the Air Vent

Because the world needs another opinion

SteveM Got a Little Press

Posted by Jeff Id on August 21, 2009

On the 18th Kevin Libin of the National post picked up the HadCrut idiocy.  As a good reporter should he tried to make sens of the claims made at face value.  Its fun to watch him struggle with it because I sure have.

Kevin Libin: You’ll just have to take our word on the global warming stuff

But probably nothing could damage the credibility of climate change believers than the recent revelation by the Climatic Research Unit (CRU) that it has lost or destroyed all the original data used to construct historic global temperature records. The CRU, at the University of East Anglia in the UK, which has been using information collected from weather stations across the globe for decades, is probably the most widely cited source worldwide for those mounting a case that the earth has exhibited an inexorable warming trend: its website boasts that CRU’s research has “set the agenda for the major research effort in, and political preoccupation with, climate research.” The critical raw climate data responsible, which scientists of all climate-creeds have a natural interest in, is now gone, apparently, forever. With the exception of a handful of countries that the CRU has agreements with to sell its data, all that remains for the bulk of the statistics are “value added” versions, which is to say, consolidated, homogenized data. Actually, the CRU says it doesn’t even have all the data for countries it has data-sharing agreements with. “We know that there were others, but cannot locate them, possibly as we’ve moved offices several times during the 1980s,” the CRU writes in a rather embarrassing explanation for all this posted on its website.

It’s good to see the press going after some of this ridiculous covering of data.  It’s very clear that HadCrut has the data, after all they post global averages from it.  Claims they don’t have raw data are untrue in my opinion, they are acting like people with something to hide.  Interesting considering they have the highest warming trend of any temperature dataset.

The real point was picked up in this article though.  Kevin Libin seems quite a bit braver than Nature in exposing the reality of climate science.

Professor Phil Jones, the fellow in charge of maintaining the CRU data set, told an Australian researcher a few years back that he refused to publicly share his statistics. “We have 25 or so years invested in the work. Why should I make the data available to you, when your aim is to try and find something wrong with it?”

Exactly Dr. Jones.   I should mention that without naming any sources there is credible information that discussions between primary publishers in climate science have been had regarding the best means to stall those who want data.  The point is that I know damn well from several avenues these boys don’t wan you to know what they’re doing.  There is at least some active planning in the field behind the scenes to withhold the code and data,  how thorough it is I don’t know but it took a long time for RC to admit that the code wasn’t released for Steig et al (we haven’t got it yet and I’ve grown tired of asking for it anyway) and it also took a long time to get the data (7 months).

Here’s a great SteveM comment:

“We have never had sufficient resources to keep track of the exact source of each individual monthly value,” they explained in coming up dry for the FOI requests. As Stephen McIntyre, the Canadian economist famous for his addiction to poring through volumes of mind-numbing climate statistics, and occasionally finding errors (as he did, with Ross McKitrick, in deconstructing and undermining the famous “hockey stick” graph), writes on his Climate Audit blog, it appears that the impoverished CRU even lacked filing cabinets in which to store its records.

Kevin Libin finished the article nicely with a scientific point that should already have been obvious to the community.

Given that the Unit has admitted now that it cannot fully substantiate its work, it raises the uncomfortable question of whether CRU’s historic climate research should be used any longer at all.

Many of us live with the unproven belief that GISS exaggerates warming trend by some amount.  We know the corrections are so loose that they defy any reasonable engineering explanation yet GISS trends are less than HadCRUT.  Those who are intellectually honest will also question whether the government funding motive is possibly driving the result.  It is a valid question, especially in climate science where quote after quote comes out of the leftist organizations from leaders advocating ends justify means scientific reporting. Greenpeace is the latest example where they actually got caught explaining that their real motive was less prosperity for America and developed nations.  Not a nobody either, the CEO.  It is one of many similar incidents.  We deserve to know the truth, if the leftist government organizations want power through dishonesty, what does that say about motive.

The corrections to both of these datasets are nearly the same magnitude as the signal they proclaim.   What would it mean if we discovered there was no good explanation for the source and corrections to the data? We know the financial motive exists, we know the fame motive exists, we’ve seen endpoint filtering used to manipulate impressions of this data, we’ve seen fake hockey sticks for a variety of BS warming  Is it such a stretch to suspect the data homogenization may not be perfectly Kosher.

We’ve seen the other datasets closely from satellite and ground sources.  From that, I  believe the probability that HadCRUT has not inflated the warming signal through systematic overcorrections as very unlikely.  This is the real problem, just thought I’d say it.  I’m glad SteveM has kept at this and it’s good that one paper picked it up.  This is NOT a small issue just as Greenpeace’s admission of their lies and true intent was not small.  Google greenpeace in the news section, only 16 news stories for this admission, unbelievable.

5 Responses to “SteveM Got a Little Press”

  1. Al S. said

    Jeff, this is somewhat related.
    I have been following E.M Smith’s blog. ( He’s a retired FORTRAN & mainframe guy, who has a flavor of GISStemp code running at home, and has shown a much-reduced warming in the long-record GHCN stations. He believes much of “Global Warming” is a short-term station artifact in the actual GHCN reported data.
    I was hoping to have somebody who’s into data analysis weigh in on what is really going on. The eyeball test can only take me so far…

  2. Kenneth Fritsch said

    Jeff ID, I think it is important to keep in mind what CRU claims about data. I think Libin has it in mind in the version that I do.

    CRU lost the “original” data sometime in the 1980s during a moving of their records. The original data would be that raw data from its sources and its CRU adjustments that goes all the way back to the start of the CRU data series and is a requirement for justifying the “value added” adjustments that CRU has made over the years.

    The current raw station data, going back in time to the lost data of the 1980s, are data that CRU obtains from its sources and then adjusts using the value added data from the 1980s as a basis. It is these data, in my mind, that CRU inadvertently posted on its web site and was withholding from Steve M and others on their requests – and not the original data.

    What CRU has admitted is very damning for an owner of a data series and, in my mind and as Libin suggests, it could very well be fatal to its future use. The lost original data would be the clincher, but admission of the sloppiness of losing the claimed contracts with their sources and the original data compounded with their replies to FOI requests adds significantly to their ownership problems and image.

    I cannot believe an even halfway rational organization would fatally, or at least potentially fatally, wound itself to keep from revealing their raw and original data.

    I also do not think one has to impose or suspect any intentional biasing of the data through adjustments by the owners of temperature data series. What can, and I am not saying that it necessarily does, happen, when a large part of the temperature trend (in the US) can be attributed to adjustment, is that those making the adjustments can have valid reasons for the adjustments made but unfortunately will tend to neglect the need for adjustments in the other direction.

  3. E.M.Smith said

    Many of us live with the unproven belief that GISS exaggerates warming trend by some amount. We know the corrections are so loose that they defy any reasonable engineering explanation yet GISS trends are less than HadCRUT

    It does not need to be an “unproven belief” it can be measured. I’ve started with a benchmark to “characterize the data” that I’m now running through GIStemp. I’ve already found one bit of code that warms the entire data set by about 1/1000 C in one line of code. (It does this by warming 1/10 of the records by 1/100 C).

    Then I looked at the overall effect of the first steps of GIStemp. The “temperature” STEPs up to the point where it makes the temp data into “zones”. The effect on the benchmark are clearly to amplify any “warming signal”:

    There is a specific effect of about 1/2 C that is seen in the “delta” to the benchmark. Note that this is in addition to any warming IN the data, this is just what GIStemp adds.

    There is a reasonable argument to be made that when the thermometer locations are “boxed and gridded” there will be more warming. (N. Hemisphere impact will be reduced, due to lower surface area, while Tropical / Equatorial impact will be amplified due to more surface area.) I’ve not done that analysis yet (still working on it) but in making the benchmark to do this, I discovered that there was a significant impact in the data prior to GIStemp from changes of thermometer locations. This will undoubtedly impact what GIStemp does and GIStemp zones will not catch it, so they ought to amplify some of it (via the “Reference Station Method”):

    While this is a lot of work, and some of it is fairly technical, it is not beyond what any programmer can do. I make the source code available and the data are on public servers, so anyone who wants to do so can check these numbers.

    BTW, my “flavor” of GIStemp is essentially pure GIStemp. The changes I made do not change the product significantly and are mostly housekeeping changes (i.e. put the source code in a secure directory, compile once in a Makefile). It will run on Linux just fine and ought to run on any Unix box as well. You do need both an f77 and f90 era FORTRAN compiler to make it go, but I have pointers to a free f95 (g95) on the site. I’m ready, willing, and able to help folks make their own copy “go” if anyone would like to “do this at home” 😉


  4. E.M.Smith said

    (It does this by warming 1/10 of the records by 1/100 C).

    That ought to have said “1/100 of the records by 1/10 C” … no net effect, but still, I ought to have proofed prior to post…

    OK, a “value add”. Here is how to select the top set of stations (in terms of length of station records) and look at their temperature trends over time. When you do this, you find no warming in these long lived stable thermometers. Drop it from 3000 to the Top Ten Percent, and the data become even more stable (i.e. less change of temperature):

  5. BarryW said

    Don’t know why anyone is surprised. NASA “lost” the high definition video of the moon landing, once of the most important pieces of historical documentation in the 20th century (IMO). Why would you expect anything better from these hobby shoppers.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: