For some reason, Michael Tobis seems to think that the Team is busy cranking out responses to an endless stream of my data requests. Nothing could be further from the truth. For me, getting data from the Team is no easier than posting at realclimate. At this point, it seems to be Team policy to simply not provide any data to me. So it doesn’t take them any time at all.
I racked my brains trying to think of whether I got any data from the Team last year (and by Team here, I mean the core Hockey Team, not the IPCC). The only thing that I could think of was the list of stations used in Jones et al 1990, which was eventually provided under a UK FOI request. I encountered an interesting little back story about this request today.
I reported on these exchanges last year, commencing with my initial request to Phil Jones on Feb 22, 2007, their first refusal and my concurrent request for re-consideration on the grounds of unresponsiveness on March 12, 2007, their partial acquiescence on April 3, 2007, my follow-up on the part where they remained unresponsiveness and their final refusal.. Immediately on receipt of this information, I wrote some interesting posts on Chinese stations here here here . Doug Keenan followed up on this information as well.
I was browsing NCAR sites this morning in an attempt to find out what they were ( a “federally funded research and development center” turns out to be a sui generis class of institution.) I could hardly resist looking at a presentation entitled:
Tom Karl & Larry Tyminski
Co-Chairs: NOAA Data Management Committee
NOAA Science Advisory Board’s Data Archiving and Access Requirements Working Group (DAARWG)
May 24-25, 2007 Chicago, IL
Skipping through this presentation, on page 13, to my surprise, I found the following slide, containing a request in a style that was not unrecognizable:
On March 12, 2007, after CRU had stonewalled my initial request for the identity of the Jones et al 1990 stations, in addition to requesting that they re-consider this matter, I had also sent an FOI request to NOAA in the form shown in the Karl presentation. Interestingly, I never received any acknowledgement from NOAA on this FOI request at all. Since CRU re-considered and provided the requested information on April 3, 2007, I didn’t follow up with NOAA or appeal with them.
So even though they never answered my request in any way whatever, in May 2007, a couple of months later, my request was being used in NOAA presentations on data archiving.
It continues. On Aug 23, 2007, the Data Archiving and Access Requirements Working Group (DAARWG) issued a report entitled Report to the Science Advisory Board National Oceanic and Atmospheric Administration, containing the following slides:
Amused by this, I sent the following letter today to Thomas Karl (forwarding my oridinal request of March 12, 2007):
Dear Dr Karl,
I notice that the following FOI request was discussed in two recent NOAA presentations:Tom Karl & Larry Tyminski, Co-Chairs: NOAA Data Management Committee. NOAA Science Advisory Board’s Data Archiving and Access Requirements Working Group (DAARWG), May 24-25, 2007
Data Archiving and Access Requirements Working Group (DAARWG), Report to the Science Advisory Board National Oceanic and Atmospheric Administration, 23 August, 2007
May I observe a certain irony in respect to the use of this FOI request in your presentations. I did not actually receive any reply from NOAA to this FOI request, not even an acknowledgement. The situation later became moot when the U.K. authors re-considered their prior refusal to provide the information. If NOAA had a hand in this, I’d like to express my appreciation, but I’d still like to note for the record, that the FOI request referred to in these presentations never received a reply.
Regards, Steve McIntyre
NOAA archives an extraordinary amount of data. In paleoclimate, NOAA maintains the excellent World Data Center for Paleoclimatology, a facility that I try to support at every opportunity. And yet how often do we hear paleoclimate scientists whine about how expensive it would be to archive data? All they have to do is email their data to WDCP, who look after everything else in a professional manner. NOAA’s been very courteous to Anthony Watts’ project.
Although NOAA never responded to my FOI request, I’m wondering if maybe they whispered into CRU’s ear and maybe that’s why CRU re-considered their prior refusal. Just wondering. (See follow-up to this here.)
Climate Audit 
34 Comments
Sounds like a hint of progress. Get yourself invited to the conversation.
Remember radio’s Paul Harvey? Sometimes we do get to hear…..”The rest of the story” in real life. Too funny. If they are reconsidering their archiving policy after one FOI request, then I guess Steve demonstrated that this is a valid issue to consider. Great work!
Well, at least you set off some alarm bells.
I’d say, so much for those claiming Steve’s work has no impact and no benefit.
Paragraph three has HTML issues for me.
Why would Tom Karl feature an FOIA request that was neither acknowledged nor answered? Do they receive so few FOIA requests that all TK has as an example is one they never served? It’s not likely that TK anticipated that you’d happen on his presentation, Steve, and so there should have been no ulterior twit Steve M motive in choosing your FOIA request from the dozens to hundreds of such requests for data just surely received by NOAA every year.
One can only conclude that NOAA receives very few FOIA requests, and yours was featured because it was the only one. On the other hand, whether to archive given data should not depend on the number of requests for them. It should depend, for a government lab certainly, on how important are the data for policy decisions. In general, data should be archived as evidence for the validity of past conclusions, and to facilitate future work. Journal articles are a form of archiving, for example, and I have had repeated need to go back into quite old (60 years old) journal articles to read, especially materials and methods sections, to see how past workers treated their materials or synthesized their molecules. Raw data and methods should probably be archived for a good 100 years. One never knows when they’ll be useful again, or for what.
On another note, how would anyone at NOAA even know that you had requested data from CRU unless someone at CRU had told them? If there was email traffic concerning your requests, Steve, then that might imply that you’re upsetting an international equanimity. Perturbations in the global equanimity field — a whole new arena for teleconnection science.
I think that we know that the Team has exchanged my IP address as I’ve been blocked from Team sites e.g. U of Arizona. I can’t imagine how Hughes would get my IP address other than via Mann and/or realclimate. We are not talking about ordinary pettiness here.
They don’t keep the raw data? They changed the original dataset and it’s no longer available?
What, are they keeping this in an Excel workbook or something? The NOAA doesn’t have a database or a document imagining system to archive data? How convenient.
I’m not saying this is any type of fraud, but it sure it mismanagement. But since NOAA is a government agency, I don’t expect any repercussions.
#8. The Jones et al 1990 example is probably not a good one in several respects, as things have changed quite a bit since 1990. Plus Karl was an ancillary author, with the calcs probably done at CRU. Also in fairness people the issue of versions is something that people think about more.
Also I don’t really want to get into NOAA as a type case for the situations that have plagued me as they manage huge amounts of data and deliver a LOT of data online. CRU is very different, Thompson is different.
So I’d appreciate it if people would simply view this as an interesting story and not get excited about NOAA.
They have been very courteous to Anthony Watts’ project because he has photos that are compelling to the average Joe and make the NOAA look like complete incompetent fools.
It is always worth asking for a review of an FOI request if there is anything slightly odd.
OT here, but I recently had a case where I asked for copies of documents related to a decision not to provide a bus service; four months late I did not receive any documents but instead a discursive reply saying that there would have been insufficient demand. The problem was that I already had a copy of a statement hinting that their models had said that the necessary level of service to meet projected demand would cause serious traffic congestion. The FOI review produced a quicker response and some real documents leading up to the decision, mainly because somebody more professional and disinterested had to conduct it. The irony was that the post-hoc rationalisation was in my view probably more accurate, even though it was not the cause of the decision.
Steve Mc, at least they referred to the request as coming from an “international researcher” instead of the “the entity” or the “depths of Hades”.
There are three plausible reasons for this sort of review:
1. They could be reviewing data retention policies with the intent of preventing skeptics from acquiring and auditing data via FOIA requests.
2. They could be attempting to eliminate the appearance that their existing policies have the effect of preventing skeptics from acquiring data.
3. They could genuinely see the value in retaining old versions of data sets and being able to provide them to the public that they serve.
In this case, I think that #2 and #3 are substantially more plausible than #1.
Steve wanted data to audit an important paper. The data was not available in a form which would allow Steve to replicate the original results. It is perfectly natural to ask whether or not this situation needs to be addressed.
I think that those alarmists who are most hostile to the scientific process and most unwilling to share their data are perfectly capable of making old data sets disappear without producing a readily accessible paper trail. In other words, if the primary objective was #1, we wouldn’t have a presentation in which they openly ask which data retention policies will best allow them to accomplish their mission.
You have the right to appeal this [denial or partial denial] of the FOIA request. Your appeal must be received within 30 calendar days of the date of initial denial letter by the Assistant General Counsel for Administration. Address your appeal to the following office: Assistant General Counsel for Administration (Office)
Room 5898-C
U.S. Department of Commerce
14th and Constitution Avenue, N.W.
Washington, D.C. 20230.
Re: #6
Pat, I suspect that the salient aspect of Steve’s request for the purposes of Karl’s presentation was the age of the data requested. The request is a nice real-world instance of 20-year-old work suddenly in demand again.
Don’t want to be a stick in the mud, but are most countries FOIA laws only for the benefit of citizens of that country?
(Speculating….) Thus non-citizenship might mean that a FOIA request isn’t valid and not even requiring of a response? If that is the case, might one then need to consult any treaties/agreements to see if they extend FOIA coverage?
Pete:
Curiously enough, that’s not the case. Although FOIA legislation are usually framed in terms of the [national] public good, requests for information under FOIA can come from almost anywhere (including Canada).
Certainly there is no reason for NOAA/NASA to refuse the FOIA request on the grounds of Steve’s nationality, especially when there’s the “obstructive jerk” reason.
Steve,
Databases of legal reports normally include signs such as colored flags to indicate if the legal precedent is under challenge, it is a strong law etc. And this is done by private legal services. Considering the potential cost of developing and responding to public policies that are developed based on scientific papers based on “top secret” information, will it not be a public service if those papers are rated. i.e. DAC- (Data availability and codes)-five star extensively posted in the internet including raw data, adjustment and analysis codes four star-all the data are available but no raw data but all the codes three star- adjusted data are available include codes for adjustment and codes for analysis two stars- adjusted data and codes for analysis available etc, TS for top secret. Then an overall index of the paper reliability based on available information to verify and recalculate the findings. 1 star- highly unrelaible , 2 star apply with caution, 3 etc. As Wegman reported for a very specialized field the reviewers are limited and may form a cluster or a network.
#15 — Armand, I have data going back to 1985 on my pc at work, and could provide it in the space of hours. We have data tapes that go back into the 1970’s pretty well documented with labels. There would be a problem with reading them these days (VAX TK50 tapes, if you can believe it), but at least we’ve got the data. All of our data files include a column for raw data, which remains unchanged during processing. So, this whole business of the survival only of processed versions or grey data seems like a bogus argument or else evidence of incompetence.
It’s not really that far back to 1990 in terms of data structure and even in 1990 we had efficient binary to ascii converters, headers and all. Very ordinary data archiving should have been sufficient to make that data recoverable. A place like CRU ought to be well able to update old data files as computer and data storage systems change. Doing so ought to be a matter of standard business at CRU (and NOAA), where data production is a large part of their raison d’être.
Those data in particular keep their value because they provide an extremely useful climate baseline. It’s really hard to imagine they’d let those data slip into disrepair or loss. I just don’t see much of an excuse for non-possession, non-provision, in a claim of age, referring to 1990 data in general, and especially 1990 data that retain current value.
The glaring question: on what basis can anyone assume that their conclusions are correct… when there is no data??
On the basis of their 20-year old memories? On the memories of those long retired/departed?
On their solemn assurances that the results are absolutely correct?
#20. The 18-year old analysis was peer reviewed and of course the data was checked at the time. Or was it?
Well Pat, you can read a TK50 tape in a TK70…
😉
Here is a link to a list of all FFRDCs, which is quite a bit longer than I had remembered even though I used to work for one.
http://www.nsf.gov/statistics/nsf01328/sectb.htm
Re:#19
Pat, I agree, and would think that Karl does as well. I would think he included the example purely to forestall possible objections about “old” data/papers.
The NOAA slide is really interesting. It is unlikely to be related to why CRU changed their minds about releasing the data though. The reason is explained in this comment:
http://www.climateaudit.org/?p=1323#comment-100985
Briefly, I drafted a letter to the U.K. Information Commissioner’s Office, alleging that the university was in violation of statute, and sent the draft to the university, asking them to let me know if they believed the letter to be inaccurate; the data was then released. (The dates and my conversations with the university make that clear.)
On another thread, Steve has posted a reply from Tom Karl
David Jay,
TK70??
Well, I have a newer unit that is read compatible with both if any one needs it! 8>)
If the bureaucrats at NOAA are anything like those at my University, their #1 priority in life is to not be hassled (e.g., they never find students guilty of cheating, regardless of the evidence, because they do not want to be hassled by angry parents).
I would bet Steve’s request, and the response from the purported data-holders that the data did not exist, set off alarm bells. The last thing the bureaucrats want to do is be hassled by Steve about the lack of data, especially if he could appeal to a formal process that would involve lots of paperwork for them.
I am just starting a project at NASA to resurrect 45 year old analog tape drives to digitize original first generation images from a certain series of spacecraft (sorry can’t be more precise right now). There is a HUGE problem with data and data formats older than about 15-20 years now as there is never any money to update these old databases. We just got lucky as these images are in demand right now.
Dennis 29:
Of course it is practically impossible to transfer and re-transfer a steadily growing amount of data to every new generation of storage media that comes along. But then, this is not the way to do it anyhow IMHO. I come from a quite different area of expertise (digitization and restoration of early audio and video recordings), but the problems are similar: To get the best possible results, one always needs to go back to the first-generation “data” (whether digital or analog); each transfer generation invariably accumulates errors of some kind and is never truly better than its source. New methods of signal processing can often extract detail from original sources not even thought to be there even some years ago. All too often however original sources have however been thrown out by archives long ago in favor of “modernized”, adjusted transfers that lack much of the finer details.
It is a mystery to me why (in your case) playback equipment not even half-a-century old is not kept in good working order although an archive of recordings playable only on such equipment exists. Maintenance of such equipment and, where necessary, construction of interfaces to connect it to the currently used machinery, so that every bit of data can be retrieved from its original format whenever needed, would probably cost but a small fraction of constantly re-transferring (and thereby involuntarily degrading) the recorded data to new formats every other year or so.
ChrisZ,
“It is a mystery to me why (in your case) playback equipment not even half-a-century old is not kept in good working order although an archive of recordings playable only on such equipment exists.”
I don’t believe it’s such a mystery because these things require someone to budget for it and for that budget submit to make it up to the funding approval folks. Unless you have a specific tangible need like Dennis, it’s too hard to make a compelling case. Especially when the case may involve a statement like; “we don’t know exactly what we might need it for, but it’s good practice so that we’ll be ready quickly if the need does arise.”
Never mind that that the cost is probably higher than what the approvers notion of a reasonable cost is. I can just see the thoughts in their heads: $500 for some old equipment off e-Bay, 3 days to get it working, 2 days to get the old data, 3 days to run it through the free internet download data conversion application, 2 days to wrap it up and 2 days to give us a brief.
That’s 2 people at 12 days @ $120/hr, $500 equipment. That’s about $23.5K How hard is this?
Request approved, but just do this within your existing budget.
It is VERY hard to keep 60’s era analog instrumentation tape drives in operation for 45 years. We are going to have to replace all of the capacitors and probably all of the germanium transistors with modern versions. I don’t think that most people today realize how touchy and maintanence intensive that era’s equipment is. Also, in this case, there is only one remaining read head company in the world that refurbishes the helical scan air bearing rotating heads. When they quit the business in a year or so the cost for refurbishment goes up by two orders of magnitude.
Just about all complex equipment has quirks that the experienced engineers that kept these things working knew by heart but never transferred that data from their brains to paper. We have a retired Ampex engineer working with us but our equipment is unique (was used for a lot of classified work in the 60’s) and the amount of information on this equipment is extremely limited.
John Christy got his start in the climate business when figuring out how to download satellite temperature data from 1979 (he did this in 88 and 89 when I worked down the hall from him) that was already almost impossible to download. A lot of those data set digital formats are peculiar to the machine that originally recorded them and these things get tossed almost as soon as the contract is ended.
I worked on a project in the 80’s to digitize the drawings from the B1- Bomber (5 million drawings) and Rockwell bid the job at almost a quarter of a billion dollars.
Bottom line is that this is NOT easy or cheap. It is way better to update every few years the original data sets rather than waiting a few decades.
Dennis Wingo, feel free to contact me through Steve Mc. I can put you in touch with an out of work tech with Ampex Quad experience if you are interested, or who also may be able to point you to others in the broadcast community who may have some experience you need.
Nevermind, I’m sending an email with contact information to your company.
5 Trackbacks
[…] end of the world predictions on has somehow recently just disappeared. This was after scientists asking for over a decade for it to be released so they could audit it. As a result there is now almost NO data support for […]
[…] my surprise, as I reported at CA here, I saw two NOAA presentations on the internet referring to my FOI request (not identifying me by […]
[…] end of the world predictions on has somehow recently just disappeared. This was after scientists asking for over a decade for it to be released so they could audit it. As a result there is now almost NO data support for […]
[…] “Climate Audit and NOAA FOI Policy“, Steve McIntyre, Climate Audit, 3 July 2008. […]
[…] received the following response today from Tom Karl regarding my 2007 FOI request (See here): As previously indicated in my e-mail last Thursday, NCDC did indeed put together a response to […]