Friday, September 25, 2015

GWPF wimps out

In April, there was a big story summarized by a headline in the Telegraph: "Top Scientists Start To Examine Fiddled Global Warming Figures" I wrote about it here. The GWPF announced an inquiry into global temperature adjustments. They had assembled a panel of apparently reasonably well qualified scientists. They promulgated terms of reference (which seemed to borrow heavily from the terminology of Paul Homewood). They put out a call for submissions, with a deadline of 30 June. They said "After review by the panel, all submissions will be published and can be examined and commented upon by anyone who is interested." and set up a page for this purpose here. All this promptly echoed, eg here and here.

I immediately thought about putting in a submission, and did in fact write one. I sent it to the prescribed email address on 2 June. No response. So I emailed again to ask if it had been received, on 14 June. Still nothing. So then I wrote to the GWPF general email, and got a prompt and courteous response from none other than Andrew Montford. He said he couldn't find my submission there, so I sent him a copy, which he received. Encouraging.

Still no response from the actual panel though. I kept an eye on the site, especially the submissions page. I thought they might say that they had received (with thanks) x submissions, or some such. But AFAICS, the site didn't change at all.

There was one link on the page that said "news", which was an obvious place to try, but it didn't seem to connect to anything. Three months later, still wondering, I got a helpful direct link from a correspondent, to this page. And there I find, dated July 22, this information:
"The team has decided that its principal output will be peer-reviewed papers rather than a report.
Further announcements will follow in due course."


So...no report! So what happens to the terms of reference? The submissions? How do they interact with "peer-reviewed papers"?

And of course one may ask who (if anyone) will ever write those papers? And about what?

I wonder what changed their minds?

BTW, here is a Wayback snapshot from June. I don't think anything has changed, except for the "news".

39 comments:

  1. The home page still says that all submissions will be published. And going to papers, rather than a report, shouldn't stop them publishing their submissions. So all you can tell from the outside is that they're being slow. You don't know that they won't publish them.

    Oh course we all know that the actual answer is something like one of (a) they have an embarassingly small number of submissions; (b) apart from people like you telling them they are bozos, but politely, they have any number of nutters telling them that the GHE doesn't exist (c) their dog ate them because they're clueless idiots (d) something even less respectable.

    ReplyDelete
    Replies
    1. I can't see that there is any organization actually assembling the submissions. That is based partly on my own experience, but also on the total lack of comment on the site. You'd expect to see something like - thanks to all those 159 people who sent in submissions; the committee is considering them, and we will post them shortly. I expected to get an email saying something like that. But three months after deadline, there is nothing.

      And what is the point of submissions if there is no report?

      Delete
    2. > the point of submissions if there is no report?

      Perhaps they are planning to steal your valuable input and put it into a paper?

      Delete
    3. Dr. Connolley is probably right on some accounts. But I also want to add one more hindrance. Doubters don't have any real funding, it is all activist labour done without a fee. There is no way the pompous declarations could fulfilled. Repeat. There is no real funding, no manpower. If they are right on any account, they can't prove it. Congratulations.

      In a certain manner I find this extremely funny. There has been only little warming during the last 15 years, but the amount of centennial warming during the last 15 years has still increased a lot, thanks to adjustments. And you, Dr. Connolley, have much more power than you ever wished for. The little adjustments to Wikipedia may have a huge impact on what people believe. I really hope you understand your responsibility.

      There is only a limited amount stupidity the Earth can take before the fabric of space-time collapses into singularity. Somebody said 'plastic bag' or 'wind farm' with too much chutzpah, and shbang!


      Delete
  2. Thanks for the update, Nick. I was wondering about this myself, not that I expected anything to come of it. In fact I speculated at the time that it was "nothing but a political stunt by a denier lobby group, to try to get people to doubt that climate change is happening. "

    http://blog.hotwhopper.com/2015/04/denier-weirdness-mock-delegation-from.html

    I didn't expect there to be any report. The GWPF said "No timetable has been set for the panel to report." So I figured that if the panel were to write something that passed GWPF approval, it would release it in time for Paris, or the bits they liked. Otherwise we'd never hear of it again.

    There's still time if they can get a compliant member of the review team.

    ReplyDelete
    Replies
    1. "There's still time if they can get a compliant member of the review team."
      Sou, the panel together does not seem compliant; I think this decision is a major embarrassment for the GWPF, and their relations with the panel must now be poor. A solo rant from a panel member would not have much impact.

      Delete
  3. After calling on all people to send submissions for their "review", the least the GW Policy Foundation can do is to publish these and comment upon them. Whether they then want to write a report of articles about that, that is their choice. But after making so much work for other people, it is indecent not to respond to that. It would naturally fit to the behaviour of their members.

    ReplyDelete
    Replies
    1. Victor,
      I suspect that the number of people who labored to make submissions is actually small. I think the more important aspect is the wimping out. There has been no let-up in the stream of silly articles discovering that adjusted data has been adjusted (eg WUWT here). But when they try to put a more serious case, it just collapses. And the nonsense stream continues.

      This kind of development is somewhat encouraging. Not that Abbott's environment minister is a model of enlightenment, but that he has found it to his advantage to promote himself in this way.

      Delete
    2. If there is a small number of replies it would be even easier for them to reply to the submissions. That would make it even more indecent. People have wasted their precious short time on Earth on this charade, the least the Policy Foundation can do is publish the submissions and respond to them.

      Let's see how small the number is. The mitigation sceptics like to claim that there is no consensus. Thus I am looking forward to all those submissions from scientists in the field that support the Policy Foundation. I would like to read what their quality arguments are.

      I have pointed the Policy Foundation to your post on twitter. Three hours ago. No reply yet.

      Delete
    3. I've also asked them. They've ignored me too. I'm not that surprised, though.

      Delete
  4. If Dr Connolley is to modest to mention it himself. He now also has a post on the lack of feedback from the Policy Foundation. Plus an appraisal of the review team.

    What if you gave a review and nobody came?

    ReplyDelete
  5. I might be persuaded that the submissions are being looked at by 1)Evan Jones and 2)Richard Tol - and they'll be right on it as soon as 1 finishes his project/paper for AW and 2 locates the missing #TOL 300

    ReplyDelete
    Replies
    1. Perhaps they've decided the world will end on Monday so they don't need to bother.

      http://www.mirror.co.uk/news/weird-news/world-end-september-28-blood-6400240

      Delete
  6. > Doubters don't have any real funding, it is all activist labour done without a fee

    That's not entirely true. The GWPF has ~£250k / year funding (or did in 2013 when it last published accounts, IIRC) and was spending £160 k / year on salaries for 4 people at that point.

    ReplyDelete
    Replies
    1. But, but, they have to spend that all on PR which unfortunately leaves no money anymore for science.

      Delete
    2. Compare that to Greenpeace. About $400 million US, right?

      £160 k / year isn't very much for salaries, when you include overhead. Do you know if there is a list of who it's paying and what percentage effort that is?

      Delete
    3. Carrick, I have no idea, but yes they have a lot of popular support.

      At least it is the appropriate comparison the Policy Foundation and Greenpeace are both political activist organisations.

      Delete
    4. Greenpeace probably does more real science.

      Delete
  7. BTW, I've added the exciting temperature review to their wiki page: https://en.wikipedia.org/wiki/Global_Warming_Policy_Foundation#Inquiry_into_temperature_records

    ReplyDelete
  8. All you have to do is look at this graph
    https://wattsupwiththat.files.wordpress.com/2015/09/nino-3-4-ceres-and-noaa-sst2.jpg
    Look at that incredible agreement between the two curves, one a satellite and one a thermometer record.
    The denialists such as woeful willis have no understanding of the hypocrisy in what they post.

    If the data collected matches that exactly, does that mean that there is no measurement noise in the system? Or did he screw up and plot the equivalent of x=x?



    ReplyDelete
    Replies
    1. I was set to scoff at the idea that Eschenbach downloaded the CERES data, extract the upwelling radiant heat flux, back-calculated the surface temperature in K and arrived at perfectly matching data.

      But I don't like to judge unfairly, so I downloaded the CERES data for the Nino 3.4 region from the auxiliary dataset skin temperature, figured out what the temperature field was, brought it into MATLAB along with ERSSTv4 Nino 3.4 and... reproduced his plot perfectly. So, point to Willis.

      However the near-perfect match with the NOAA ERSST data was so unexpected and physically implausible (with 50 1.0 sq degree boxes of different temperatures radiating T^4 and not uniformly sampled over the course of 1 month) that I looked a little farther and found this on the CERES site:

      "No single instrument is suficient for characterizing the complexity of the Earth System. As such, data from various platforms must be combined. Moreover, parameters that can not be directly measured may be useful for such an analysis. In order to provide a comprehensive data set, CERES team put together a variety of Auxiliary Data at the same spatial and temporal resolution as the CERES derived parameters. As CERES algorithms refine - requiring more auxiliary input, and scientific demand increases, CERES team plans on expanding its auxiliary data."

      In other words, they are likely computing the skin temperature in the auxiliary data set using ERSSTv4, so without "screwing up" Eschenbach probably plotted x=x.

      Delete
    2. Correction: that's 500 (50 x 10) one-degree square boxes in the Nino Region 3.4, not 50.

      Delete
    3. Regarding how close surface and satellite tmperatutes can be, check out the Merchant et al. 2012 paper on the ARC surface temperature data. It reports an excellent match between infrared measurements of ocean temps and HadSST3. The authors emphasise that their satellite approach is independent.

      Delete
    4. @ Matter: looks like I have to read both that paper and several other recent ones on the topic quite closely. While I didn't doubt that a radiometer-derived surface temperature accuracy of +/- 0.1 K would be obtainable under ideal conditions the overall correlation with gridded SST measurements is much closer than I would have thought possible.

      Delete
    5. Magma, great work. Time series data that doesn't show a trend is very difficult to correlate. Eyeballing that particular match it had a correlation coefficient well over 0.9. Your skills are needed at the Azimuth Project. We are doing this kind of analysis with respect to QBO and ENSO. https://forum.azimuthproject.org/discussions

      We are getting good correlations with models that only include oscillating angular momentum shifts.

      Delete
  9. You'll have seen their waste-of-time reply, http://www.tempdatareview.org/news/2015/9/29/international-temperature-data-panel-status-update, I think. Including the priceless "the numerous submissions made to the panel by members of the public".

    ReplyDelete
    Replies
    1. Thanks, William, no I hadn't. Link here. There is a bit of a time difference down here. I saw, briefly, your appropiate query/comment, too, asking how many submissions and when they would be published. It seems it was rapidly erased.

      Yes, the update doesn't add anything. Looks like the panel was not going to agree on a report, but someone wants to soldier on and talk about Paraguay. Or maybe it is just bravado.

      Delete
    2. > I saw, briefly, your appropiate query/comment, too, asking how many submissions and when they would be published. It seems it was rapidly erased.

      Ooh, exciting, I didn't see that and don't know what you mean. Where did it appear?

      Delete
    3. It was at the bottom of that page. It was odd - it didn't seem to name an author. At the time I saw it, it was timestamped "Eight minutes ago". It just asked how many submissions were received and when they would be published. Similar to your query to Dr Kealey; maybe someone there had attached part of that to the statement, then thought better of it.

      Delete
    4. That was me. My two civil questions were:

      "How many are "numerous submissions"?

      Any timeline for when these submissions will be published?"

      Seems that these questions were to difficult for the Policy Foundation.

      Delete
  10. "One of our projects is an analysis of a subset of the submissions made to the panel by members of the public which have a common theme of the impact of adjustments on individual station histories. The stations histories provided by the public are from many different geographical regions and from several different adjusted data sets."

    It looks like they are doing a collection in best Goddard or Homewood style under the label "tampering with data", etc.
    I have tested the god will of those charlatans by posting e g GHCN Amundsen-Scott, but examples of "warming the past, cooling the present" don't pass moderation.
    I guess that GWPF will receive a lot of examples of "cooling the past, warming the present", but very few with a contrary view, since sensible people are not cherry-pickers and/or don't bother to do submissions to such organisations.

    It is possible to cherry-pick examples in both directions, but the best evidence for me is to see how adjustments affect an operational global temperature dataset. Nick, You did such a comparison here
    http://www.moyhu.blogspot.com.au/2015/02/homogenisation-makes-little-difference.html#more
    I suggest that you make TempLSmesh adjusted a permanent member of the active graph http://www.moyhu.blogspot.com.au/p/latest-ice-and-temperature-data.html#Drag
    making it possible to follow effects of adjustment up-to-date
    (The old version from Feb 5 demonstrates that GHCN adjustments actually cool the trend from year 2000 by 0.3 C/ century, but TempLS mesh has been updated since then)

    ReplyDelete
    Replies
    1. Olof, where did you find that quote. I cannot find it.

      If they could program a little, they could have thousands of examples of time series with no warming, that are warming after homogenization. No need to ask citizens to look through all the data for a well-defined case, computers are great at that.

      They could also find examples of series cooling more than 5°C before homogenization, but I guess that even their uncritical audience would start wondering if so much cooling is realistic and whether that might not be a data problem that needs to be taken into account.

      Delete
    2. Olof,
      "I suggest that you make TempLSmesh adjusted a permanent member of the active graph "
      Yes, I can do that. I will have to add it to the regular download on a monthly frequency.

      Victor,
      "If they could program a little, they could have thousands of examples of time series with no warming"
      I did a Google Maps exercise here which allows you to filter by size of adjustment. It will tell big positive or negative adjustments by trend, but won't tell whether they turn negative slope into positive. I featured that post in my submission.

      Delete
    3. Thanks Nick, that would be really nice, and useful.
      One day, when GHCN V4/ISTI hopefully solve the problem with the Arctic cooling bias, the TempLSmesh adjusted will surely be the first choice..

      Victor, follow Nick's clickable link above, 1.53 AM

      Delete
    4. Olof, found it. Was not in Google yet.

      With the much larger ISTI station dataset, the erroneous trend reductions in the Arctic should become less in the next version of GHCN.

      Now the statistical assumption is that all stations in a network have the same climate signal. It would be interesting to include more climatological knowledge in this step and allow for smaller differences in the climate signals depending on how climatologically similar you expect the stations to be. If anyone has any idea how to do that. Especially how to formalise how climatologically similar stations are.

      Delete
  11. Nick,

    This is tangential, but I wondered if you had any insight. I was pointed to some wayback versions of the GISS Meteorological Station global average:

    August 2012: https://web.archive.org/web/20120825064307/http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A.txt

    November 2012: https://web.archive.org/web/20121106184504/http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A.txt

    The difference plot between these looks like this and the November version is similar to present. Between these 2012 dates GISS moved from GHCN v3.1 to v3.2 but I was under the impression that the difference between the GHCN datasets was not this big. The change over time also seems oddly linear. Did the GISS difference occur because they removed their own urban adjustments when switching to GHCNv3.2? Or has there been some conceptual change in how the Met Station average is produced?

    ReplyDelete
    Replies
    1. Paul,
      I don't have any insight - I don't follow the Met Stations analysis much. It gives huge weighting to coastal and island stations, which makes it much less stable. As you say, the GHCN change should have been small, and there is nothing else recorded on the GISS updates page.

      Delete
  12. Replies
    1. Thanks, Victor. The story about the allegedly prospective paper for which they become confidential data is a transparent device to avoid publication. It will take many years to find a journal interested in an analysis of such a collection of wisdom. If the submissions are that important, the journals would publish them, not the analysis.

      Delete