HADCRUT Global Temperature Anomaly (part 1)

A quick squizz at the latest data with a side order of cogitation.

The HADCRUT global temperature anomaly combines the CRUTEM5 land surface air temperature and HADSST4 sea surface temperature anomaly data series in order to get some idea of what the world is doing. Details on this stalwart effort by us Brits may be found here. NOAA does similar, as does Berkeley (BEST), ECMWF (ERA5) and NASA (GISS). The work involved in deriving these time series from raw data is mind-boggling; neither should we underestimate the amount of adjustment, estimation, modelling and infill that takes place.

A member of the public looking at these magnificent time series can be forgiven into thinking they are looking at real world observations made over time. Certainly, a real temperature measurement at some place at some time gets the ball rolling, but these individual location measurements are then subject to a great deal of processing and come with a stupendous amount of bias. I’ll be taking at closer look at this in future newsletters

All of these major data players use the same basic method. The surface of the globe is divided into a large number of grids, then a series for each grid is derived from whatever nearby weather station, ship or buoy can be sensibly utilised. The surface of the Earth is vast in comparison to the total number of observation points and so the major players have to get very creative in how they fill the many holes. Again, I shall be taking a closer look at this in future newsletters.

The key thing to realise here is that it is impossible to derive a comprehensive measure of Earth’s surface temperature that dates back +100 years and yet remains consistent in methodology. We’d need millions of observation points all measuring the same thing in the same rigorous way without fail, day in day out for +36,500 days. That never happened, so all we can do is estimate the historic stuff from the patchy data we do have. Here’s Professor Phil Jones of the Climatic Research Unit of the University of East Anglia chatting about this in a leaked email:

We can take this text two ways: we can either declare in alarming fashion that CRU/UEA are making things up or we can acknowledge that there are times when we have to make educated guesses to fill the many holes that are par for the course if we are to salvage anything from history. All this and more needs to be borne in mind when we try to draw firm conclusions from a wiggly line on a graph.

My Version Of Events

You could write a book on the derivation of HADCRUT5 with its strengths and weaknesses so I am going to put my own little spin on things, starting with a nice plot of the in-filled series with estimated uncertainties:

Take a look at those dotted grey lines marking the envelope of uncertainty (95% confidence interval). From 1965 onward it looks like we’ve been pretty certain about what we are measuring, largely owing the plethora of observation points that got started. Whilst this is good news it is also bad news in that a land station that opened in 1966 will not have a record stretching back to 1850: we simply have no idea what it would have recorded back then. The assumption made is that it wouldn’t have done anything wildly different from other stations that were in continuous operation from 1850 onward. Whilst this is a sensible assumption it remains an assumption. In a future newsletter I shall reveal how this assumption can lead climatologists astray.

Old Is The New Normal

I have deliberately added (1961-1990 normalised) to the title to remind us that the Hadley crew decided to take this period as the reference climatological normal. When climatological normals were first devised the idea was to take a 30-year reference period for measurement of weather events, thence to pop the period along by 30 years to update the baseline. Strictly speaking the new normal should be the period 1991-2020, which should have become the international standard in 2021. This didn’t happen and I reckon subscribers will have already figured out why.

The Hadley crew are not only aware of the changes brought about by revision of the climatological normal but have also thoughtfully provided a comparison on their landing page. Herewith a screenshot:

By using the cooler earlier period as a baseline the anomaly peaks at around +1.25°C instead of +0.90°C. What they haven’t done is include (c) Anomalies relative to 1991-2020 and give us an idea of what would transpire if we stuck to the rules of the climatological normal. I have thus done this for them in a slide that permits direct comparison:

We now observe a peak anomaly bouncing around +0.30°C instead of +0.90°C. To my way of thinking revising the climatological normal would be a cheaper and quicker way of avoiding that dreaded +2.00°C IPCC target than the political and economic circus in which we are all now embroiled. Another way to view this is we can make two topologically equivalent statements: the Earth is hotter these days; or: the Earth was cooler back then. Journalists and politicians prefer the former for some reason.

Some may consider this facetious but it enshrines messages about impact and adaptability. We’re now at +0.90°C and climbing, with the Earth showing signs of increased greening, as plant food (carbon dioxide) becomes more abundant and temperatures more favourable. I’ve never lived on a cold planet but my ancestors did and I gather it was pretty grim. Anybody who wants the planet to cool should first try growing carrots on a glacier.

I Beg Your Pardon!

But the HADCRUT5 evidence is right there before our very eyes! Since 1965 the globe has plunged into near perfect linear warming and this isn’t going to stop unless we stop using fossil fuels. What tea are you drinking Mr Dee?!!!!

Two Curious Wrinkles

Before I change my blend of tea there are a couple of curious wrinkles we need to consider and we shall start by downloading atmospheric CO2 concentrations derived from in situ air measurements at Mauna Loa, Observatory, Hawaii, then plotting HADCRUT5 monthly global mean anomaly against monthly atmospheric CO2:

Now that is what I call curious! Back in the early part of this time period we see a negative relationship between atmospheric CO2 and global anomaly; that is, as atmospheric CO2 increased, the global anomaly decreased. The greenhouse gas theory of global warming only started making sense from concentrations of 330ppm CO2 onward, a fact that we seem to have overlooked!

Just after 380ppm CO2 the theory starts to run out of steam with little change in the global temperature anomaly despite increasing levels of CO2. Just to rub it in, we observe what appears to be a reversal from 410ppm CO2 onward with anomalies stagnating or even in decline. There’s a lot I can say at this point but this will have to wait for future newsletters: all we need to realise right now is that nothing is straightforward despite the impression given by those experts with vested interests.

If that slide doesn’t get folk scratching their head the next one will, for it is a cross-correlation plot of the monthly Mauna Loa atmospheric CO2 series with the HADCRUT5 anomaly:

What this analysis is telling us is that over the period 1958 - present the monthly temperature anomaly generally started to rise before the atmospheric concentration of CO2 once we account for seasonal effects, as evidenced by that big wedge of positive values at negative lags. Mind blowing or what? This is not exactly new news since climatologists and Earth systems modellers have known about positive feedback in the carbon cycle from the outset; this being the rise in CO2 as a result of the planet warming. Geologists will talk about ocean degassing and stuff that rocks do, then there’s the biosphere cranking up in response to warmer conditions. Stuff like that.

There’s a great deal more than these two wrinkles to chew on but I reckon this newsletter is long enough as it is. Wrinkles like these suggest to me that the linear thinking that leads to cult-like alarmism is somewhat naïve and probably wrong. The fact that I can sip a cuppa and type stuff into substack today means that the planet didn’t opt for runaway global warming back when it had several good opportunities to do so, which would suggest our climate is more complex than we think.

Kettle On!