A court challenge to the validity of the New Zealand temperature record [NZTR] has concluded. The Judgement refused all three parts of the challenge to the NZTR.
The challenge had been initiated by a group of climate researchers called The New Zealand Climate Science Education Trust [the Trust] against the government funded scientific body which prepared the NZTR, the National Institute of Water and Atmospheric Research LTD [NIWA].
The Trust issued a Statement of Claim [SOC] seeking:
A declaration that the New Zealand Temperature Record is not a full
and accurate record of changes in the average surface temperatures
recorded in New Zealand since 1900.
The Trust supported this claim on the basis of a pronounced difference in the temperature trend from the raw temperature data at the seven sites used to produce the NZTR and the trend after adjustment by NIWA. When enquiries were made to NIWA for an explanation the Trust was referred to a 1993 paper by Rhoades and Salinger [RS93], a leading statistician and climate researcher respectively, which provided criteria for adjusting raw temperature data.
When the Trust applied the criteria for adjusting temperature from RS93 they found the adjusted trend was still much less than the trend used by NIWA:
| Series | Trend (°C/century) |
|---|---|
| Unadjusted | 0.23 |
| NIWA method | 0.91 |
| Rhoades & Salinger method | 0.34 |
It should be noted that there are cogent reasons for adjusting raw temperature data. In an entertaining paper by Torak some of the reasons for adjusting were compiled and can be viewed at Table 2.2; they included the involvement of beer in the data collection process, unruly birds and other wildlife, inferior readings by women and a suspicion that readings were exaggerated because extra rations were given if the temperature went over 100°F.
However, the challenge against the NZTR was not about the need for adjusting the temperature record but that the official criteria for adjusting the temperature was not followed by NIWA.
The Defence filed by NIWA asserted that the NZTR was not a public or official record. The NZTR was only for research purposes. The Defence distinguished between the data from the official temperature sites which it admitted was a public resource and the NZTR record which it claimed was not a public resource.
In response to the Defence an amended SOC was filed. The amended SOC considered the 2010 review by NIWA of the then 7 official temperature sites and the replacement of those by a new 11-site-based temperature record in 2009. The amended SOC noted at paragraph 41:
The data and calculations used in the Review Report, in particular
the adjustments made to the raw temperature data, differ markedly
from those utilised in producing the 7SS. Notwithstanding these
significant changes the NZT7 shows a coincidental, and for the
reasons set out below a scientifically untenable, century-long
warming trend similar to that shown in the 7SS.
In effect NIWA had replaced the old temperature network and employed a new and/or revised adjustment process and criteria and still found approximately the same temperature trend.
In his Judgement Venning, J. allowed the challenge in principle because NIWA was a Crown research Institute [CRI] and no other method of challenging it was available. In effect while Venning, J. agreed the NZTR was not a public record [paragraph 38] the Public had a right to challenge actions of NIWA such as producing a NZTR.
However Venning, J. dismissed that right to challenge on the basis of criteria outlined at paragraph 47 of his Judgement:
I consider that unless the Trust can point to some defect in NIWA’s decision-making process or show that the decision was clearly wrong in principle or in law, this Court will not intervene. This Court should not seek to determine or resolve scientific questions demanding the evaluation of contentious expert opinion.
The error “in principle or in law” which the Trust tried to establish was that NIWA did not adjust temperature according to its own criteria as described in RS93.
NIWA’s own expert witnesses admitted that they did not follow the RS93 criteria in a 2010 “Review” of the NZTR [paragraph 147] but had done so in forming the original NZTR which had been changed in 2009 from a 7-site network into an 11-site network which in turn was reviewed on the basis of other unspecified criteria for adjusting the raw data.
NIWA’s expert witnesses stated it did not matter what criteria they used to adjust the raw temperature data or what sites they adjusted the data from because the temperature trend results were the same as the trend produced by the RS93 adjustment criteria [paragraphs 148-149].
To double-check that the alternative method[s] of adjusting the raw temperature data produced the same temperature trend as the RS93 criteria for adjusting would have done if used in the 2010 “Review” NIWA’s experts did apply RS93 to the 2010 Review of the NZTR sites [paragraph 144]. The NIWA experts concluded that the Trust expert witnesses had been incorrect when they applied RS93 to the raw data and found a lesser trend than found by NIWA when it applied both RS93 and the other adjustment criteria.
This thoroughness by NIWA in checking its own adjustment procedures was approved by Venning, J. who concluded at paragraph 157:
I accept that NIWA could have recalculated the temperature adjustments in a different way yet still have arrived at a similar result which would strengthen the robustness and validity of the previous results.
Venning, J. could do nothing other than accept the statements of the NIWA experts because earlier he had dismissed the expertise of the Trust witnesses [paragraphs 49-54].
However, in accepting that NIWA could produce the official trend of 0.91°C from different adjustment methods Venning, J. provided legal interpretation of two principles of the AGW debate.
The first is that experts working for government or ‘official’ scientific institutions [CRIs] are given greater evidentiary weight than witnesses not working for CRIs. On this basis the authority of the AGW science has been judicially endorsed.
Secondly, at paragraph 80, Venning J. elaborates on the issue of “officially recognised scientific opinion.” Venning J. recognises NIWA’s expert opinion that in respect of temperature adjusting and climate matters generally, there is no one “immutable” standard and indeed could not be.
NIWA’s adjustments of temperature are, therefore, essentially subjective in nature, and, like the definition of “excellence” which the Trust claimed was the standard to which NIWA should be held accountable, not “readily susceptible to judicial assessment” [paragraph 74]. What the courts will do, as Venning J’s judgement shows, is give considerable discretion to CRIs in their preparation of temperature networks and trends by adjusting raw data.
This discretion extends to subjective aspects of temperature adjustment where the institution can choose the method to adjust temperature data regardless of other methods and criteria; and in matters of fact where the institution has used or not used a method of adjustment of temperature data the institution is the preferred authority for concluding the fact of what it has done. On this basis the AGW science has been judicially settled.
New Zealand case law is not binding on other Commonwealth jurisdictions like Australia, but still persuasive.
In Australia the Bureau of Meteorology [BOM] compiles the Australian temperature network in a similar fashion to NIWA, by adjusting raw data. BOM has just replaced its old High Quality temperature network [HQ] with a new network, ACORN. BOM did this because it recognised problems with the HQ network which possibly made it inaccurate. ACORN has supposedly rectified those inaccuracies but has still produced a similar temperature trend to the HQ network. ACORN was peer-reviewed by a panel of climate experts including NIWA representatives.
Questions were raised from outside BOM about the reliability of the HQ network. Questions are again being raised about the reliability of ACORN.
However, based on the New Zealand temperature case, any questions raised from outside the government scientific institution, BOM, especially from sources which the court does not recognise as being “officially recognised scientific opinion”, are unlikely to have much credence given to them by judicial appraisal.
Views: 402
A good summary of the issues and the case.
I have one comment, as the person who did the RS93 analysis, on the statement by Venning:
Unlike Justice Venning, I have NO confidence that NIWA could have recalculated using RS93 and arrived at a similar result. Had Venning not excluded my evidence, he would have noted that the RS93 analysis I did showed conclusively that the trend would have come out at about a third of NIWA’s over the last 100 years.
Did I do the RS93 analysis correctly? Yes, even NIWA had to admit that, and in desperation they have resorted to claiming that I did it “too rigorously”. One laughs.
One main question remains:
Why, when NIWA had the chance to redo the NZ7T in 2010 with taxpayer money, did they not perform a RS93 analysis? It’s not that difficult – I did it, and nobody paid me! Why go off and use a crude method from a student’s thesis in 1981 instead?
Applying Occam’s Razor, Bob, the most likely explanation is that you were just wrong. Again.
Once again Rob, you outdo yourself on ill-considered comments.
So according to Rob Taylor, the reason NIWA didn’t use RS93 in their Review in 2010, instead choosing a simplistic method from a student’s thesis written in 1981, was that I personally was wrong in my analysis written a year later.
Brilliant, one stands in awe.
Clearly, your “analysis” was amateurish and wrong, Bob.
What ARE your professional qualifications, exactly?
This blog used to have some interesting discussions until Rob turned up.
Why, Andy, you flatter me.. you’ve taught me so much, too.
Andy & Rob,
Drop it. You’re both capable of more than a mere slanging match.
That’s it – my irony meter’s just blown again! Damn…
I was wondering how long it would take the Hot Topic trolls to turn up after Rob Taylor posted the “breaking news” about Bob’s real identity over there.
I hope you’re joking, Andy…
Please remember, don’t give them any oxygen. Andy? Andy??!!
I’m not giving them any oxygen, though I am happy to donate some spare carbon dioxide I have lying around.
Just the thing.
And now the hypocrisy meter’s blown a fuse, too! 😉
“Clearly, your “analysis” was amateurish and wrong, Bob”
And so were the 3 professional statisticians that reviewed it presumably?
“the 3 professional statisticians that reviewed it” – and who were they, Richard. Part of the “anonymous science” team perhaps?
You might find out from the appeal Ken – be patient.
Or you might find out if the ‘Statistical Audit’ is published – that’s a possibility.
Strange though, that the judge didn’t ask the same question.
>Clearly, your “analysis” was amateurish and wrong, Bob.
Hmm, possibly. Strange that NIWA said the opposite then. The only “criticisms” raised were that the RS93 period chosen (1 and 2 years either side of a site change) were too short. This of course is incorrect – the central theme of RS93 is one shortens the time period specifically to exclude gradual or sudden undocumented effects, while using monthly values to ensure statistical significance.
In RS93 only 1 or 2 years are mentioned, and the example specifically uses 2 years. They claimed that somehow in following the RS93 method I was “too rigorous”.
NIWA has no leg to stand on here.
So, Bob, what ARE your professional qualifications, exactly?
You describe yourself as a professional statistician – prove it.
Or are you just another wannabe internet Galileo in your lunchtime?
I do? Where please?
From my affidavit (publicly available):
I’ll tell you what, Rob. Check these two papers and let me know where the statistical errors are please. I’m sole author on one and lead author on the other, so the mistakes will all be mine.
http://researchspace.csir.co.za/dspace/bitstream/10204/1656/3/dedekind_1994.pdf
http://researchspace.csir.co.za/dspace/bitstream/10204/591/1/dedekind_1996.pdf
Then let me know why you believe I don’t have a “sound grounding in the practical application of established statistical techniques”.
So Bob – are you Dedekind? The anonymous co-author with Treadgold of that “paper” Are we getting warmer yet? The one with the huge statistical cock up?
Ah, Ken catches up. Welcome Ken.
I’m sorry to have to say it, Ken, but I suspect you wouldn’t know a huge statistical cock up even if one fell on you.
See here. And yes, I am the same one who pointed out to you and Renowden that you were amusingly incorrect and busy making fools of yourselves bagging NIWA.
Speaking of cock-ups, Matthew, Bob, or whatever your name is, getting the figure wrong by a factor of three suggests your “sound grounding in the practical application of established statistical techniques” was not gained at a university.
Which also explains the frequent schoolboy howlers, such as believing that heat flow is unidirectional.
That clammy feeling is your credibility, evaporating….
Really? Are you sure of that?
So Rob, about those statistical errors in the papers.
Also, since I know so little about heat flow, as a bonus you can tell me about the errors there too, since the boundary conditions are defined using transient and steady-state thermal analyses.
Want to go on?
As an aside to other readers, notice how these people operate – trying to discredit me, not my arguments? Not a single one of these folk have ever shown they even understand the RS93 method, let alone argue its relative merits.
Bob,
Yes, I noticed. Nil illegitimum carborundum.
🙂
Bob – you are embarrassed aren’t you?
If you check my publications you will find many more papers than 2 where I have used statistical analysis. And far more recently too – despite the fact I have been retired for quite a while.
And you couldn’t even see the need for site adjustments even though it stuck out like the proverbial.
Now wonder Justice Venner was unimpressed. 2 publications, unrelated field, many years ago. Where have you been in the last 20 years?
Ken,
You might well be more qualified in statistics than Bob and more recently published, but that will never invalidate his qualifications. Shame on your scientific mind for implying it. And it’s a pity your spelling is so consistently awful.
As I’ve explained to you more than once, site adjustments weren’t part of our intention. You don’t listen too well, do you? If you mention it as a snide remark again, you’ll have to visit the naughty step.
“And you couldn’t even see the need for site adjustments even though it stuck out like the proverbial”
Ken you haven’t got a clue have you?
Read ‘Statistical Audit of the NIWA 7-Station Review’:-
https://www.climateconversation.org.nz/docs/Statistical%20Audit%20of%20the%20NIWA%207-Station%20Review%20Aug%202011.pdf
Pages of……..SITE ADJUSTMENTS.
I’m not embarrassed in the slightest, and I base my assessment of your statistical prowess on this.
Still banging that drum, eh? Richard T has already dealt with that, many times over. If you don’t wish to listen, then so be it, but I’m not going to keep trying to convince the unconvincible (is that a word, Richard?).
Well, inconvincible is probably a better form. The context is certainly apt. – RT
Who said I only wrote 2 papers? I said it was because I was the sole author on one and lead author on the other. And they’re both online in full and easy to access.
In New Zealand for 16 of them, and New Zealand doesn’t have these sorts of gas turbine engineering facilities. So I changed career and went into IT. Never looked back, to be honest.
My papers weren’t even presented in any of our affidavits, as you’d know if you’d ever bothered to read them.
I might also add that the vast bulk of our work was unpublishable, due to commercial and political constraints, as is true in almost all defence environments worldwide.
See here for one example of the work we were doing. Publish? I think not.
Bob/Otto you are embarrassed about “Are we getting warmer yet? – as is Treadgold. No adjustments necessary you both declared and implied any adjustment by NIWA were fraud.
Why did Treadgold keep you hidden then? Why did you want to be anonymous (or the “anonymous science team”)? Why would you not respond to critiques made by myself and David Winter at the time? Why did you nott criticise your international mates who used your “paper” to “prove” our climate scientists were committing fraud?
Where was your integrity then, let alone now?
Ken, you’re being idiotic. I explained at the time that anonymity was requested for professional reasons. Stop pressing this point. It’s not relevant.
Don’t you love people who tell you what your feelings and thoughts are? Rob does the same – it’s very amusing.
Specifically because of people like you and Rob Taylor, Richard Christie, etc. People who simply will not debate the science and issues at stake, but insist on hurling insults, screaming loudly and making personal attacks. Richard very kindly volunteered to take the brunt of it and he has, thanks Richard.
Might I just say here (as an immigrant, albeit some time back) that in this process I have seen the best of the New Zealand character (Richard T), and the worst (Rob Taylor, Ken Perrott, Richard Christie, etc.) You really should be ashamed of the way you have behaved.
Because your criticisms were both facile and puerile. You showed no sign of having done even the slightest bit of work to check any of our assertions, you just wanted to shout everyone down, even after Richard T pointed out repeatedly that you were quoting one line out of context. You still do.
I have no control over what people say about anything, just as I’ve never seen you criticize your mates over at HT when they come out with the most patently false accusations and pseudoscience.
The paper was what it was. All it did was show everyone the adjustments, station by station. We drew attention to the fact that NIWA had never even mentioned that adjustments had been carried out, nor that the unusual warming trend came purely from those adjustments. The station histories (we checked them, contrary to your ridiculous statement that we didn’t even think about adjustments) showed no reasons for the adjustments at that time. Since then NIWA have updated their records (they found errors when they checked), and they have, in the Review, published the rationale for each adjustment check.
As it turned out NIWA themselves had no idea about those adjustments. They had to admit they had lost them. The paper was therefore entirely vindicated – NIWA really had no idea about how they got the apparent warming trend – they had just assumed it was correct.
Intact. I have now checked each of the adjustments using the method NIWA kept touting as their own (RS93), and I found that had NIWA actually used RS93, they would have got a different answer.
NIWA have been unable to show any fundamental error in my workings, and have now fallen back to complaining that my analysis was “too rigorous”, and I should relax the RS93 parameters until I get the answer they do.
It’s hardly surprising that I disagree with their opinion, since by doing that I may as well not use RS93 at all.
Sorry Bob, that should be Manfred, Not Otto. I will have to get used to it.
Please note I do not disparage your qualifications as Treadgold says (horrible guy with the truth isn’t he?). But just agree with the judge who said your qualifications were not appropriate to make the claim that you could rule on what was “official accepted methodology” in climate science. Your opinion was not that of a qualified climate scientist and therefore irrelevant. My qualifications would not have been suitable either – but I would not have gone to court with such a silly claim.
The “paper “Are we getting warmer yet?” Is like the proverbial albatross isn’t it. Poor old Richard points to the later document (refereed by 3 “anonymous” statisticians (??? – bit of a pattern here) – both “papers” can’t be right and you refuse to acknowledge the first one was so obviously wrong? Mind you that would require an apology and you guys find difficulty saying sorry, don’t you.
Embarrassing all round I guess – especially as you guys must now find the money to pay for your silliness.
Ken, would you like to fill us in with your definition of a “qualified climate scientist” and give us some examples of people that fit this description?
“These engines also have a high rate of thrust increase and, consequently, flight acceleration, which is especially critical for front-line fighters”
Not a big market in NZ now, I see why you went to IT Bob.
Andy, perhaps you could “fill us in” as to why Manfred has the sort of qualifications that would enable the judge to accept his personal opinion as to what an “official accepted methodology” was in climate science. An arrogance on his part I would think? Especially as Venning did quote from “Are we getting warmer yet?” (Rather damming) – and Manfred had given it as evidence of his qualifications for such an opinion!! The very first time he acknowledged his involvement in this sorry document.
Ken, J Venning preferred (favoured, contrary to his Judicial Oath) the opinion of a climate scientist (Dr Mullen) over the expertise of professional statisticians (reviewers) when deliberating over a statistical question in [144].
Does this mean one climate scientist is a better statistician than three statisticians are?
If so, what are the climate science qualifications that override statistical qualifications?
Or has the Judge made a boo boo?
Not at all. Everybody calls me Bob, even though I was christened Manfred.
It was that short-sighted thinking that we were hoping to challenge, but unfortunately the judge took the easy off-ramp (in my opinion) and decided not to even look at the evidence, based purely on my qualifications. Case closed.
NIWA got away with it in the end, but had we got around to addressing the science the result may have been different.
Peer reviewers are anonymous in almost all journals. Are you going to criticise all the journals too? Had we not had the paper reviewed you’d be complaining that we should have done.
And I’d like to point out something here. You should only be criticising our reviewers if you have found an error! You have not. In fact you’ve found nothing at all, because you still haven’t even looked, and don’t even understand the adjustment methods used, nor the rationale for them. Nor do you seem to have downloaded any data and checked for yourself, nor have you ever engaged in a rational, intelligent discussion about the merits of the Review or the Audit. I don’t think you’ve even read them.
And yet you try to lecture us on how our paper was an albatross? Don’t make me laugh – it was the trigger that got the whole thing going. We now have a better understanding of the 7SS than NIWA does, because (unlike NIWA) we have actually done a RS93 analysis. Our analysis, being based on an internationally-accredited method, has more standing than theirs.
No Ken, I want you to tell me what qualifies someone as a “climate scientist”?
Which particular combination of degrees does Michael Mann have, for example, that leads him to his position in the holy ordained ministry of Climatology?
he has degrees in Maths and Geophysics I believe. Furthermore, he published his (in) famous Hockey Stick paper when he was quite young.
So, Richard, these 3 (anonymous professional statisticians) gave evidence at the High Court – must have been interesting.
Were you not aware that the court action had nothing to do with Manfred’s little exercise – it was an attempt to discredit the scientist who produced the NIWA temperature record. Manfred’s document was not scrutinised – no reason to.
So Andy, you can’t find anything in Manfred’s qualifications to enable him to confidently assure us about “official climate science methodology” (especially when he opposed working climate researchers) either. You can understand why Justice Venning was not impressed by Manfred’s assurances.
There’s no basis for the opinion of a climate scientist to be favoured over statistical reviewers in this case.
If the data for each 7SS location were anonymized, no-one would have a clue that they were looking at temperature records for Hokitika say, if the file names for the contributing Hokitika stations were merely: XYZ_123_a, XYZ_123_b, XYZ_123_c etc.
The process then becomes a statistical break-point analysis of anonymous data that is of indeterminate origin.
Manfred you say “Our analysis, being based on an internationally-accredited method, has more standing than theirs.” So why should we take your opinion about suitable methodology when you have no qualifications in the area? Especially as you claim to know more that working climate researcher about suitable methodology. (By the way – where is the method you favour “accredited” ?) You didn’t convince Justice Venning, why should we believe you?
In the old days, Ken, we used to go to University to learn how to learn.
I don’t really see why you need some special kind of characteristics to understand the concepts behind temperature records and how they fit into the local climate.
It really isn’t rocket science. It requires patience, diligence and technical competency.
Bob has show that he has all these characteristics. Basing an argument on what “qualifications” one has is a non-starter in my book.
Once again, Ken, you show your misunderstanding. The methodology was NIWA’s, not mine. NIWA gave RS93 as the internationally-accepted technique. I simply used it.
How difficult is it for you to grasp this one fact? You yourself went on about it constantly in 2009 and 2010, don’t you remember? You kept repeating how everyone knew the method used for the adjustment was Rhoades and Salinger (1993)?
“Manfred’s document was not scrutinised” – Right
“no reason to” – Wrong
The Judge should have determined the statistical validity (professional review) of the document’s workings (one simple question would have achieved that) – not the expertise of the lead author.
But that will be sorted out in an appeal I’m thinkin.
Richard – one minute its anonymous referees giving evidence in court – now you are prattling on about anonymous data! No records – only your personal recommendation! Pull the other leg!
Explain then why Manfred didn’t look at his anonymous data and notice the glaring site effect. Simple statistical analysis (site 1, site 2, etc – anonymous enough for you or did they assume only one site?) would have shown it – just as a simple look at the plotted data (which sensible statisticians usually do first thing) is slightly easier .
Isn’t this funny? Ken makes another significant blunder, accusing me of recommending a new adjustment methodology (RS93), when in fact it was NIWA’s method all along.
Then, when I point this out to him, he simply changes the subject and talks about eye-balling data as a reasonable statistical technique.
“one minute its anonymous referees giving evidence in court”
No Ken, that’s your serial misunderstanding and ignorance kicking in again.
The rest of your comment is irrelevant gobbledygook so I’ll not waste my time responding.
BTW – Anonymization of statistical data http://spdp.dti.unimi.it/papers/it2011.pdf
“NIWA’s adjustments of temperature are, therefore, essentially subjective in nature”
‘Report on the Review’ wrt NIWA’s methodology says on 13:-
“The details of the methods used to identify the necessary adjustments are provided in the individual reports on the locations used in producing NIWA’s “seven-station” temperature series.”
Then on page 17:-
‘Creating a Composite Temperature Series for Auckland’
“In February 2010, NIWA documented the adjustments in use at that time (see web
link above). These adjustments to the multiple sites comprising the ‘seven-station’
series were calculated by Salinger et al. (1992), using the methodology of Rhoades
and Salinger (1993), which extended the early work on New Zealand temperatures by
Salinger (1981). Subsequent to 1992, the time series have been updated regularly,
taking account of further site changes as circumstances required.”
http://www.niwa.co.nz/sites/default/files/import/attachments/Report-on-the-Review-of-NIWAas-Seven-Station-Temperature-Series_v3.pdf
The key elements being “using the methodology of Rhoades and Salinger (1993)” and “the time series have been updated regularly”
R&S93 was the initial methodology applied loosely, NIWA followed that method (also loosely “therefore, essentially subjective in nature”) when updating the series since 1993.
The irony being that NZCSET 7SS and NIWA 7SS are not much different compared to BEST NZ. BEST NZ being the outlier in extremis.
Ho hum… tell it to the judge, Richard.
Ooops, you already tried that, didn’t you?
Will you even once address the topic, Mr Taylor?
He won’t, which was my point above. Not to promote a flame war, but that content-free ad hominem arguments are just increasing the noise to signal ratio on this blog.
There’s no denying it. Thank heaven for moderators…
Dead right Andy, although to be fair to RT it must be like herding cats.
I suggest, in a post like this, tight moderation à la Climate Audit sometimes and perhaps more loose in a post of lessor significance à la Twilight Zone to enable certain people to vent their spleen.
“Not a single one of these folk have ever shown they even understand the RS93 method, let alone argue its relative merits”
From ‘Statistical Audit’
Appendix B
Comments on NIWA’s method
Weightings
NIWA do not weight the individual neighbour station shifts at all. The correct R&S method for weighting is described fully in section 2.3 on page 905 of R&S:
The weights {wi , i = 1, . . . , n} are based on correlations between the target station and neighbouring stations. It is better to use correlations between the differenced series {y(i)} (with k = 1) than between the raw series {x(i)}
The only time NIWA uses correlations at all is when comparing annual values between neighbouring stations, and then only to justify the use of each station. The correlation period (1972 onwards) also doesn’t coincide with the times of most of the station changes.
Error Analysis
The most significant error in the NIWA method is the failure to base each adjustment on the clearly-defined confidence levels as specified on page 904 of R&S (emphasis added):
A [see pdf] per cent confidence interval for the site change is [see pdf] . Following what seems to be the standard convention, we adjust for the site change only if the change is significant at the 5 per cent level, i.e. if the 95 per cent confidence interval does not contain zero.
In other words, one can only adjust a station record up or down if one has first demonstrated that the comparative difference in monthly values is significant at the 95% confidence level.
NIWA have shown no indication that they have even performed error analyses on these adjustments to date, and they appear not to have used any technique whatsoever in determining whether an adjustment should or shouldn’t be made. This means the NIWA method appears to be 100% open to Type I (false positive) errors, since no attempt has been made to prevent them.
Why is this important? In the 7-station series, the ‘noisiness’ of temperature values can cause apparent shifts at times when no site change occurred.
To illustrate the point, we performed an R&S shift analysis for a period when we knew no site change occurred: Hokitika Aero on 1 June 1970 (chosen at random). We used the same neighbouring stations as the 1967 shift check.
The result obtained for k=1 was 0.21 ± 0.50 °C, and for k=2 the result was 0.28 ± 0.35 °C. According to R&S, this result is not significant at the 95% confidence level, and so (correctly) no adjustment should be made.
NIWA, on the other hand, would in this case have made the adjustment of (0.21 + 0.28)/2 = 0.25°C to all pre- June 1970 temperatures simply because they didn’t calculate the correct monthly comparison confidence limits, or use them to determine the validity of their adjustments. This of course would greatly influence the trend of the whole series.
https://www.climateconversation.org.nz/docs/Statistical%20Audit%20of%20the%20NIWA%207-Station%20Review%20Aug%202011.pdf
Maybe this will help.
Bob D.:
Yes, Bob, I’ve still got my D.E. lecture notes somewhere too, but I know my limitations, whereas you seem to have collided with yours, in a very public debacle.
No problem there, your pseudo-scientific beliefs never had anything behind them but ideology and wishful thinking, but I would have thought you’d have at least learned SOMETHING from the experience!
As Einstein once said,
Umm, yes. Whatever you’re going on about. Keeping lecture notes is not quite the same as publishing in the foremost international journal for an area of expertise, is it? Or perhaps you see them as equivalent?
Still waiting for you to show me all the errors in those papers, both statistical and thermal.
C’mon Rob, it can’t be that difficult. After all, I keep making “schoolboy howlers”.
Rob,
Rob, make a point, with references, AND stop attacking people’s imagined intentions and character, or get lost. This is a supreme waste of time and talent. Your university education isn’t doing us a lot of good right now because you’re talking like a fool.
The essence of insanity is doing the same thing over and over again, yet expecting different results…
Sums up your debating strategy Rob….denigrate….denigrate…..denigrate…..denigrate…..
Then repeat, expecting us to accede to you every time on every different issue.
No, RC, I only expect you to accede where you are wrong in fact and in science; which (ahem) seems to be most of the time….
Let’s see you cut-and-paste your way out of this legal debacle!
ROFLASM
Ah yes, and when you’ve finished showing me my errors in those papers, we can get on to the actual topic of the thread, which is the RS93 analysis NIWA did. Oh wait…
Rob, we might believe we’re wrong more often if only you gave reasons, cited references and ANSWERED OUR QUESTIONS. Sorry I shouted. But you confine yourself to hurling abuse, which does nothing to open anyone’s mind. You’re a complete failure as a communicator. Please go away.
“…which (ahem) seems to be most of the time….”
This from one who avoided the CO2 forcing challenge like the plague.
Yep, good point, RC.
Nonsense, I pointed out the general ignorance, on this blog, of the concept of nett climate forcing.
CO2 is only one such forcing; you can educate yourselves here:
http://www.skepticalscience.com/CO2-is-not-the-only-driver-of-climate-intermediate.htm
Have you watched the damning video yet, Richard, or are you another John Key, refusing to view any inconvenient evidence?
Here’s the link again:
http://www.youtube.com/watch?v=CY4Yecsx_-s&feature=youtu.be&t=6m4s
And STILL you avoid the CO2 forcing challenge like the plague (IPCC vs Leckner/Hottel).
Wouldn’t do to give THAT too much exposure would it?
Rob,
That’s so obvious, I have to ask why you feel the need to say so?
Ken:
Hardly. You have missed the salient point (again) that RS93 was given by NIWA as their approved methodology. We simply used the methodology specified by NIWA. It is in fact NIWA who didn’t use it – they used an unpublished, crude method from a student’s thesis in 1981. Foot, meet mouth, again.
Try to keep up, Ken. The judge dismissed the scientific evidence based on my qualifications alone. He said specifically he wasn’t going to make a judgement on the science.
No, Bob – he didn’t make judgements on the science, despite you guys asking for it. That would have been a sad day in NZ history if judges decided on science.
But he did judge that your opinion about “(accredited??) official methodology in climate science” was of no more value than mine – especially alongside the opinion of working climate science researchers. His judgement was on the integrity of the scientist vs the “integrity” of the science deniers.
If the judge “dismissed” the evidence of your three anonymous statistically-inclined friends – what about presenting the actual evidence here (and their names would also help – because, you know what – your assurance about their skill and independence are just not convincing).
Once again you miss the point. NIWA should have used their own methodology – they didn’t. That is something that can be challenged in court, and goes to the heart of excellence in science, and maintaining high standards.
But the judge simply dismissed our evidence showing that they hadn’t used their own methodology, based purely on our qualifications (not on the evidence), and so it was an easy decision for him – the only opinion left was NIWA’s. End of case.
“Science deniers”
Wow this gets better. I see Ken has posted some drivel on his blog about “deniers” losing the battle to get hold of the emails of Michael Mann
In Ken’s world, anyone who dares to ask questions of the holy scriptures of the High Priests of Climatology, even if they paid for these scriptures out of their tax money and have the legal right to that information, is a “science denier”
Bob/Manfred, you are just making excuses.
In my experience researchers use the most appropriate methodology for their situation and time – there is a danger in a naive understanding of “official” “accredited” methodology which makes me suspect you have little experience of research in natural systems. NIWA was clear about why they used the methodology they did and the mistakes you made in your own analysis (just as well your document was not at issue, isn’t it?).
It really does come back to extreme arrogance on your part to think your opinion (motivated of course) on what is acceptable methodology in climate science would have any value. No more value than mine.
Perhaps Ken you can actually address the issues rather than continually question people’s motives and integrity
“NIWA was clear about why they used the methodology they did”
The issue is HOW they used (applied) the methodology they did, not why.
“and the mistakes you made in your own analysis”
Dr Mullen was definitely NOT clear about any “mistakes” (not even the word used – that was “too rigourous” elsewhere) in [144].
There’s no “mistakes” Ken. Or if there are in your opinion, why don’t you identify them specifically?
Ken:
You’re really trying desperately to make a save here, but it’s not going to work.
You’re saying that the most appropriate methodology for the 7SS was RS93 (after all, that’s the method they used, no? – you said so yourself, back in 2009 and 2010), but now suddenly a better technique is an unpublished, untested, non peer-reviewed technique from a student’s thesis written in 1981?
What happened between 1999 and 2010 that caused them to reach that conclusion, Ken, and drop RS93?
And why did you not mention that back in 2009 when you were banging on about how everybody knew RS93 was NIWA’s technique, being internationally peer-reviewed and all?
Manfred/Bob – you are showing all the symptoms of being in denial. Working hard to confuse people with a spin. As I asked before – where is your integrity.
NIWA explained how they did their review independently – not just using exactly the same values as previously. You have done nothing to show thei8r methods are inappropriate – nothing at all – just presented your opinion (which is of no more value than mine). And to cap it off there were faults in the way you tried to apply one method.
No-one has dropped any method – just used what was appropriate and where necessary independent.
Treadgold still having a long lunch? some people are lucky.
“You have done nothing to show thei8r methods are inappropriate – nothing at all – just presented your opinion”
Ken. why don’t you actually read ‘Statistical Audit’ before making an idiot of yourself?
“And to cap it off there were faults in the way you tried to apply one method”
And those were specifically? Stop hand waving, bluffing and casting innuendo Ken – you’ve got nothing to back up your claims..
“NIWA was clear about why they used the methodology they did.”
They were? I must have missed it. Come on Ken, do tell ….. what was the reason they dropped the Rhoades & Salinger method, after all these years?
Andy – clearly deniers describe people who are in denial – a common human failing. Try to objectively look around you and see how you guys are working hard to deny the validity of Justice Venning’s decisions. Blaming everyone under the sun and refusing to face up to your own mistakes – classic denial.
Mind you it’s going to be much harder to face the bill for NIWA’s costs – no matter how hard you go into denial.
Congratulations Ken, you have managed to create yet another comment with no less than 4 uses of the word “denial”, or variants thereof, in it.
“working hard to deny the validity of Justice Venning’s decisions”
Rubbish. Analysis, critique and scrutiny are turning up the grounds for appeal.
Why would the Court system support an appeal process if every judicial decision was perfect Ken?
Andy, I am taking advantage of Treadgold’s long lunch (or is it the siesta afterwards) – and having a bit of fun. But denial is a recognised psychological condition, nevertheless.
Just in the nick of time, Josh provides a bit of light relief
http://www.bishop-hill.net/blog/2012/9/19/save-your-scientists-josh-185.html
Thanks, Andy – a brilliant explosion of light relief.
Hey Richard – careful about counting those chickens. It’s backfired on you so often. You say:
“Analysis, critique and scrutiny are turning up the grounds for appeal. Why would the Court system support an appeal process if every judicial decision was perfect?”
So you have already appealed, got support from the court and won your case? You are in denial aren’t you?
Pull the other one. I imagine all you will be appealing is the size of the costs awarded against you.
“So you have already appealed, got support from the court and won your case?” – No
Wait for the appeal process to run it’s course Ken – don’t get all preemptive and premature yourself.
Besides I’m not a member of NZCSC/ET so it’s not me that’s appealing. I’m just identifying the obvious grounds.
Oh please, please, please appeal the decision, and bring Monckton down to support your case!
Where can I donate to help get things off the ground?
Can I book tickets in advance? I just love a good comedy…