Some Prompt Data Responses

A couple of prompt and favorable responses to inquiries for data from Parker 2006 and Nyberg et al 2007 (a recent article on hurricanes)

Parker 2006
I wrote Parker today inquiring about the sites used in Parker 2006 and for data for unarchived sites. He replied promptly as follows:

Here are the station lists. The “Word” document gives station details and the text documents specify sources.

Regarding the data from national sources, you can download Central England temperature from

I attach also the data for Lerwick and Eskdalemuir, UK. In these files, Jan, Feb…Dec for a given year and day of the month runs along a row. Valuse are in tenths deg C with -999 = missing (generally denoting nonexistent days).

For the other national-source data you will need to apply to the people listed in the Acknowledgements of my 2006 J Climate paper. The Danish Met Service may charge a fee for the Greenland and Thorshavn data.


David Parker

I’ve uploaded the files to and interested parties can examine the lists. Given the policies of WMO and IPCC, it’s ridiculous that unarchived data should be used in these studies, although one cannot hold Parker to fault for this, other than for failing to criticize this practice as an author.

Nyberg et al 2007
Nyberg et al 2007 is an interesting article in Nature using proxies to study long-term lack of trend in hurricanes. I’ll talk about this article at some point. It uses four coral luminescence series which have not been archived and David Black’s percentage G bulloides series from Cariaco (in the last 2 years, Nature has published articles in which percent G bulloides, a proxy that I’ve discussed on a number of occasions in the past, has been used as a proxy for NH temperature, Asian precipitation and Atlantic hurricanes – a versatile proxy indeed.) I wrote to the second author, Bjorn Malmgren (not readily locating Nyberg’s email online) requesting that the coral luminescence data be archived at WDCP and received the following cordial reply (which he authorized me to print):

Dear Steve,

Many thanks for your letter. I do not have access to the luminiscence records myself, they are stored with the senior author of the Nature article, Johan Nyberg, a former Ph. D. student of mine. I will contact him immediately to ask him to send you the data directly or to send them to me after which I can forward them to you.

By the way, I am an avid reader of Climate Audit, so from me you receive a proper response. In fact, I download the articles to my cell phone and read them with great interest every day. Many thanks for so relentlessly contributing these articles to Climate Audit.

Kind regards, Bjorn

Within the next day, he confirmed that the data would be archived at WDCP and also sent me a text file with the data.

So not everyone is as obstinate as Jones, Mann, Bradley etc. BTW if I recall correctly, William Connolley or some such have taken the view that I like to complain about unavailability of data. I don’t. It’s a waste of my time and the obstruction is a waste of the author’s time. Rather than get annoyed about it, my practice, as you know, has been to publicize the obstruction at the blog. Sciencemag responded to this pressure when private emails got nowhere.

But I’d much rather write notes like this where a positive result is obtained. In the Nyberg case, the authors obviously had no objections to archiving data, they just hadn’t done it. It’s too bad that the lax refereeing at Nature doesn’t attend to these things, but that’s another story.


  1. jae
    Posted Jun 14, 2007 at 12:19 PM | Permalink

    I salute these authors! True scientists and gentlemen.

  2. steven mosher
    Posted Jun 14, 2007 at 1:01 PM | Permalink

    Kudos to these guys!

    You know, there is an unseen benefit in opening yur data to CA

    one of two things happen.

    1. there is no issue and you eclipse the Mann.
    2. there is an issue and you get a chance to correct it, and eclipse the mann.

    As a political animal I would demand that CA audit my stuff. everything else
    is hitting girls.

  3. steven mosher
    Posted Jun 14, 2007 at 1:59 PM | Permalink

    Spot check.

    Selected USA sites. Picked site I had no clue about.

    Olympia Wash. GPS and google earth shows a Airport site.
    46 58 -122 54

    that took 1 minute to check. so 300 sites would be… a days work?

    My sense is that the local windspeed at an airport location
    is going to be an interesting animal, depending on where and how
    it is measured. Just a question.

    So, is an Airport location URBAN or RURAL? I dunno. Look at it.

    That is one site. You guys pick another. better yet would be a routine
    that took a list of sites lat lon and created an album of the sat photos the highest


  4. steven mosher
    Posted Jun 14, 2007 at 4:02 PM | Permalink

    Ok this is dumb as dirt. Fourth site picked

    Phoenix 33 26 -112 01 Airport.

    5th site:

    Syracuse: 43 07 -76 07 Airport.

    6th site:

    Grand junction: 39 07 -108 32 Airport.

    OK. Random canadian site. 68 18 -133 29 Inuvik :

    Now you’ll go figure. How were the 290 sites Parker selected?

  5. Steve McIntyre
    Posted Jun 14, 2007 at 4:06 PM | Permalink

    I think that I mentioned before that the GSN network is nearly all urban airports. The dirty little secret about CRU, GISS etc is that they haven’t bothered incorporating any riral information for the past 15 years; they are just parasiting off the urban airport information. One of the reasons why they are so vituperative about the non-existence of UHI is because they are so depednent on urban airport data.

    Parker’s 290 are a subset of this and are mostly airports. How did he chose his subset? I can’t tell so far.

  6. steven mosher
    Posted Jun 14, 2007 at 4:20 PM | Permalink

    ok, more funny things.

    Checked a norway site: on the coast
    Checked a greenland site: on the coast
    Checked a swedish site: on the coast.
    Didnt need to check Inchon Korea

    hmmm well key west was an airport by the ocean!

  7. steven mosher
    Posted Jun 14, 2007 at 4:33 PM | Permalink

    Sorry SteveM I missed that alert.

    I just bought a bunch of lottery tickets figuring that I had some great nack for
    randomly picking airport sites. Wanna buy some? I’ll give you a good deal.

    The nice thing about airport sites is that ordinarly they are selected for a reason.

    Flat. few obstructtions. Predictable prevailing winds.

    I’m not sure it’s a random sampling of places to measure climate.

  8. hswiseman
    Posted Jun 14, 2007 at 11:07 PM | Permalink

    I suspect that Parker saw the ASOS data and felt that these sites had good equipment, consistent TOB, and solid data archives and site histories. In other words, data collected through the practice of some consistent conventions applied over hundreds of sites. Pretty good fodder for the statistical engines. And he is correct. The ASOS sites seem to be producing good data from what I could tell through a casual inspection. That does not mean they are necessarily good sites for observation of GW or AGW. JFK/BOS/PHL/ are 24/7/365 blast furnace operations over square miles of paving. Small airports in reasonable proximity with comparable geography to these majors could be a check point Compare BOS/PLY, JFK/ISP, PHL/TTN for example.

  9. Willis Eschenbach
    Posted Jun 14, 2007 at 11:30 PM | Permalink

    The other thing about airports is that they watch the wind pretty carefully … which makes it odd that he used gridded NCEP/NCAR reanalysis winds.


  10. Steve McIntyre
    Posted Jun 15, 2007 at 6:59 AM | Permalink

    Wheels within wheels. I was unable to identify all of Parker’s sites in any online NOAA version, so I asked him for a URL for his data. Although there are many online data sets, it turns out that he used unarchived data obtained from Ron Ray at NOAA. I’ve written Ray asking him for a copy (or a link) for the data that he sent to Parker.

  11. BradH
    Posted Jun 16, 2007 at 6:23 AM | Permalink

    I’m kind of overcome by the professionalism and openess of these guys.

    Perhaps there is some hope, after all.

  12. MrPete
    Posted Jun 16, 2007 at 8:23 AM | Permalink

    Speaking of data and sources, NOAA maintains a very nice site with map/text search access to quite a few databases, from weather to paleo info. Appears to even provide details on exact location of of tree ring sites.

  13. Steve McIntyre
    Posted Jun 16, 2007 at 8:41 AM | Permalink

    For the paleo data, this is simply a new wrapper for the same data base that’s been operating for many years. I’ve mentioned the WDCP World Data Center for Paleoclimatoloyg or ITRDB International Tree Ring Data Base quite often – these are the archives to which I encourage scientists to constribute. That’s what you’re locating here. It’s an excellent archive and I’ve attempted to commend it on all possible occasions.

  14. Barry
    Posted Jun 16, 2007 at 2:06 PM | Permalink

    RE # 9 . There is a site providing photos and metadata for the ASOS sites that were being used in a study. Provides photos from all of the major compass points.

  15. Steve McIntyre
    Posted Jun 16, 2007 at 2:27 PM | Permalink

    #15. I’ve posted up a couple of “pages” – see left frame Station Data and Station Locations in which I’ve been posting up links as I notice them. I appreciate additions. (This one I’d notice, but there’s lotgs of information and a big job to collate. Right now I’m having trouble getting lists of GSN stations. Or rather I’ve got 3 lists, each inconsistent and none complete)

  16. Steve McIntyre
    Posted Jun 17, 2007 at 8:28 AM | Permalink

    Ron Ray responded that the data was at . This file does not appear to contain any US sites, while the list of sites that Parker obtained from Ron Ray does. I’ve asked Ron Ray the following:

    Thanks for this. David Parker used a number of US stations. When I unpacked, I identified 451 different station numbers in the file, none of which seemed to include Parker’s US stations. I’ve also had difficulty locating a list of GSN stations that synchronizes with the station numbers. I matched lists from, , all of which overlapped, but none of which yielded all the station identifications. Is there a list of stations in the gsndy file with particulars?

    Secondly, I’ve also seen a directory with GSN daily data at . Can you explain what the connection is between the file and the data in the other directory?

    Also, in your directory , I notice that there are two subdirectories rawdata and reformatted . What is the connection of the data in these directories with the file and the data at ghcn?

    Regards, Steve McIntyre

%d bloggers like this: