Rhoades & Salinger (1993)
The late Michael Crichton declared that ‘global warming’ had become a religion, with the IPCC reports comprising its bible. The central dogma of the religion rests on the global temperature record, which ‘proves’ recent global warming. It has a bible of its own, largely written by CRU’s “hockey team” — those conniving, partisan, anti-sceptical scientists of Climategate infamy.
Above and beyond everything else it might be, a bible inspires belief. That is the natural result of the veneration a bible gains by long use. And you don’t ignore something you venerate.
Around the world amateur, unpaid investigations reveal that recent warming trends don’t really exist. In fact, they are largely created by “homogeneity adjustments” to the actual thermometer readings. Chapter and verse for these rituals are laid out in a seminal journal paper with 21 authors, Homogeneity adjustments of in situ atmospheric climate data: a review, by Peterson, Easterling, et al. (1998).
The Peterson Review outlined the different but valid approaches taken in Australia, Austria, Canada, Finland, New Zealand, Norway, Sweden and USA. It discussed marine temperatures and global records, confidence levels and likelihood ratios. It reached a crucial conclusion (page 1513):
On very large spatial scales, positive and negative homogeneity adjustments in individual station’s maximum and minimum temperature time series largely balance out, so when averaged into a single time series, the adjusted and unadjusted trends were similar.
New Zealand’s acceptance of the Peterson Review is the basis for the repeated claim that “NIWA uses internationally accepted techniques for its adjustment methodology.” This sentence has become like a mantra in recent years, used frequently on NIWA’s website and in other publications, and featuring regularly in answers to Parliamentary Questions.
The particular New Zealand technique summarised in the Peterson Review is fully described in a paper published five years earlier in the same learned journal (the International Journal of Climatology): Rhoades & Salinger (1993), Adjustment of temperature and rainfall records for site changes.
This is New Zealand’s own bible on temperature adjustments, and NIWA’s veneration of it is strong. Whenever Parliamentary Questions come up concerning authorities, NIWA never fails to cite the Rhoades & Salinger paper. It appeared on the website with New Zealand’s old temperature record, the 7SS, and it’s still right there with the new one as well (the NZT7). It appears in the Overview portion of the 2010 Review report, and in the discussion on each and every one of the seven stations (on pages 27, 55, 73, 95, 119, 144 and 165).
NIWA never tires of the statement “the 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),” since it appears no less than seven times.
New Zealand has produced only two peer-reviewed journal papers dealing with the assessment and adjustment of climate records. One was Hessell (1980), “Apparent trends of mean temperature in New Zealand since 1930”, which found no warming and is hardly mentioned at all by NIWA. The other was Rhoades & Salinger, which also found little warming but offered a comprehensive methodology for adjusting old records. NIWA worships it and mentions it endlessly.
So who were the evangelists behind this holy book?
Prof David Rhoades is a leading New Zealand statistician and seismologist, whose current research interests lie primarily in earthquake measurement and forecasting.
Dr James Salinger is New Zealand’s best-known climatologist, who invented a novel adjustment method in the 1970s and imported it to NIWA in 1992. A member of the “hockey team”, he has co-written 11 papers with Prof Phil Jones, and gets frequent mention throughout the Climategate emails.
It was no accident that Rhoades was the lead author. Adjusting historical records is essentially a data manipulation exercise — peppered with statistical pitfalls — and there is persistent worldwide criticism that climate scientists have very limited statistical skills.
The climate scientist’s knowledge of atmospheric physics plays a useful role, of course, particularly in the interplay of available records for rainfall, wind, climate zones and oscillations. At the strategic level, they are well placed to design a methodology to reveal whether temperature jumps are meteorological or measurement-related, and detect whether gradual change is driven by such factors as UHI or shelter. Also, their familiarity with climate data will be useful for the unavoidable discretionary choices.
It is at the tactical level, and especially the execution, that the statisticians will shine. They know how to identify statistical anomalies, how to maximise the probabilities and exclude the ‘black swans’. They understand proportionality and correlation, and the potential to reduce error bars by testing each step in the process. At the end, they can mathematically test the ‘significance’ of findings, identify and measure the resultant trends and report on aggregate confidence levels.
In doing all these things, the statisticians apply established mathematical methods of general application, which can be understood and/or challenged by lots of people outside the cloistered halls of public climate science. There will be nothing mysterious in these established methods.
So, the provenance of New Zealand’s adjustment bible is respectable enough. But it still has its mysteries, just like the ecclesiastical version. Here are some of them:
(1) How did NIWA’s 1992 adjustments apply the methodology of Rhoades and Salinger, which wasn’t published until the following year? Did they work from a draft? Or did they use only the Salinger-designed process without the Rhoades-influenced execution?
(2) Why did almost all the 7SS adjustments push the trend in the same direction (in favour of warming), rather than largely balance out, as contended in the Peterson Review?
(3) If the same methodology was used in both the 7SS and the NZT7, why were the resulting adjustments so different?
(4) Why couldn’t NIWA’s Review immediately report the confidence levels of the NZT7 adjustments, rather than say (page 5) that “further research is under way”? The assessment of uncertainties should be an integral part of the process. Further work shouldn’t be necessary.
(5) If home-grown peer-reviewed scientific papers are holy writ, why isn’t the Hessell paper accorded similar adoration to R & S?
Views: 40
Pingback: New Zealand Media – A Scandalous Silence | New Zealand Climate Change