The Temperature Of The UK Over The Last 100 Years (part 6)

A butcher's at daily maxima and minima over the last 100 years. Are things getting worse, and what does ‘worse’ mean?

In part 5 of this series we ate another helping of spaghetti Bolognese, this being the colourful plot of tmin anomalies for the 34 UK stations in my sample, tmin being a lazy way of writing mean minimum daily temperature. The Bolognese was reduced into a clarified slide that revealed a rather wobbly grand mean anomaly. Night temperatures have certainly been getting warmer, but they’ve also been oscillating between warmer and cooler phases which cannot easily be explained by such simplistic notions such as carbon = warming.

After sticking a straight line through all that deliciously curvy behaviour we arrived at a warming rate estimate of 0.8°C per century for the mean minimum daily temperature as measured at 34 UK stations over the period 1865-2021. This is a lot less than the 1.5°C mean global warming offered by the IPCC in Special Report 15 and so we donned our sleuthing hats to go and find the reasons why.

We started by picking cherries and boy what ripe cherries they were! In next to no time the IPCC went from the kitchen fridge (The Little Ice Age) as their reference period to use of non-existent sea surface temperature data prior to 1900; and I promised to continue sleuthing. In this article I’m going to try and replicate the IPCC estimate by carefully selecting time periods so we may all see how a grand illusion may be performed. Get the coffee on and get the biscuit tin open…

Time Lords

The IPCC and experts who shovel model outputs their way are excellent time lords. Choose the right data period and you get the ‘right’ answer, choose the wrong data period and you get the ‘right’ answer, as it were. So far we’ve established that the official pre-industrial reference period for global modelling is 1850-1900, this being tainted by the end of the ‘Little Ice Age’, and we noted that few land-based stations were in operation back then, with even fewer ship records.

We now need to establish the period used for the post-industrial era, and this turns out to be the 30-year span ending in 2017. Let’s take a look at my tmin and tmax anomaly charts for my sample of 34 UK weather stations to see how these periods sit in relation to the data:

My series starts in 1865 and not 1850 so I’ve just plonked down the end point for the pre-industrial reference period, this being a grey dashed line. First impressions as a statistician of 38 years standing are that you couldn’t cherry pick two 30-year periods better if you tried!

N.B. In case folk are not familiar with the term cherry picking, in regard to time series analysis this is the popular act of carefully choosing periods to support your hypothesis. In this instance we need the coolest period possible followed by the warmest period possible to maximise the difference and thus ‘prove’ human-induced warming.

At this point I could resort to some fancy modelling but all I am going to do this morning is compare the overall mean anomaly for 1865-1900 with the overall mean anomaly for 1988-2017, thence to derive differences. For tmax this yields a warming of 1.05°C and for tmin this yields a warming of 1.15°C, and so we see my estimates using real world data approach the simulated 1.5°C of the IPCC.

This isn’t at all bad considering I’m only using land-based data and not combined land and sea surface temperature data, and I’m only using 34 UK weather stations instead of several thousand global weather stations. Neither am I using a multi-million dollar simulation but we’re roughly in the right ball park, folks!

Given that I’m in the right ball park we can deduce that less cherry picking on behalf of the IPCC would likely result in a lower global estimate for so-called “human-induced warming”. This is why I call them the Time Lords - they ain’t going to analyse data that will give them the wrong answer!

I’ve had a go at summarising the situation in a single annotated slide:

Now that’s the way to stack your deck - there’s no way you can fail to find a difference! Run all this within Earth systems simulations that blames those changes on humans, that feeds into a political organisation hell-bent on promoting the anthropogenic climate above all else and there you have it. Job done.

Ideally, and being courageous modellers, they’d now go and choose two trickier bits from the middle. If their climate simulations were up to scratch (and the theory 100% correct) they should still be able to come up with similar estimates for the rate of human-induced warming; that is to say their theory should hold for all historic periods and not just the cherry-picked ones.

Hairy-Scary Research

Now that’s the sort of poking-around, hairy-scary research I’d like to see embodied in all IPCC reports. But we don’t see it and there’s a jolly good reason why:

This is taken from a document called Principles Governing IPCC Work, which may be found here. In a nutshell the IPCC was not established to further understanding of climate science but to promote a single idea. It’s also made very clear that outputs are influenced by government suits who then use these important documents to further their own aims.

How Red Are My Cherries?

Whilst it’s pretty clear that the two data periods in question will procure a whopping great rise in the tmin mean anomaly, we may ask if 1988-2017 is the reddest cherry the IPCC could have chosen. We could spend hours fiddling with different data periods and regressions so what I’ve done for baking speed is stick with 1853-1900 as my pre-industrial reference and slide a 30-year period along every 5 years from 1903-1932 all the way up to the IPCC post-industrial period of 1988-2017. If we calculate the net change in the tmin mean anomaly from pre- to post-industrial periods and get the crayons out we end up with this slide of slidings:

It transpires that the IPCC didn’t choose the very juiciest of cherries - this being the period 1978-2007 - instead they chose the second juiciest cherry. It would be a delicious dream to mandate them to re-work SR15 using 1983-2012 as their post-industrial period, or even 1968-1997 but they’d probably run a mile if a mandate came their way to test their Earth systems models using 1958-1987!

A Weed-Free Narrative

In his refreshingly candid and most informative talk Solar v Anthropogenic Causes of Global Warming Dr Nir Shaviv drops this little slide in at the 16:40 mark that serves as a fine example of IPCC gov-suits weeding out that which muddies the narrative:

What we have here is a slide from an early draft of IPCC’s fourth assessment report that summarises the outcomes of two modelling strategies for human-induced warming. To the right is a sharp hump depicting a narrow range of warming outcomes if Earth systems simulations are restricted to a handful of easily understood factors. To the left is a broad hump produced by the very same simulation systems if modellers dial in a few more factors. What Dr Shaviv has done with his green arrow is point out that a proportion of those broader scenarios predict global cooling even if we continue to use fossil fuels at an exponential rate. What?!!!

That’s right, those fancy Earth systems models that once ran under CMIP4 and CMIP5, that are now running under CMIP6, will negate any notion of human-induced warming if you allow them to become more complex. Needless to say this slide was dropped from the final report, the decision no doubt being taken by eager gov-suits keen to please and rise through the ranks.

With that I shall wrap this newsletter up and ready myself for part 7, in which we pop back to analysing the mean daily minimum temperature record (tmin) and ask where all the recent warming has gone. Until then…

Kettle On!