NIWA withholds evidence, loses face

About ten to four this afternoon Outlook went ‘bing-bong’ and a mail from NIWA arrived on my desktop. It was John Morgan, refusing my second request for publicly-owned information. The time elapsed since we first asked for it has reached 21 weeks.

Readers might remember I asked again on 20 February:

So would you please provide a copy of the scientific literature that approves of the measurement technique used by NIWA in the Review. This request is reiterated under the Official Information Act.

Here’s what Mr Morgan decided to say about it:

Dear Mr Treadgold

I write in response to your email dated 20 February 2014 indicating that you believe that my reply of 21 November 2013 did not answer the question raised in your email of 22 October 2013.

I believe that my reply of 21 November 2013 is in accordance with NIWA’s obligations under the Official Information Act 1982.

I remain firmly of the view that the question you have raised with respect to the methodology applied by NIWA in the Review has been thoroughly canvassed before the Court and, being a matter of expert scientific opinion, is not the correct subject matter of an Official Information request.

Section 28(3) of the Official Information Act allows you to complain to the Ombudsman regarding this response.
Regards
John Morgan
Chief Executive Officer

That’s balderdash, of course. He shamelessly corrupts a request for documents into “a matter of expert scientific opinion” which has been “thoroughly canvassed before the Court” and is not the “correct” subject for an OIA request. Stuff and nonsense, John, what are you drinking?

The documents which he brings to our attention because, he claims, they “recognise” NIWA’s methods are just documents, not opinions; they have been mentioned in, but strangely not produced to, the Court (the judge let them get away with that), therefore they are not available to us (so we must ask for them); but because they’re documents, the request is an ideal subject for an OIA request. He’s twisting everything.

So all right, I’m angry, but I replied carefully:

Dear Mr Morgan,

Thank you for your reply, but I’m having trouble understanding it. Perhaps you misunderstand my request, because I’m not asking for a scientific opinion. Please note also that this request is unconnected with any judicial proceedings.

I asked simply for a copy of the scientific literature that approves of the measurement technique used by NIWA in the Review. In other words, a copy of the documents you referred to in the Herald published on 16 October 2013, when you stated that the methodology applied by NIWA in the Review was internationally recognised. I ask for nothing more and nothing less and I have asked for it twice. Confirmation of your statement can only be found in documents describing the international recognition you cite, so nothing else will do; I ask you to produce them because you have not produced them anywhere else.

In the absence of your confirmation, doubts arise as to whether the documents exist. Yet as a prestigious institution steeped in the ways of science, NIWA surely understands the value to credibility of producing evidence.

My readers and I want to see a copy of those documents you cited in which international recognition of NIWA’s methods in the Review is described, or an Internet reference to such documents. As it concerns material whose existence you confirmed in a public forum, our request is an appropriate matter under the Official Information Act 1982. It doesn’t require scientific input. You can answer it easily with some photocopied pages.

Regards,

Richard Treadgold

What are staff in other departments thinking when they see their CEO blundering about with baffling replies and obfuscating our simple request?

There’s not much more I can say. I mean, why should I bother to sabotage a man so intent on sabotaging himself? The more often he intentionally “misconstrues” my request, the more clearly everyone can see his obfuscation. We have to wonder what he’s hiding.

Whatever it is must be simple. There are only a few likely sources of the “international recognition” Mr Morgan has claimed.

  1. A friendly Australian climate scientist.
  2. Jim Salinger, who studied at the University of East Anglia, which is overseas.
  3. The chief climate scientist of Senegal.
  4. Egos of NIWA.

I hope it’s not 4. Readers may think of other possibilities, but any real scientific source, like a university or a national meteorological office somewhere, is highly unlikely, or he would have told us by now. I know this will annoy him, but I’ve had enough.

Just send me the documents, John.

23 Thoughts on “NIWA withholds evidence, loses face

  1. They won’t give you the documents because the documents won’t support them. NIWA are full of it RT, the ombudsman is the way to go.

  2. I hope you’re not completely right but I fear you are. Certainly the ombudsman is our backstop.

  3. Richard C (NZ) on March 21, 2014 at 8:42 am said:

    >”We have to wonder what he’s hiding”

    He’s not hiding anything, he has nothing to hide – that’s his problem.

  4. Richard C (NZ) on March 21, 2014 at 9:59 am said:

    >”being a matter of expert scientific opinion”

    With these words, John Morgan concedes there is no citation NIWA can offer. NIWA’s method is only their own in-house opinion which, apparently, is “expert scientific” even though it has no basis in the scientific literature whatsoever.

    NZCSET cite Rhoades and Salinger (1993) for their 7SS. BOM cite Menne and Williams (2009) for ACORN-SAT.

    NIWA cite nothing.

  5. Alexander K on March 21, 2014 at 8:02 pm said:

    John Morgan is behaving in the manner of a classic corporate bully whose modus operandi is that of a small, spoilt boy. It is childish of him to use the ‘argument from authority’ like said small boy shouting ‘My father’s a Pleeceman, so there!!’
    Go to the Ombudsman.

  6. Richard C (NZ) on March 22, 2014 at 10:45 am said:

    >”Go to the Ombudsman”

    There’s no point going to the Ombudsman Alexander because NIWA (Morgan) are conceding there’s no document or citation they can produce.

    They only have opinion to produce – (Morgan) ”being a matter of expert scientific opinion”. Opinion is all they’ve got to offer and that’s all the Ombudsman would be able to extract i.e. a repetition of exactly the same as in Morgan’s letter above which is effectively a letter of concession.

    Basically, NIWA’s 7SS method has no basis in scientific literature. Morgan effectively confirms that in his letter. NIWA’s 7SS basis is only in the mind-of-Mullen now that Salinger has left NIWA. J Venning took the word of the mind-of-Mullen in his decision – nothing else.

    And NIWA cannot give the mind-of-Mullen to the Ombudsman for him to pass it on to the NZCSC unless by some Hannibal-like surgical procedure.

  7. “There’s no point going to the Ombudsman”

    Nice analysis. However, I’m not asking for confirmation of the method, but of the claimed “international recognition.” So I disagree that NIWA have nothing to give us. Obviously, if there really is a written reference as Morgan claimed in the Herald, he can give us that. But if there’s not, he can still give us a statement that there is not. In other words, he can tell the people of New Zealand the flippin’ truth.

  8. Richard C (NZ) on March 22, 2014 at 11:24 am said:

    >”But if there’s not [a written reference], he can still give us a statement that there is not. In other words, he can tell the truth to the people of New Zealand.”

    Yes true, and a fundamentally new course IMO. What you’re moving to now is confirmation from NIWA as an organization of Morgan’s effective concession above that there is no literature that can be referenced that stipulates NIWA’s 7SS methodology or “international recognition” of similar applications i.e. no independent scientific replication of NIWA’s 7SS is possible whether of NZ’s 7SS or applications anywhere else in the world. Morgan can confirm that, there’s no need for the Ombudsman.

    In other words, NIWA’s 7SS is opinion-based only but scientifically and statistically invalid in terms of replication. NZCSET proved that.

  9. Correct. Could you spell out, I wonder, how you see Morgan’s letter as conceding there’s no authority for his claim of recognition?

  10. Richard C (NZ) on March 22, 2014 at 11:44 am said:

    >”or applications anywhere else in the world”

    Keeping in mind that anyone attempting an application (let alone a replication) must first know the mind-of-Mullen (or the mind-of-Salinger) in order to apply the undefined subjectivity of Mullen (or Salinger) to non-NZ data.

    I think John Morgan would have difficulty pointing to even one instance of that.

  11. You’re saying that John Morgan lied to the Herald.

  12. Richard C (NZ) on March 22, 2014 at 12:36 pm said:

    >”Could you spell out, I wonder, how you see Morgan’s letter as conceding there’s no authority for his claim of recognition?”

    It’s not an explicit concession, which is why confirmation is required, but it is certainly an implicit concession based on the following:

    1) Omission by Morgan of any reference to any authority recognizing NIWA’s 7SS method as being defined, applicable, and replicable, for any non-contigious station data homogenization. Or to the NZ 7SS either (BOM’s Review just checked spelling and arithmetic – not “technical” as they put it).

    2) Morgan’s statement implicitly concedes no other authority other than in-house NIWA opinion (as accepted by Venning) – “I remain firmly of the view that the question you have raised with respect to the methodology applied by NIWA in the Review has been thoroughly canvassed before the Court and, being a matter of expert scientific opinion….’.

    Morgan, “firmly of the view”, does not consider NIWA’s methodology to be a matter of definition-by-documentation in the scientific literature and international recognition to be of that. He considers it to be “a matter of expert scientific opinion” only where recognition, presumably, is by unquestioned adoption of it e.g. by HadCRU and IPCC i.e. they trust (recognize) NIWA’s “expert scientific opinion” by default because there’s no other basis on which to make an assessment.

    3) Omission by Morgan of any non-NIWA non-NZ scientifically expert opinion specifically endorsing NIWA’s method. None were produced in court – just Mullen and Wratt from NIWA.

    4) In respect to your “…please provide a copy of the scientific literature that approves of the measurement technique used by NIWA in the Review. This request is reiterated under the Official Information Act”, Morgan is correct to state – “…..the question you have raised……..is not the correct subject matter of an Official Information request”.

    The subject matter is opinion (the subjective “mind-of-Mullen”) therefore there is no information to be provided in documentary form and the OIA is not the correct channel. That would be an Official Opinion Act request if it were law.

    As I see it Morgan has, perhaps inadvertently, made a major statement of implicit concession – no documentation exists – none can be provided.

  13. Richard C (NZ) on March 22, 2014 at 12:52 pm said:

    >”You’re saying that John Morgan lied to the Herald.”

    No. See 2) above, specifically re “recognition”:

    “where recognition, presumably, is by unquestioned adoption of it e.g. by HadCRU and IPCC i.e. they trust (recognize) NIWA’s “expert scientific opinion” by default because there’s no other basis on which to make an assessment.”

    “International recognition” can mean anything i.e. anyone (e.g. BOM, HadCRU. IPCC) from outside NZ viewing NIWA’s 7SS “recognizes” NIWA’s method by default i.e. they don’t audit and attempt to replicate as NZCSET did but by their adoption and acceptance of it they are “recognizing” (endorsing) it but by no other standard than NIWA’s in-house “expert scientific opinion”. If they were to attempt application or replication they would run into the same situation as NZCSET – no documentation.

  14. Richard C (NZ) on March 22, 2014 at 7:31 pm said:

    I’ve been going back through the initial email thread ‘Cumulative step adjustments in the NZT7 BOM/NIWA review report’ (51 emails in all, provoking new threads after that) to ascertain the what/when/why’how events leading to the dual realization by the participant parties to the thread that an established statistical method for break adjustment was necessary for replication of a NZ 7SS and that NIWA were not following established adjustment methodology for their compilation of it.

    Here’s a (very) abbreviated synopsis of the key points starting with Bob’s early identification of a discrepancy in the Masterton case study we were using (Barry’s 7:32 PM mention of Rhoades & Salinger is the first instance of that paper in the thread):

    +++++++++++++++++++++++++++++++++
    Bob Dedekind, Sat, Feb 12, 2011 at 5:28 PM
    Using my method of checking the differences using the 8 years pre- versus post-site change, I get -0.09 for the 1920 change, where NIWA gets -0.21, and I get -0.16 for the 1942 adjustment, where NIWA gets -0.26.

    Now to me, my method is more valid, as I’m checking the station against itself. Ie: if the site change had an effect, surely the first place we would see it is in the station itself. Hessel uses this approach. Why use distant stations to do exactly the same analysis?

    -Bob

    PS: Barry, I’m copying you in on this discussion as I think you could have some useful inputs.
    +++++++++++++++++++++++++++++
    Barry Brill, Sat, Feb 12, 2011 at 7:32 PM
    Whilst it makes sense to check for area-wide changes, this can only be done when nearby stations are available. In the case of Albert Park, the only sites that could possibly be described as “neighbouring” are both subject to Forest Heat Island effects – of unknown severity.

    But both Hessell and Rhoades & Salinger regard the station’s own record as the first and best indicator to check.

    As to cumulation, NIWA argues that all sites, screens, etc at any location (eg Masterton) can only be spliced together after all are made compatible with one ‘reference’ site – usually the current open one, but they select East Taratahi EWS (ET) in the Masterton case. Because Site 6 allegedly had a warmer baseline than ET, 0.08° needs to be subtracted from all its data. Site 5 (Essex St) was 0.26° warmer than Site 6, and therefore 0.34° warmer than ET. So 0.34° needs to be subtracted from its records.

    Is this a correct description?

    Barry
    ++++++++++++++++++++++++++++++++
    Bob Dedekind Sat, Feb 12, 2011 at 7:34 PM
    What confuses me is I haven’t seen any discussion in their docs explaining why this most obvious, simple method isn’t used.

    But I haven’t read them all in detail yet. Have you seen a discussion?

    -B
    ++++++++++++++++++++++++++++++++
    Bob Dedekind Sat, Feb 12, 2011 at 7:46 PM
    >Is this a correct description?
    Yes. Spot on.

    To me, I’m appalled at just how bad the data is. We have a ‘long’ station record that is in fact nothing of the sort. It was moved from place to place and spliced together. Now NIWA is going through and inventing even more adjustments.

    It is very difficult to keep up, as they own the metadata, as well as the raw records (paper-based?). They also have many more resources than we do.

    I think your approach is the best, to question why they are using these techniques, when they aren’t generally accepted. Everything depends on their correlation diagram, which gives them the confidence to use remote station comparisons.

    -Bob
    ++++++++++++++++++++++++++++++
    Barry Brill Sun, Feb 13, 2011 at 9:59 AM
    Bob
    We know their comparison-station selection method is shonky, and now it seems their comparison method is equally unorthodox.

    The Review relies upon the assertion (p.11) “trends and interannual variations at nearby sites are very similar”, which perhaps recognises that inter-monthly variations are not in lock-step.

    I take it your comparisons gel with R&S in using monthly data for brief before-and-after periods, but I presume you have used simple averages (rather than “CUSUMs” and a weighting by the fourth power of correlation)?

    I have a raft of questions:

    # What is the status of a calculated figure which is much smaller than the margins of error? eg your first figure seems to have error bars 30 times larger than itself. At what point does the figure become statistically insignificant?
    # “averaging over the year and then calculating differences is never going to work” – why is that?
    # Can you calculate NIWA’s standard errors? eg Waingawa 1942. Are their actual calculations accurate?
    # Does the R&S quote (p904) mean that potential adjustments of less than 0.1°C should be ignored?
    # What is the meaning of the sentences following the quote: “The significance level x is, of course, a matter of choice. It has a profound effect on the statistical properties of the adjusted series”? (Rhoades is a statistician – shows, doesn’t it?).

    I’ve asked [redacted] to recommend an academic statistician who could do some work for us on contract. My aim is to pre-prepare a series of well-defined statistical questions, recognising his/her lack of prior climate science knowledge.

    Barry
    +++++++++++++++++ [End of relevant points]

    Anyone else attempting the same NIWA methodology replication (or application) exercise would have to go through the same process as above. Bob’s confusion (Sat, Feb 12, 2011 at 7:34 PM) arises from the inability of anyone to refer to NIWA’s method definition-by-documentation.

    Now John Morgan is stating (implicitly) that there is no NIWA method definition-by-documentation. He is stating (explicitly) that he is “firmly of the view that …………. with respect to the methodology applied by NIWA in the Review has been thoroughly canvassed before the Court and, being a matter of expert scientific opinion…” i.e. definition-by-undocumented-opinion.

    So instead of definition-by-documentation, NIWA’s method is (according to Morgan) definition-by-undocumented-opinion therefore no reference can be made to it and no guidance that can be followed by anyone else obviously. The only way the 7SS can be compiled independently is by recourse to established documented methodology that NIWA applies only arbitrarily and subjectively i.e. an invalid scientific and statistical approach impossible to replicate – hence the discrepancy when R&S93 is applied, as it was by NZCSET eventually in lieu of any access to definition-by-documentation by NIWA.

    Clearly, method definition-by-undocumented-opinion is non-scientific and non-statistical.

  15. Alexander K on March 23, 2014 at 2:29 pm said:

    Little wonder that the head of NIWA is being stubborn – he is sitting on a real Pandora’s box of issues! Perhaps he is relying upon the lack of credibility of ‘established climate science authorities’ ensuring they don’t blow the whistle on his ‘argument from authority’. I remain convinced that he is a corporate bully and also remain more than a little concerned by Justice Denning’s willingness to be duped by said corporate bully.

  16. Magoo on March 23, 2014 at 3:32 pm said:

    He could just be sitting on the documents to be a jerk and to make things as difficult as possible for those who dare question the oracles of the high priests at NIWA.

  17. Richard C (NZ) on March 23, 2014 at 6:16 pm said:

    I think John Morgan is being pragmatic rather than obstinate. He knows full well that NIWA’s 7SS method is only in the heads of former employee(s) Salinger (and Renwick maybe) and a few current employees, Wratt and Mullen being those who know the history and left carrying Salinger’s can (possibly none too happy about that too). But I suspect there will be others updating the recent data (e.g. Georgina Griffiths perhaps) who would be unable to reconstruct the early data as per Salinger’s head.

    Morgan’s “international recognition” of the 7SS is meaningless in terms of its validity. Osama Bin Laden was an engineer who had “international recognition” too. He was engineering expert enough to know that the Twin Tower structures could be brought down completely by exploiting the design weakness. But I don’t think Bin Laden will be remembered along side Brunel or Eiffel for the same reasons they will be even though all three were engineers of international recognition.

    HadCRU insert NIWA’s 7SS into HadCRUT4 complete, that’s “international recognition” but so what? Their quality control doesn’t extend to questions of validity, they just accept the product on the word of the supplier. It doesn’t work that way in the commercial world though. For example, the kiwifruit export season is about to start this week. Zespri specifies the quality of product for post harvest operators to supply. But they don’t just assume the product supplied meets their specification – they check it at the port before it is loaded on the ships. If it isn’t to spec it is rejected and sent back to the supplier.

    The NZCSC (via Richard Treadgold’s probing) has John Morgan over a barrel now. J Venning’s court rationale (accepting opinion in lieu of documentation), rather than letting NIWA off the hook, is now coming back to dog Morgan. He can’t just throw in the towel (and Wratt and Mullen under a bus), so he has to make the best of NIWA’s predicament and the facts as they stand, for as long as he can, while projecting an air of confidence in NIWA’s position, real or not.

    But now the truth is being exposed. NIWA’s 7SS has no documentation enabling replication, no documented definition, no specification, no quality assessment possible by methodology (not NIWA’s anyway), no scientific or statistical validity because its all in a mind or two and not necessarily minds at NIWA either. John Morgan has yet to publicly confront this dawning reality (I’m sure he has privately though). I’m hoping Richard T’s next round of communication with John Morgan (not the return letter above but the next round) will confront him with it in no uncertain terms.

    In any event, unless John Morgan ignores further communications from NZCSC completely, he cannot claim scientific and statistical validity for the NIWA 7SS because it cannot possibly be replicated independently from established and documented methodology that doesn’t exist.

  18. Richard C (NZ) on March 23, 2014 at 6:34 pm said:

    >”He [Morgan] knows full well that NIWA’s 7SS method is only in the heads of former employee(s) Salinger (and Renwick maybe) and a few current employees,”

    Citation:
    Mullan, A.B; Stuart, S.J; Hadfield, M.G; Smith, M.J (2010).
    Report on the Review of NIWA’s ‘Seven-Station’ Temperature Series
    NIWA Information Series No. 78. 175 p.

    http://www.niwa.co.nz/sites/default/files/import/attachments/Report-on-the-Review-of-NIWAas-Seven-Station-Temperature-Series_v3.pdf

    Mullan, Stuart, Hadfield, and Smith are those current NIWA employees (if they’re all still there) in whose heads the NIWA 7SS “opinion” methodology resides.

  19. Richard C (NZ) on March 23, 2014 at 7:42 pm said:

    >”its all in a mind or two and not necessarily minds at NIWA either”
    >”Mullan, Stuart, Hadfield, and Smith are those current NIWA employees (if they’re all still there)”

    Full Staff List – Climate

    Brett Mullan, Principal Scientist – Climate
    Mr Stephen Stuart, Climate Scientist

    http://www.niwa.co.nz/our-science/climate/key-contacts

    The minds of Hadfield and Smith have moved on apparently (maybe they were seconded for the Review). Mullan’s and Stuart’s still there.

  20. “I think John Morgan is being pragmatic rather than obstinate.”

    You present a good set of reflections which I mostly agree with. As far as the man in the street is concerned, he’ll be briefly disgusted with the misrepresentation of NIWA’s methods and quickly move on to finding lunch. More thoughtful people will see the long-term implications of dishonesty so they will object to the perpetrators being reappointed to senior NIWA posts. That’s how the system responds to persistent disdain and dupery. This will hurt them in the end.

  21. Richard C (NZ) on March 24, 2014 at 3:10 pm said:

    >”More thoughtful people will see the long-term implications…..”

    Apart from NZ being represented historically by a series that can’t be independently replicated from scientific literature, the one critical long-term implication of consequence as I see it is (in a world where scientific integrity and quality prevails) that the NZT7 composite in HadCRUT4 prior to about 1970 will have to be replaced by an independently validated series accompanied by documented methodology that can be replicated.

    Pigs might fly but imagine the flak NIWA would be in for if that ever transpired. It would be nice if the current senior heads were still around for that. I’d like to be still around too.

    NIWA would have rather different “international recognition” of their 7SS to what Morgan is currently alluding to.

  22. “NIWA would have rather different “international recognition” of their 7SS”

    Touché.

Leave a Reply

Your email address will not be published. Required fields are marked *

Post Navigation