From The Daily Sceptic

By Chris Morrison

A bitter row has broken out following Dr Roy Spencer’s backing of the UK Met Office’s controversial temperature record amid accusations of Met Office bias and poorly sited weather stations.

Dr Spencer asserts that his simple bias reduction modelling of daily high (Tmax) and low (Tmin) UK summer temperatures over the last 65 years produces variations in average temperature “essentially identical” to those published by the Met Office. But there are those who point out that coming to the same conclusion by using two different ways to measure the same faulty data doesn’t alter the fact the data are faulty in the first place.

Ray Sanders, who has undertaken a forensic examination of almost all the Met Office’s 400-plus nationwide temperature measuring sites, is unimpressed, calling Spencer’s findings, “a load of twaddle”. Speaking to the Daily Sceptic, he added: “The whole point is that you can torture numbers as much as you want, but if they were no damn good in the first place, they will never get any better.”

His criticism strikes at the heart of Dr Spencer’s work. Spencer argues that a measuring station can be placed in a non-natural, anomalously warm urban environment, but as long as that environment stays the same over time, “it can probably still be used for climate change monitoring”. He went on to state that it “might be” that there has been little additional urban heat warming in London over the last 200 years, something that seems highly unlikely given that the city has grown in the period from a low-rise centre with numerous surrounding small villages to a major metropolitan area full of everything from tall glass buildings to airports. Sanders says that Dr Spencer’s opinions on UHIs (Urban Heat Islands) are “risible”. Simply saying things were bad in the past so they can be compared to now – with the caveat of ‘if things haven’t changed’ – is irrelevant, he states.

Spencer’s model starts with a baseline of three sets of data from sites that are said to have a 126-year continuous temperature measuring record. They are the Central England Temperature Series (CET), a group of three stations, Armagh and Stornoway Airport. This is the first problem. The CET has shown a number of station changes in that period, while the Stornoway record was compiled at another site nearly three miles away until its re-location to the Airport in 1968. Spencer’s high/low calculations go back to 1960 so the Stornoway move, unhelpfully not identified by the Met Office, will likely affect his figures.

The current CET portfolio includes two CIMO Class 4 site with international unnatural ‘uncertainties’ of 2°C, a junk classification also shared by Armagh and Stornoway. In addition, Paul Homewood has discovered that the Met Office has retrospectively cooled the near past and boosted CET readings from the last 20 years. Sanders has done considerable research on all the junk stations and dismisses them as “truly crap sites of incredible bad provenance”. For good measure, he notes that the Armagh weather station used to be housed in a metal box bolted to the side of a building by a first floor window.

Once he has established his initial baseline, Spencer incrementally adds further stations which adjusts the overall composite average. Each time a station is introduced he adds or subtracts its average difference (the bias) over the overlapping years until it lines up with the baseline. Further and better particulars can be obtained from his original article here. This method assumes long-term biases at each station stay mostly the same over time and using that assumption seems to produce the same warming trends as the Met Office. The uncharitable thought, ‘garbage in, garbage out’ springs immediately to mind.

Something that does not seem to have been considered in Spencer’s work is the effect of the change over the last few decades of measuring devices from liquid in glass manual thermometers to automatic electronic systems. This is a largely unaddressed issue in meteorological science but it is highlighted in recent work compiled by Dr Eric Huxter. Looking at the daily Tmax published by the Met Office throughout last May, he found that on average about 0.89°C was added to the maximum recording by temporary and most likely unnatural blips that would not have been picked up by slower-reacting glass bulb devices. Dr Huxter has told the Daily Sceptic that he is confident that electronic devices are biased to recording higher Tmax temperatures, especially given siting issues.

Since the Met Office picks up its daily Tmax from the one minute record, one way to remove some of the distortion would be to average the high over a longer period. Thus the national record declared at RAF Coningsby on July 19th 2022 at 3.12pm falls from a 60-second spike of 40.3°C to a five-minute average of 39.96°C. Over 10 minutes, a possible time frame for movement in liquid in glass bulb thermometers, the fall is 0.5°C. Still hot of course, but not BBC-ready ‘record’ headline hot.

Dr Roy Spencer is a brilliant sceptical scientist who is behind the valuable UAH satellite temperature record – an uncorrupted series that since 1979 shows less global warming than (frequently-adjusted) global surface datasets. He is one of the authors of the excellent climate report published recently by the US Department of Energy that has helped push back the notion of ‘settled’ political science. He is also an authority on urban heat island corruption. In a recent peer-reviewed paper with fellow Energy Department report author Professor John Christy it was calculated that the UHI effect accounted for about 22% of the observed summer warming across all US stations since 1895. For some of the suburban and urban locations – stations that Ray Sanders might refer to as “crap” – the UHI warming bias was no less than 65%.

Chris Morrison is the Daily Sceptic’s Environment Editor. Follow him on X.

I would like to wish all my readers a very Happy Christmas and a prosperous, sceptical New Year. I am enormously encouraged by all the support you give me during the year and the fantastic contributions that some of you make when you respond to the publication. Your input often leads me to think harder and sometimes follow up leads that I had not previously considered. Such support and contributions are a great help to those of us who labour at the copy face. Thank you.