

    GISS took an approach which eliminated the bump between pre 2000
    and and post 1999 data, but they could take a different approach which
    would leave the old “leaderboard” intact.

  103. Mark T.
    Posted Aug 9, 2007 at 9:29 AM | Permalink | Reply

    Re: #41

    The question I have, as I know a good number of others do too, is: was this data which you argue was “created in error” in fact created for a purpose?

    That’s a completely different fish to fry, and the fact that the auditors have the data, old and new, is likely sufficient. I’m sure GISS still has copies of the old data/software as well. No need for them to advertise that fact on their own webpage, however.

    That said, I’m a firm believer that most of the “agenda” I (and many others) sometimes gripe about is most probably confirmation bias wrapped up with a bit of ego. Even the most ethical scientist feels threatened when his work is challenged and once he’s on the train, it’s hard to get off.

    In other news, it looks as if the blogosphere has caught this one. It will be hard to ignore.

    Mark

  104. Posted Aug 10, 2007 at 1:57 AM | Permalink | Reply

    Has anyone attempted to go through the office of the President of the United States to compel NASA to release the data and software? If you can show that you have first attempted to go through lower channels without success, I wouldn’t think there would be anything wrong with going up the chain. Explain why it is so important and document what you have done to get the information in more conventional ways and basically petition the President to compel NASA to make the information public.

    You aren’t asking for a reprimand or passing judgment on anyone, you are simply asking for data and software. I believe the President could get this done with the stoke of a pen. And if a letter doesn’t work, the White House spokesman, Tony Snow, seems to be pretty accessible to people including bloggers. Framing the question along the likes of “shouldn’t the data that we are being asked to modify our lives around and the programs used to process that data be open and accessible so that others may check and recheck it in order to make sure it is correct before we invest so much time, effort and money into the conclusions drawn from them?”

    It just seems that having transparency of that sort would be a good idea going forward no matter WHO is in office in the future. It would be completely apolitical. It’s just data and software.

  105. Posted Aug 10, 2007 at 2:04 AM | Permalink | Reply

    #103. Gavin also states that although the US is a large proportion of the data it is only 2% of world land area and so the problem has a minimal effect globally. But by having the bulk of the data, the US is the best sampled, and possibly the most reliable area to judge change over decadal scales. So the US trends carry more importance than a simple measure of land area would indicate. Change in the trend in 2% of the land area affects the confidence in the global trend more than 2%.

  106. Posted Aug 10, 2007 at 2:08 AM | Permalink | Reply

    What if other networks in other parts of the world use the same process that Hansen uses? What if NASA shares that with them or collects their raw data, processes it, and gives it back? Does anyone know if anyone else might use NASA’s process?

  107. Posted Aug 10, 2007 at 2:09 AM | Permalink | Reply

    Given the current political “climate”, it would be political suicide for the President’s office to even appear to give aid and comfort to anyone perceived as a part of the “denier” camp.

  108. Peter Bickle
    Posted Aug 10, 2007 at 2:10 AM | Permalink | Reply

    Hi all

    Great work Steve. No doubt the beleivers are working out their response, after consulting their comrads.
    Would be good to see what other countries data is like.

    Regards
    Peter Bickle

  109. Posted Aug 10, 2007 at 2:13 AM | Permalink | Reply

    Jeez:

    It wouldn’t have to be aimed at anyone at all. Simply order all climate data and software to be publicly available. That’s it. No need to name names, it would be a matter of policy.

  110. cce
    Posted Aug 10, 2007 at 3:02 AM | Permalink | Reply

    The correction was due to the transition to a different dataset for the US. It has no effect on any other measurements anywhere in the world.

  111. PaddikJ
    Posted Aug 10, 2007 at 3:13 AM | Permalink | Reply

    #69:

    Gore’s highly publicised conspiracy theory twaddle of yesterday.

    Did I miss something? Could you fill me in?

  112. TAC
    Posted Aug 10, 2007 at 3:39 AM | Permalink | Reply

    SteveM (#103),

    Also as far as the global data goes – the big question is why the US data with its high proportion of rural data – has a negligible trend, while the rest of the world has a very strong trend.

    Is there a plausible physical explanation for this, specifically one that does not include a term related to (unrecognized) UHI?

    Assuming the USHCN/GISS data are now OK [an unrealistic assumption, but for now...], we have some explaining to do: The 1930s is now the decade with the highest number of peaks (not the same as the warmest decade, but still interesting), followed by the 1990s, with only one top-10 appearance from the 2000s (so far), the 1920s, and the 1950s. At present, the U.S. temperature anomaly record seems at odds with both AGW theory (that CO2 should cause substantial global warming) as well as with the non-U.S. temperature data.

    On the other hand, Antarctica seems to have been cooling for decades. It makes me wonder about SteveM’s question: Perhaps temperature data coming from the not-very-urban penguin homeland and the not-very-urban U.S. have something in common?

  113. Posted Aug 10, 2007 at 3:39 AM | Permalink | Reply

    “What if other networks in other parts of the world use the same process that Hansen uses? What if NASA shares that with them or collects their raw data, processes it, and gives it back? Does anyone know if anyone else might use NASA’s process?”

    From what I understand not even NOAA uses NASA data. 2005 is ranked as warmest year by NASA, but not by some other records indicated by NOAA. My guess is that there is even less correspondence between NASA and other stations, although it’s always possible that there is better.

    …………

    More to the point, I feel that this discovery, while interesting, isn’t the breakthrough that everyone thinks it is. Errors likely exist in all datasets, whether in climate science or other fields. Errors also exist in understanding data. I think that some here are simply echoing a skeptical argument that has been at the heart of all science since its inception. Even Darwin spoke of bad data. And remember: corrections in errors can skew the record not just down, but also up, and there is absolutely no way to predict what future corrections will indicate: they could easily make the trend steeper.

    I don’t deny that information audits should be completed. But given the fact that errors are detected anyway through audits that are completed anyway, what real difference would adding another auditor make? 1 or 2 errors and corrections that have no overall impact?

    What’s interesting is that some here choose, on the one hand, to doubt NASA’s ability to compile and process data, yet on the other hand derive that very diagnosis from the fact that NASA accepted that an error existed. Question the validator, and then use it to help validate? Kudos.

    Anyway, congrats on the finding.

    Justin

  114. Bill Tarver
    Posted Aug 10, 2007 at 4:30 AM | Permalink | Reply

    “Cannot you understand that it’s a big deal that 1930s were officially the hottest decade in the US on record?”

    In the US, perhaps. But what about GLOBALLY. As I said before, there is a world beyond your borders. Regional effects, while interesting, do not make for global trends.

    “If there are problems with the US data adjustments do you really think that other parts of the world are even close to ours in terms of reliability of measurements?”

    So people in Europe, Asia, India, Africa etc.etc (i.e. the rest of the world that isn’t covered by the USA) are incapable of making temperature measurements? Why shouldn’t they be MORE reliable? After all, one of (or just about the only) bugbears on this site is the poor quality of temperature measurements in the US, chiefly owing to bad positioning. Fair point. BUT that doesn’t alter the fact that the trend in global temperatures is upward and that the climate models point to AGW as the culprit. Retro-tested, the models explain the data very closely.

    “I see many more True Believers in the AGW crowd than those of us who follow this site.”

    I’m not a believer: I follow where the evidence and the science indicate. Until 18 months ago I was as skeptical as you. But there comes a point when the evidence becomes undeniable and if the science provides a model that works, the conclusion is ineluctable: AGW is a reality.

  115. Don Keiller
    Posted Aug 10, 2007 at 5:03 AM | Permalink | Reply

    Great work, Steve.
    By the way another pillar of AGW is under attack- the “insignificant” role of UHI in contaminating the Global Temperature record.
    See http://www.informath.org/WCWF07a.pdf

    Jones may finally get his come-uppance!

  116. MarkW
    Posted Aug 10, 2007 at 5:46 AM | Permalink | Reply

    Bill,

    You show the typical response of a true believer.

    Claim: The US data is better quality than many other parts of the world.
    Response: What? You don’t believe that anyone else in the world is capable of taking data?

    If you don’t see the disconnect between claim and response, ask nicely and I will try to point it out.

    You state that it doesn’t matter that the data has problems, because the data matches the models.

    Of course it does. The models have been carefully calibrated to match the data.
    Regardless, being able to match bad data is not an attribute I would typically brag on.

    You claim that the evidence convinced you? What evidence was that?

  117. T J Olson
    Posted Aug 10, 2007 at 5:51 AM | Permalink | Reply

    Steven McIntyre-

    Winner of the “Emporers New Clothes Award” for scientific scrutiny.
    Second time.

  118. Posted Aug 10, 2007 at 6:34 PM | Permalink | Reply

    Realclimate has a response. I urge everyone to put away the party hats and be rational about this.

    1934 and All That

    Justin

  119. jae
    Posted Aug 10, 2007 at 8:09 PM | Permalink | Reply

    121: We know.

  120. jae
    Posted Aug 10, 2007 at 8:19 PM | Permalink | Reply

    127: RC has a point, but they actually admitted that Steve Mc is right. Do you think they will now embrace the idea of auditing? And who knows how much “error” exists for the rest of the world’s “Global Average Temperature,” whatever the heck that is.

  121. Al
    Posted Aug 10, 2007 at 9:10 PM | Permalink | Reply

    The fundamental point to take away from this is:

    If you’d open the flipping source to peer review, this would have been fixed 6+ years ago.

  122. bernie
    Posted Aug 10, 2007 at 9:16 PM | Permalink | Reply

    Gavin’s latest missif has two standard GISS graphs – but they use different scales on the y-axis. I clumsily adjusted the scales and what becomes apparent is that the Global temperatures series for the 30s is where the US diverges from the ROW and that this divergence contributes significantly to the apparent trend. Yet much of the world during this period was dramatically unstable. How can these records be reviewed and checked?

  123. Posted Aug 10, 2007 at 10:10 PM | Permalink | Reply

    [off-topic]

  124. Posted Aug 11, 2007 at 9:51 AM | Permalink | Reply

    Bernie,

    Some one has checked some of the records and has found problems:

    http://powerandcontrol.blogspot.com/2007/08/problems-in-asia.html

  125. Posted Aug 11, 2007 at 1:41 PM | Permalink | Reply

    #120 Justin,

    There is always the possibility that the next correction will swamp the “heating” data. Or a number of “insignificant” changes will all add up the “wrong” way.

    You prevent propagation of errors by correcting them as soon as possible. In the review stage is best. However, immediately after publication of methods, data, and conclusions (which ought to happen simultaneously) would be good. Just as in industry. The farther down the chain the errors propagate the more costly they are to fix.

    No scientific conclusion reached on the basis secrets (keeping the data and methods hidden) can be considered reliable.

    Peer review is just not up to the task of doing what 10,000 interested and 50 qualified amateurs can do.

    IMO we are actually entering a golden age of science. The beginning of a transformation of how science will be done.

    I may be retired. I have not quit. [snip]

  126. bernie
    Posted Aug 11, 2007 at 6:30 PM | Permalink | Reply

    #133
    That is a good reference. Doug Keenan has a powerful way of presenting an argument. He is also gracious to others, except when they are not straightforward. I can just imagine his calls to Jones and Wang. The link you provided leads to only a foreshortened 4 page PDF copy of his article. I am looking for the complete version now.

  127. bernie
    Posted Aug 11, 2007 at 6:48 PM | Permalink | Reply

    #133
    M. Simon (A first name would make life easier or at least friendlier)
    I made a mistake it is a four pager – but ends in a kind of abrupt way. More to the point, he raises the same point that I noted elsewhere on the suposed continuity of records for large areas of East Asia. In the 30s and 40s it was war with the Japanese and the among the warlords and in the 60s it was Mao’s revolutionary guard. The whole China record for these periods needs to be examined very very carefully or discarded. The more recent record is likely subject to pretty extreme UHI, micro-site and Keenan’s favorite – location shifts as China experiences tremendous urban growth.
    I would certainlylike to see the record sans the Chinese data.

  128. Posted Aug 11, 2007 at 10:02 PM | Permalink | Reply

    bernie,

    My friends call me Simon. The M. is to honor my dad who signed that way.

    Let me repost something here that I just put up that goes to the heart of the matter:

    http://powerandcontrol.blogspot.com/2007/08/default-interpretation.html

    I was reading the comments at Coyote Blog since I just finished a bit about a climate article they had put up and I came across this little gem by dearieme posted Aug 9, 2007 12:15:51 PM:

    “Government scientists ..refuse to publicly release their temperature adjustment algorithms or software”: the default interpretation of that is that they are crooks.

    It seems like a lot of people are coming to that conclusion.

    ===

    It looks like it has gone beyond simple mistakes and errors of judgment.

    BTW the 1221 US stations – the quality stations – are 1/2 the global network. Taking out the Chinese and Russian data will leave big gaps in the coverage.

  129. bernie
    Posted Aug 12, 2007 at 6:39 AM | Permalink | Reply

    Simon:
    The number of US stations impacts only the precison of the US data. My understanding is that the Global figures are based on 5 degree long. x 5 degree lat. blocks for the entire world. Poor or non-existent records for a particular area means a huge amount of interpolation and estimation or simply bad numbers for numerous of estimated 1600 odd land blocks. (30% x 72 x 72) North America represent some 16% of these and the US 6 %. Someone who understands the actual specifics can provide more of the details.

  130. Posted Aug 12, 2007 at 11:43 AM | Permalink | Reply

    bernie,

    I understand your point.

    However, it is acknowledged that the USA has the “best” system in the world. We are discovering that the USA network as a whole is at best “fair” in terms of quality. We know there are problems with Russia and China. How do you suppose things are in Africa? South America? Where there are far fewer “quality” stations? None of which have yet been audited, to my knowledge.

    The only way out is a release of all data and methods, plus a proper survey of all stations used in the models.

  131. Dr. Veritatis
    Posted Aug 16, 2007 at 9:14 PM | Permalink | Reply

    Isn’t 1954 tied for 10th?

  132. bernie
    Posted Aug 16, 2007 at 9:46 PM | Permalink | Reply

    There was some serious work being done earlier examining records in CHina. With Keenan’s article is it worthwhile revisiting that to see if the temperature record for the 30s is as robust as Gavin et al would have us believe?

  133. uchicago
    Posted Aug 18, 2007 at 6:50 PM | Permalink | Reply

    It is my understanding that the US is 6% of the world land area.

    United States — Area (Land): 9,161,923 SQ KM
    World — Area (Land): 148.94 Million SQ KM

    https://www.cia.gov/library/publications/the-world-factbook/

  134. uchicago
    Posted Aug 18, 2007 at 7:32 PM | Permalink | Reply

    The US is about 1.8% of the world surface area.

    United States — Area (Land): 9,161,923 SQ KM
    Earth — Surface Area: 510,065,600 km2

    According to http://en.wikipedia.org/wiki/Earth

  135. bernie
    Posted Aug 18, 2007 at 7:48 PM | Permalink | Reply

    #143
    One issue is the nature and scope of the corrections. These are, of course, limited to the US which is as you say a small % of the surface and land mass. However, this issue puts into question how records are kept and checked in both the US and ROW. Second, the fact that the 30s were brought into the spotlight as a warm decade raises additiona questions as to the integrity of the record for the ROW during that decade – a period of widespread turmoil. While it is certainly possible that the warmth of the 30s was purely a North American phenomenon, it would be unwise to assume that was the case without closer scrutiny of the temperature records. Given the paucity of well kept records outside of North America, Europe and selected other countries, there has to be a significant amount of “adjustments” and interpolations.
    The finding of the GISS errors has some secondary effects beyond merely the correction of the US records from 2000.

  136. BarryW
    Posted Aug 19, 2007 at 11:09 AM | Permalink | Reply

    Re #144

    It should be noted that what we are seeing is not continuous records from one reporting station but the splicing of different records to make what appears to be a continuous record. Instrumentation differs, different units (Kelvin vs Celsius), locations differ. They look at the data and make guesses as to how to correct the data, even where they have no documentation as to what might have happened to the site. A general description of the changes that are made and reference to papers generally describing the process are provided. If Hansen can’t catch a data error (spliced data from two different sources not a Y2K error) that’s been there for six years in the US data, how can anyone have confidence in anything he says? I thought his QA was supposed to catch those errors?

  137. bernie
    Posted Aug 19, 2007 at 10:01 PM | Permalink | Reply

    Barry:
    Your point about the quality of the data is well taken.
    I just analyzed the population data for 37 GISS Brazilian station (see Brazil thread). To put it mildly, the population numbers used to designate stations as rural, suburban and urban are hopelessly out of date. Yet it would take but an intern to keep these updated. Hansen should probably spend a bit more time checking the quality of his data rather than politiking.

  138. henry
    Posted Aug 22, 2007 at 1:26 PM | Permalink | Reply

    Recently, I also looked at the “Contiguous 48 U.S. Surface Air Temperature Anomaly (C) chart, and have a few quick questions:

    1. How accurate is a chart listing the 48 Contiguous states temp, when we didn’t get 48 states till 1912?, True, some of the states were long time territories before statehood, but when the chart starts in 1880, when we had 38 states, is that representitive of the whole U.S? In other words did expanding land usage (depletion of forests) cause some problems?

    2. When did we start including Alaska and Hawaii (or do we) in this listing? If we don’t, how are those temp stations included in the world averages?

    3. In 1934 we had an extreme drought (the “dust bowl”). If 1998 was only off by .02 degrees (a statistical dead heat), why no corresponding “disaster”. Is it possible that the socio-economic world conditions (after the depression, etc) may have led to some problems (burning more wood because of costs, etc?)

  139. henry
    Posted Aug 23, 2007 at 11:04 AM | Permalink | Reply

    Re #137:

    BTW the 1221 US stations – the quality stations – are 1/2 the global network. Taking out the Chinese and Russian data will leave big gaps in the coverage.

    Is this right? That 1/2 of the global network is being used to monitor 2% of the total land area, and STILL only has a small effect on the global averages?

    Somebody please try someting – remove ALL of the US records, and see what that does to the global averages by year. See if it’s really us or not…

  140. JerryB
    Posted Sep 15, 2007 at 10:06 AM | Permalink | Reply

    A new, new, leaderboard has arisen as a result of the change of USHCN input file
    at GISS.

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

    mentions the change of the file, and

    http://data.giss.nasa.gov/gistemp/graphs/Fig.D.txt

    has the new GISS estimates of US contiguous 48 temperature “anomalies” by year.

  141. JerryB
    Posted Sep 15, 2007 at 10:42 AM | Permalink | Reply

    It looks like 1934 and 1998 are now tied, and 1921, and 2006 have swapped places.

  142. BarryW
    Posted Sep 15, 2007 at 10:42 AM | Permalink | Reply

    Re #150
    New top ten US

    Year Temp
    1934 1.24 (tie)
    1998 1.24 (tie)
    2006 1.15
    1921 1.13
    1931 1.08
    1999 0.94
    1953 0.9
    1990 0.88
    1938 0.85
    1939 0.84

  143. _Jim
    Posted Sep 15, 2007 at 11:15 AM | Permalink | Reply

    Can I say it – I have to say it (if no one else will) – “It looks like ‘the fix’ is finally in”.

  144. JerryB
    Posted Sep 15, 2007 at 4:31 PM | Permalink | Reply

    One caveat regarding the statement at GISS:

    “We switched to the current version of USHCN data set which includes data through
    2005.”

    They switched to the edition at the CDIAC server. There is an edition at the NCDC
    server which includes 2006 data. Except that it does not include the 2006 data,
    the edition at the CDIAC server is a duplicate of the edition at the NCDC server.

    I use the word edition, instead of the word version, in order to avoid confusion
    with the NCDC terminology which regards the various editions of USHCN as all
    being version 1, and the forthcoming edition as being version 2.

  145. Steve McIntyre
    Posted Sep 15, 2007 at 5:06 PM | Permalink | Reply

    This new leaderboard is really something else. I’m going to post on this: but if the SHAP version was what they used for the past decade, it’s a little – shall we say – “convenient” to decide in Sept 2007 that they are going to switch to the FILNET version (without announcing it on their website) and then, surprise, surprise, 1998 is now tied for the warmest year. This is going to send shivers up the spine of any readers familiar with accounting principles.

  146. Larry
    Posted Sep 15, 2007 at 5:46 PM | Permalink | Reply

    It would appear that after telling the world over at RC that the relative positions of 1934 and 1998 didn’t mean anything, that it matters a lot to someone at NASA. Some considerable effort went into this achievement.

  147. jorb
    Posted Apr 25, 2008 at 12:10 PM | Permalink | Reply

    I’m a little confused Steve. The details at the GISS link you provide do NOT match your leaderboard. They still show 1998 as hotter than 1934 and the numbers dont match. Has Hansen pulled a “swifty”?

  148. Steve McIntyre
    Posted Apr 25, 2008 at 1:30 PM | Permalink | Reply

    #158. The leaderboard changed again in September. If you track through the category – Surface Record: Hansen in September, you’ll see that they changed their originating data set to a different version in September and – surprise, surprise – 1934 was no longer at the top of the leaderboard. (BTW they didn’t announce the change in methodology at the time so it was rather intriguing trying to keep up with GISS machinations.)

  149. Aaron
    Posted Jun 11, 2009 at 10:35 AM | Permalink | Reply

    I know I am not the smartest guy on here but can anyone answer this for me? The reranked data from 2000-2006 here, and now 1998 is less hot then 1934? Can someone explain to me how data not touched changed?

  150. Posted Dec 22, 2009 at 12:49 PM | Permalink | Reply

    Very interesting points in your blog it sure cleared many questions that i may have had. I loved it because i did learn a lot from it because of it.

  151. Bruce Michael Albert, Ph.D. PDRA
    Posted Feb 26, 2010 at 9:45 AM | Permalink | Reply

    Dear Sirs,

    I have been following the matter of the US temp. data for some time. It came up in a book review I did for Quaternary Science Reviews (I stated that the 1930′s was probably the hottest decade of the 20th Cent., circum Dust-Bowl). How OFFICIAL is this view now (can you reiterate this point in email)? It’s not a question of belief, it’s just that I don’t think this is officially admitted by NASA yet (in Euro. and Japan, with LESS relative UHI differential through time, the 1930′s also seems to be hotter than the 1990′s).

    Bruce M. Albert, Ph.D., Leverhulme post-doc PDRA

  152. Steve M. from TN
    Posted Feb 26, 2010 at 12:31 PM | Permalink | Reply

    while only 3 of the top 10 are from the last 10 years (1998, 2006, 1999)

    Very generous of you to include 1998 and 1999 in the last 10 years. 2000-2009 would be the last 10 years, and only 1 year (2006) falls in that category.

  153. Steve M. from TN
    Posted Feb 26, 2010 at 12:36 PM | Permalink | Reply

    never mind my last post..I didn’t look at the original date for your post

  154. John Ryan
    Posted Feb 28, 2010 at 5:15 PM | Permalink | Reply

    Use NOAA data they are more accurate and the official temp keepers

  155. Posted Aug 13, 2010 at 12:26 PM | Permalink | Reply

    All these temperature measurements are sort of red herrings, if for no other reason than that they are subject to human errors. Look for changes in ecologies that indicate warming (or cooling – but there are none of those) temperatures. Migration patterns of flora and fauna. (Flora move up mountains until they get to the top and die, or they move north) Changes in time of migration. Changes in sex ratios of some amphibians. Continuously shrinking ice sheets and glaciers. Earlier melt times and later freeze times of large bodies of water and rivers. Average these over decades.

    And before some ignoramus points to more ice and snow in the antarctic, let me say that that is evidence of global warming, not cooling. This is also true of glaciers that grow in areas that, despite general warming, still seldom get above freezing. Find a climate expert to explain it to you. It has to do with the fact that a warmer antarctic or mountains are still well below freezing almost the the entire year, and the earth is still not warm enough for it to rain there.

    Nature will not be fooled.

    • Mooloo
      Posted Jul 24, 2012 at 5:40 AM | Permalink | Reply

      All these temperature measurements are sort of red herrings, if for no other reason than that they are subject to human errors. Look for changes in ecologies that indicate warming

      Why would measurements of ecology be less prone to error than simple measurement of temperature? I would suggest that determining the boundary point of a species is a terrible guide, prone to all sorts of wishful thinking.

      Mangrove swamps are a warm water phenomenon. They are declining world-wide. Are we to take this as evidence that the water is getting colder?

      No doubt we can find some area where warm temperature species are expanding. Especially if we don’t look too hard for the spots where the opposite is true.

      • transrp
        Posted Jul 25, 2012 at 3:23 AM | Permalink | Reply

        The complete scientific illiteracy here is scary. Did not a single person here bother to look at the actual data?? Did no one notice the column on the right? Surely given your right wing tendencies, you would have looked at the right hand column. The one that shows 5 year averages? And while the 30′s peaked at .58, EVERY single 5 year period after 1998 was over .6 with one exception, 2009.

        But wait… It gets worse. I actually looked at the individual years from the link that you gave.
        years yrly mean 5 year mean
        1998 1.3040 0.6268
        2006 1.2890 0.7784
        1934 1.2210 0.4088
        1921 1.1000 0.1202
        1999 1.0650 0.8234

        It is like 911 “truthers” or creationists. All of the referenced facts are lies. You just make things up. Note above that years 3&4 are not so special when you examine 5 year mean.

        Of course, given your scientific illiteracy, and willful blindness, one should not expect you to see the obvious. As to mangrove swamps shrinking. What is your point? With rising waters and more severe weather, what did you expect? But they are not moving south to get away from the cooling are they. ON the other hand, almost every other species on the planet that can relocate, including plants is either moving north or up. Perhaps you can find one that is not.

25 Trackbacks

  1. [...] hard work of Anthony Watts (www.surfacestations.org) and of Steve McIntyre (Climate Audit) has resulted in the identification of a significant error in the assessment of the rankings of [...]

  2. [...] Jules Crittenden and Small Dead Animals, this reordering of the universe—or at least our particular ball of it: Four of the top 10 are now from the 1930s: 1934, 1931, [...]

  3. [...] http://www.climateaudit.org/?p=1880 [...]

  4. [...] The example started with this, which led to this, and leading finally to this. [...]

  5. [...] auch sonst scheint es Anfang des letzten Jahrhunderts ziemlich warm gewesen zu sein, so dass die Liste der wärmsten Jahre des letzten Jhs. umgeschrieben werden musste. Zusande gekommen sind die falschen Annahmen [...]

  6. By » Blog Archive » This is very BIG news on Aug 9, 2007 at 9:06 AM

    [...] Source. [...]

  7. By Outside The Beltway | OTB on Aug 10, 2007 at 5:42 AM

    Y2K Bug Causes Global Cooling…

    Anthony Watts as generated a huge amount of blogospheric attention with his report on Steve McIntyre’s recalculation of temperature trends. It turns out that the Y2K bug skewed previous data from this decade.
    Four of the top 10 years of US CONUS…

  8. [...] of various sorts, including highly-browed calls for censorship by esteemed Professors. Steve McIntyre, the blogger/statistician that has recently discovered a bug in the software used by …, has seen his website crushed by an apparent DOS attack just hours [...]

  9. By Feeling the heat at Hoystory on Sep 24, 2007 at 1:58 AM

    [...] Hansen has done some further rejiggering to his methods and now 1931 and 1998 are tied. [...]

  10. [...] http://www.climateaudit.org/?p=1880 [...]

  11. [...] pointed out some errors and the “official” temperature data was adjusted again to show 1934 as the high point. In fact, only 3 of the top ten dates, from that set of data, are from the current [...]

  12. [...] the early afternoon of August 8 (14:51 Eastern), I wrote a short post on changes in the “leaderboard”. This short and simple post attracted a lot of [...]

  13. [...] you see, in reality Four of the top 10 are now from the 1930s: 1934, 1931, 1938 and 1939, while only 3 of the top 10 [...]

  14. [...] 1980s and 1990s which has since peaked, says the World Meteorological Organisation.Well, you see, in realityFour of the top 10 are now from the 1930s: 1934, 1931, 1938 and 1939, while only 3 of the top 10 are [...]

  15. [...] you see, in reality Four of the top 10 are now from the 1930s: 1934, 1931, 1938 and 1939, while only 3 of the top 10 [...]

  16. [...] A New Leaderboard at the U.S. Open « Climate Audit Four of the top 10 are now from the 1930s: 1934, 1931, 1938 and 1939, while only 3 of the top 10 are from the last 10 years (1998, 2006, 1999). Several years (2000, 2002, 2003, 2004) fell well down the leaderboard, behind even 1900. [...]

  17. [...] 8 2007 So why did you post that wrong link with the 2007 table???? Where is the correct link??? A New Leaderboard at the U.S. Open Climate Audit comment 31 I suggest that you read the article as well. __________________ if you stop blaming [...]

  18. [...] changes in the U.S. history were not brought to the attention of readers by GISS itself, but in this post at climateaudit. As a result of the GISS revisions, there was a change in the “leader [...]

  19. [...] But I note, looking back through NYTimes archives, that the 1930s had peek blooms in March sometimes, and sometimes in May during the bitter cold 1970s. I note that “cherry picker” scientists are careful not to go back to the earlier warm cycle, when the US reached its warmest year (1934), a record that was only adjusted out of existence after several years of tinkering. [...]

  20. [...] data NASA corrected were data for the US and Steve McIntyre was forthright in reporting this (shown in right column). In any case, the corrections were small [...]

  21. [...] data NASA corrected were data for the US and Steve McIntyre was forthright in reporting this (shown in right column). In any case, the corrections were small [...]

  22. [...]  http://climateaudit.org/2007/08/08/a-new-leaderboard-at-the-us-open/ [...]

  23. […] http://climateaudit.org/2007/08/08/a-new-leaderboard-at-the-us-open/ […]

  24. […] prediction came true quite quickly. On Sept 15, Jerry Brennan observed that the NASA U.S. temperature history had changed and that 1998 was now co-leader atop the U.S. […]

  25. By XKCDetails | DeHavelle.com on Mar 5, 2014 at 5:02 AM

    […] This is not a one-time adjustment; sometimes several times per year they adjust history over again. It took six years of adjustments before they finally got 1998 to be warmer than 1934, for example, and the difference between those two years is now even larger than it was in 2007. In […]

Post a Comment

Required fields are marked *

*
*

Follow

Get every new post delivered to your Inbox.

Join 2,875 other followers

%d bloggers like this: