I gave a talk yesterday, over Zoom , as you do nowadays, explaining that the Australian Bureau of Meteorology don’t really know how to measure temperatures anymore, so they take the recordings they have, and remodel them until it looks like how they think it should. That is then republished as an annual assessment of climate variability and change.
The assessment is published at the beginning of each year and advertised as an update, but the Bureau don’t always include the most relevant numbers – if they don’t feel like it. That might sound outrageous, and it is. But then again, the Bureau does what it wants to do. That has been the case since at least November 1996, when the transition began away from measuring temperatures using traditional liquid-in-glass thermometers which are mercury for maximum temperatures and alcohol for minimum temperatures.
Meteorologists often refer to the average temperature, but it would be misleading to think this is the average of all the temperatures measured say every hour over a day, or a month, or a year. Rather it is the highest temperature, and the lowest temperature as measured each day. These two extreme values are added together giving the daily mean.
Before November 1996, when most official maximum temperatures were measured with a mercury thermometer, the highest temperature was less extreme than it is now. This is because mercury in a glass tube, which is the essence of a mercury thermometer, takes some time to adjust to air temperature, so there is an amount of inertia.
Nowadays, the Bureau takes temperature readings every one second from what are called resistance probes and the very hottest of these instantaneous values is designated the maximum temperature for that day.
It is actually slightly more complicated.
If you will indulge my interest in all things technical – and concentrate intensely on what I write for just a moment or at least slightly longer than it took a probe to record the maximum temperature at the airport in Cordoba, Spain, a couple of weeks ago – a new record maximum for Spain for April. That was the media headline. What was not explained, because it spoils the story somewhat, and because people have trouble concentrating on detail, is that the temperature was recorded at an airport with the probe located between the runway tarmac and the taxiing apron.
In the olden days, maximum temperatures were recorded from places like botanical gardens by astronomers who sometimes also kept their telescopes there. They were fascinated with the natural environment and continually looking for new ways to measure it. Nowadays, it is the case that airports are the primary site across the world for recording temperatures, that are reported by meteorologists, often on the nightly news, with yet another record-breaking hot day more likely to increase viewer engagement. So, it is at the end of the day, month, and year that the weather report is analysed in terms of increased traffic volume metrics (for example clicks) rather than the meteorologist’s skill at predicting, for example, how much rain or snow actually fell.
This relatively new imperative is aided by not only having the probes at airports, where more frequent take offs and landings by ever more powerful jet engines, will continually increase the likelihood of a blast of hot air being recorded as the hottest day ever, but also by all the electrical noise.
For sure, I need to explain how this works.
A mercury thermometer is completely impervious to the beat of airport radar that may move across the tarmac every 30-40 seconds. Not so, what are referred to as ‘low noise’ amplifiers attempting to precisely measure the change in resistance of the platinum electrodes in the resistance probe: that is how the probes measures temperature. These electrodes may not only be energised every time the radar sweeps over them, but also by the chatter from a pilot in the airport’s control tower.
As an analogue engineer recently explained to me, because of all the radio interference at airports, it is not really a place to be recording temperatures with resistance probes. Yet this is exactly where most of these temperature recording devices are now located – and not just in Australia, but across the world. So, the average global mean temperature may not only include the blast from a jet plane landing at Cordoba, Spain, but also the chatter from pilots and the control tower because temperature is now primarily measured as changes in electrical current and at airports.
Of course, those measuring temperatures using these probes at airports would be working hard to exclude radio frequencies (RF) getting into the input of the receiving amplifier and causing random errors. At least one would hope so – unless they are jokers.
The real test of this is seeing how the readings from a mercury thermometer compare with readings from these resistance probes at airports – around the world.
The original Joker was of course the arch-criminal obsessed with absurdity and joke-based crimes: clown-like in appearance and personality, a master of intrigue and escape. And, so, I was pondering the number of times the Australian Bureau of Meteorology has escaped having to show the parallel data for my hometown of Brisbane – or any of its many other airport sites – despite multiple reviews, reports, advisory panels, and peer-reviewed studies rejecting claims that its temperature record is biased or flawed. Here I am quoting directly from the The Guardian on Sunday, an article by Graham Readfearn in which he seems to delight, perhaps as the joker’s sidekick, about how many times any proper review of these measurements has been rebuked, including a proposal by Senator Cory Bernardi, that there be some oversight by the National Audit Office.
Rather than show even Bill Johnston a limited amount of parallel data for Canberra airport back in 2014, when he made a Freedom of Information request, the Bureau destroyed it, or so the story goes.
In the ‘Batman and Robin’ episodes I watched as a child on an old black and white television, the Joker was ever elusive and in the most imaginative ways.
John Abbot, who like Johnston has made FOI requests for parallel temperature data from airports, wrote in The Australian newspaper – at about the time the new record hot temperature was being recorded from Cordoba airport in Spain – explaining how rather than provide some parallel data for Brisbane airport, the Bureau falsely claimed the data never existed.
In fact, a relatively small amount of the data that the jokers claimed did not exist was eventually provided to Abbot, by the same jokers just before Easter. That was on Thursday 6th April when the jokers sent Abbot 36 separate emails to which were variously attached a jumble of scans of over 1,000 handwritten reports.
As though anyone expects to receive scans of over 1,000 hand-written reports across 36 emails when they are told they will eventually be provided with parallel data for Brisbane airport for the three years August 2019 to July 2022.
Abbot immediate onforwarded the mess to me.
I will continue this story as part of a series I’m calling ‘Jokers, Off-Topic Reviews and Drinking from the Alcohol Thermometer’.
On the topic of alcohol following is how many times the Bureau has changed the alcohol thermometer measuring minimum temperatures at Brisbane Airport over the last two decades. One thermometer used to last 100 years.
A JUMBLE OF METADATA
Year Opened 1992/ Minimum Temperature
14/FEB/2000 INSTALL Thermometer, Alcohol, Min (Type Dobbie S/N – 18985) Surface Observations
06/MAY/2008 INSTALL Thermometer, Alcohol, Min (Type Dobbie S/N – 19048) Surface Observations
04/JUN/2012 INSTALL Thermometer, Alcohol, Min (Type Dobbie S/N – 19266) Surface Observations
06/MAY/2008 REMOVE Thermometer, Alcohol, Min (Type Dobbie S/N – 19266) Surface Observations
24/JUL/2020 REPLACE – (jtaylor2) Thermometer, Alcohol, Min (Now Amarol S/N – 33803) Surface Observations
05/AUG/2020 REPLACE – (jtaylor2) Thermometer, Alcohol, Min (Now Amarol S/N – 33849) Surface Observations
10/MAY/2019 REPLACE – (jtaylor2) Thermometer, Alcohol, Min (Now Dobbie S/N – 2546) Surface Observations
03/SEP/2015 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 19048) Surface Observations
24/NOV/2019 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 19048) Surface Observations
08/MAY/2013 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 23264) Surface Observations
06/MAY/2015 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 27637) Surface Observations
26/AUG/2015 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 27637) Surface Observations
19/MAR/2016 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 27637) Surface Observations
09/JUN/2013 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 27654) Surface Observations
29/MAY/2020 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 32894) Surface Observations
09/AUG/2020 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 32894) Surface Observations
29/JUL/2021 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 33820) Surface Observations
03/DEC/2020 REPLACE Thermometer, Alcohol, Min (Now Amarol S/N – 33824) Surface Observations
09/JUN/2004 REPLACE Thermometer, Alcohol, Min (Now Dobbie S/N – 19048) Surface Observations
21/DEC/2005 REPLACE Thermometer, Alcohol, Min (Now Dobbie S/N – 19266) Surface Observations
10/JUL/2015 REPLACE Thermometer, Alcohol, Min (Now Dobbie S/N – 2546) Surface Observations
15/MAY/2016 REPLACE Thermometer, Alcohol, Min (Now WIKA S/N – 43095) Surface Observations
B. ANNUAL MAXIMUM TEMPERATURES FROM BRISBANE AIRPORT
Notes, and links, to some more of the topics mentioned above and much thanks to energy economist Alan Moran for organising the Zoom, and I will forward my notes in due course to you for the coal miners.
1. I confirmed on 3rd March 2021, that the Australian Bureau of Meteorology admitted, as I surmised in a blog post on 10th February, that the reference value for 2021 was not actually included in its calculation of the amount of warming as published in the 2021 Annual Climate Statement, more here; https://jennifermarohasy.com/2022/03/australias-broken-temperature-record-part-3/
In short, we have a 2021 Annual Climate Statement that does not include the new 2021 value in its calculations.
2. Paul Homewood sent me an email about the new hot day in Spain, and his email linked to a BBC article that you can read here: https://www.bbc.com/news/science-environment-65403381 . Homewood has a website and you can sign up for his emails.
3. I was so disappointed that Graham Readfearn got his article published last Sunday in The Guardian, quoting Ailie Gallant from Monash University. https://www.theguardian.com/science/2023/may/07/climate-scientists-first-laughed-at-a-bizarre-campaign-against-the-bom-then-came-the-harassment#:~:text=For%20more%20than%20a,Bureau%20of%20Meteorology%27s%20temperature%20records
4. Regarding Bill Johnston, you can read more about the Canberra airport saga here: https://joannenova.com.au/2017/08/another-bom-scandal-australian-climate-data-is-being-destroyed-as-routine-practice/
5. It is the case that John Abbot had some detail of the history of his interactions with the Bureau published by THE AUSTRALIAN and then republished by the Institute of Public Affairs:
‘BOM makes heavy weather over temperature data’ ’tis here https://ipa.org.au/research/climate-change-and-energy/bom-makes-heavy-weather-over-temperature-data
Great Job Jennifer. As well as what you mention there are more. Johnson noise is the random yet ever-present generation of electricity by thermal motion of electrons inside resistances. This electrical noise is there even when they are powered down or disconnected. Rather than being from external interference, it is unavoidably created inside components of electronic systems and damages accuracy of readings. Being random AC not DC, over the long term it averages to nothing. It is removed by gathering enough readings for it to cancel its own effects when the readings are averaged or low pass filtered. By using just one or even a low number of samples for the average it is invited in to cause random errors. Any system that does not take an average of enough readings is doomed to be generating values that are all over the shop with poor repeatability.
Henry Engelking says
How is it permitted to destroyed public records? What possible rationale reason was given for destroying these records? “Rather than show even Bill Johnston a limited amount of parallel data for Canberra airport back in 2014, when he made a Freedom of Information request, the Bureau destroyed it, or so the story goes.”
Don Gaddes says
Two weeks ago, the 1 Year (Minor) Earth/Solar Orbital Dry Cycle Was ending over Portugal, circa(10 degrees West.) Thus, for April 2023, the passage of the Dry Cycle resulted in increased temperatures over Spain and associated Longitudes. This Dry Cycle started over the North Island of New Zealand, (170 Degrees E) on May 1, 2022 – and the vanguard moved Westward at 15 degrees Longitude per 30 Day/Night Interval Month.
Eight Months into the Minor Dry Cycle, (January 2022,)the Wet/Normal Period returned – starting from 170 degrees East and replacing the Dry Cycle at 45 degrees of Longitude per 30 Day/Night Interval Month.
Richard Bennett says
This post by Jennifer highlights the importance of keeping copies of all raw temperature readings in an archive that is separate from official BOM archives. When the BOM “loses” all the inconvenient temperature truths then reputable scientists like Jennifer Mahrohasy can recover the true data from independently held archives.
Much thanks to Charles for reposting and to those who have taken part in the informative discussion following at WattsUpWithThat:
David Foster says
Very interesting, especially the point about inertia in mercury thermometers versus no inertia in electronic ones.
Regarding the RF interference issue, not all airports have radar by any means. And I would think the strongest possible effect would be from surface surveillance radar, which is intended to show traffic on the runways and taxiways, rather than airport surveillance radar, which is focused on displaying in-flight aircraft.