Hot, Hotter, Hottest (part 1)

A mini-study in which I take samples (S=50) of GHCNd stations from 15 locations across the northern hemisphere to determine regional trends in maximum daily temperature

This morning Mrs Dee asked me if the UK is to enjoy a warm summer and I confess to throwing my arms about and stating that I didn’t know. I then explained that, according to my Google news feed, the UK is going to suffer another apocalyptic heatwave (the Met Office know the exact hour and day this will start) but that we’re also going to have a month’s rainfall in one day and that summer is going to be a damp squib, but at the same time heatwaves are going to become a regular feature of a miserable life. The same feed provides scare stories about people cancelling expensive holidays abroad because of the heat; it also frightens folk about a terrible heat beast named Cerberus that is on the loose, and that China has all but melted under a record-breaking umpteen-day heatwave. I’ve even seen old school references to global warming (remember that?) instead of climate change! The Guardian newspaper leads the way as usual, being backed by the Met Office.

As I sip my morning lemon tea I realise we have no idea whether any of this is true. As I’ve stated before, it is possible to continue breaking records on a cooling planet. Re-defining heatwaves as blocks of three days instead of seven days sure helps push the count up too! Then there’s urbanisation of the global station record, along with observation bias (changing the station sample over time), and omission of error ranges for temperature estimates such that a new record is claimed even though the extra 0.1°C falls within the margin of error. Then there’s the murk that some are now calling ‘captured science’ – just who do you believe these days?

TBH I don’t believe words uttered by any science-based organisation with government ties and/or funding from billionaire philanthropists largely because of my own experience as a research manager within the public and private sectors, and especially so if the legacy media bolsters the narrative on a daily basis. The hard sell is evident.

Plat Du Jour

This brings us to the plate of the day. I decided to concoct a mini-study whereby I divide the northern hemisphere into 15 regions, plonk down a coordinate centroid in the middle of each region, then get KNMI Climate Explorer to trawl through the entire GHCNd database looking for the nearest 50 stations with maximum daily temperature records spanning at least 60 years. Climate Explorer was used to crunch the mean maximum daily temperature for each sample of 50 regional stations, and the resulting 15 time series were bunged into an Excel spreadsheet ready for baking. My stats package was then used to boil the daily series down into weekly, monthly and annual means, at which point the crayons come out.

We shall start with a simple tabulation showing the regions, their centroids and some basic summary stats for the mean maximum daily temperature (tmax):

You can fiddle with Google Earth to see where I’ve plonked the regional centroids. I’m happy to shift these about but before splitting hairs it’s worth realising that the station sample will cover a sizeable area so alterations will be entirely academic!

It’s worth noting that stations across China came online within GHCNd in 1951, so all those claims of a record-breaking heatwave need to bear this limitation in mind. The same goes for South Asia, whose stations started to roll out in 1943. I’m not sure what is happening with stations in Central America but GHCNd, as accessed via Climate Explorer, isn’t showing any data after 2013.

I’ve taken the WMO climate normal period of 1981 – 2010 to crunch some basic stats for each region over this common period. Please do remember that we’re looking at tmax data, so the column headed ‘minimum’ pertains to the average minimum daytime attained and not the average minimum nighttime, which would be much cooler. I may well consider nighttime temps (tmin) in a future article so we get to see both ends of the spectrum.

The column headed ‘mean’ is the mean maximum daytime temperature for the sample of 50 stations over the climate normal period and is the statistic that I tend to focus on for comparison. In terms of daytime heat experienced by folk South Asia gets the gold (31.71°C), with the silver going to Central America (30.34°) and the bronze going to North Africa (28.22°). This surprised me somewhat for I had bet good money on North Africa romping home. It was then I realised that nobody in their right mind would build a station in the Sahara desert, with stations being located in pleasant coastal parts of Algeria, Niger and Tunisia.

Variation = Record Breaking

A summary statistic that is often overlooked is that of the standard deviation, which tends to cower in the shadow of the almighty arithmetical mean; and this despite us being concerned with something called climate change! It baffles me just how many stats are churned out by climatologists each week that consider the mean of this or that, or the median of this or that, or the smoothed wotsit of this or that when we’re essentially dealing with a dynamic situation. I’m guessing this is because when you start to look at climate variability over time then a great of the alarmist narrative falls apart and all that linear ideology becomes cyclical. You can’t easily frighten people into submission with mention of cycles – only the biggest and straightest sticks will do!

We find Canada winning the gold medal for the greatest variability in the daytime maximum temperature record (14.28°C), followed by Russia (12.60°C) and the USA (11.12°C). This should not surprise us because these massive regions straddle some pretty hefty climate extremes and are subject to some pretty dramatic weather. In contrast nothing changes that much in South Asia (2.27°C).

If anybody is thinking there may be a relationship between mean maximum daytime temperature and the standard deviation of maximum daytime temperature they’d be right – cop this:

That’s a beauty! We now see a fairly robust negative relationship (p=0.002) that indicates a greater range in the maximum daytime temperature for cooler regions. Hot regions are hot and nothing much budges from hot. When it comes to the issue of hot, hotter, hottest this is one of those bottom-of-the-trifle-bowl basic analyses we can undertake yet I’ve yet to see an expert mention the fact that hot countries in the northern hemisphere vary less in their maximum daytime temperatures – all the action happens at the cooler end!

If we now realise that record-breaking is a function of variability (a.k.a. standard deviation) - in that the more variable a climate is going to be, then the greater the likelihood of a record being broken every once in a while - then we can see that records are more likely to tumble in the cooler climates. This is why we get fed a diet of records being broken in the UK, China, Russia, Canada, Greenland, Iceland etc – these places are always going to support the alarmist narrative owing to their variability. In sum, what we used to call weather is being re-branded.

Next up we ought to look at the annual means derived from the maximum daytime temperature data for the period 1880 – 2022 to get a handle on warming trends. Anomalous years arising from poor data capture (less than 274 daily observations per year) have been excluded from analysis.

Three Hot Regions

This is utterly fascinating.

Those early outliers for North Africa and Central America appear to be genuine in that data capture passed my 274-day rule but who knows what was going on at the one or two stations that would have provided this data! The climb to sweltering heat between 1922 and 1950 is evident in two regions and so is a phenomenon I am inclined to believe. We’re talking a ~5°C rise in the mean daytime maximum over a 28-year period, which is equivalent to a warming rate of +17.9°C per century. Wowsers! Now that is what I call global warming, and you’d think this early period of soaring regional temperatures might have been mentioned by climatologists somewhere. Fat chance; instead they witter on about the meaningless IPCC global estimate of 1.5°C per century.

Aside from that hot blip for 2016 it sure looks to me like South Asia was hotter back in the mid-‘40s than it is today. In fact, if I run a linear regression through the data cloud for South Asia I end up with a statistically insignificant warming rate of +0.29°C per century (p=0.292). South Asia ain’t warming, folks and that’s according to data supplied by NOAA.

Apart from the significant rise through to 1950 Central America looks like it might be warming a little more since then and this is confirmed by another of those linear regressions, with an estimate of +2.65°C per century (p<0.001). This could be real but it could also be down to urbanisation surrounding the Stevenson screens. Ideally we’d dig around a bit but alarmists don’t do that sort of legwork – I guess because it takes time away from sloshing soup on artworks.

The North African pattern from 1960 onward points to insignificant warming estimated at +0.22°C per century (p=0.722), with a hump in recent years that mirrors one back in the mid-‘40s.

Not exactly a clear picture of AGW is it? South Asia and North Africa are not warming and Central America might be warming. These are the three hottest regions in my sample folks, and yet there’s precious little evidence of man-made global warming. Looking at the way these curves lurch about I’m thinking along the lines of varying oceanic and atmospheric oscillations driven by something other than a steady stream of CO2 from next door’s SUV.

Five Middling Regions

What’s with the huge dip in maximum daytime temperatures for India and the Middle East? I smell natural climate variability that has changed these regions by ~2 - 3°C in the space of next to no time. That being said it could be the usual station fun and games. Bolt those dips onto the series from 1950 onward and you’ve got yourself a stonking warming trend by design!

And how about those temperatures for India prior to 1930? This is the sort of thing alarmists like to ignore, with any number of flakey reasons given. Yes, these may well be artefact but they also may well be genuine – we have no way of knowing the truth of the matter other than to flag doubt in a field that is proud to call itself settled.

Whilst scratching our head we better note the curious bump that marks out 2016 – this is the hot blip we have seen for South Asia and North Africa, which suggests a real world event rather than a data whoopsy. We’re back to thinking about oceanic and atmospheric oscillations impacting on land surface temperature measurement – in this instance of the order of 4°C. Again, the IPCC estimate of +1.5°C per century pales in comparison and at this point it is worth noting that the IPCC’s remit was to quantify man-made climate change, not to understand the role of natural processes.

At this stage we may well ask what the warming trends are for these five regions if we ignore everything prior to 1960 and accept the recent bumps as a sign of things to come instead of being… er… bumpy. Linear regression yields the following:

According to the current climatological notion of what goes up must go up then India, the Middle East and Southern Europe are heading for trouble. Except I doubt this very much: apart from the bumps likely being bumps, then we have a vast land mass (China) that isn’t warming anywhere near as much despite being on the same CO2-laden planet. I know America is on the same planet because I have friends living there, but what they may not appreciate is that the America surrounding them is not warming (p=0.629), whilst at the same time Southern Europe is about to explode into a crackling blaze. You’d think lack of warming across America would be big news, and it is in climatological circles where people are permitted to think.

Three Cooler Regions

We finally arrive at the classic warming profile you’ll see touted by experts. Nothing happened for 82 years, then it got slightly cooler for 24 years, then, suddenly man-made warming was born in 1986. Linear regression for the period 1960 – 2022 yields the following:

We’re back to roasting everybody in the near future, with Eastern Europe curiously outgunning Central Europe. The solution, as I see it, is for everybody to move to America if they can, China if not.

Four Cold Regions

Greenland looks wacko for sure, with a plummeting daytime maximum and icy-geddon breaking out in 1948. I recall my grandmother telling me of severe winters and roasting summers during the war so the big green blipper might not be as artefactual as we may think. The remainder of the spaghetti looks much of a muchness so I shall choose 1960 as my cut-off year and run another set of regressions:

Yet more roasting regions! Since 1960 everywhere has been warming apart from a vast region we call America and another vast region we call North Africa and another vast region we call South Asia. China is also big and the warming there has been much less in comparison to other roasting regions.

Putting The Scissors Away

Like most alarmists I have assumed so far that people couldn’t read a thermometer prior to 1960, so I’ve put the scissors away and permitted each time series depicted in the above slides to reveal their overall grand warming rate using linear regression. This is a rather inadequate approach to the matter but is the trick adopted by nearly everybody who likes to think in straight lines. Periodicity, natural cycles and saturation effects are all thrown out of the window once we reach for trends! With a deep frown I thus hammered out this summary table:

There appears to be two distinct groups of regions – those that exhibit gradual warming at around 2.5°C per century (10 pink regions) and those that show little or no statistically significant warming at around 0.4°C per century (5 blue regions). We’re not talking small fry regions here either with China, South Asia, India, North Africa and USA all bucking the narrative. I don’t see these facts taken up by the legacy media in their relentless broadcasting of globalist heatwave propaganda, but then again they are only doing the job they’ve been paid to do.

Kettle On!