- JoNova - https://joannenova.com.au -

Two-thirds of Australias warming due to “adjustments” — according to 84 historic stations

1954 Yearbook Brisbane

Here’s an update to the digging through our historic records we discussed a month ago, we can now include nearly twice the stations and the difference between temperatures originally recorded 100 years ago and temperatures today are even smaller.

Chris Gillham has been working with CSIR documents and official Commonwealth Year Books. Last month he used the 1953 Year Book which contained 44 weather station averages for 1911-40 to compare with 2000-14 temperatures, but has since discovered that the 1954 Year Book provides an additional 40 stations with 1911-40 data. The average rise in mean temperatures across all 84 weather stations around Australia over the last 70 years of global warming is about 0.3C. This larger dataset suggests as much as two thirds of the current official trend in Australian warming was due to post hoc adjustments, not heat recorded by thermometers.

These historic temperatures were calculated by the best scientists of the day, using the best equipment of the era (the same Stevenson Screen we use now). Yet again, global warming appears to have a “man-made” contribution. Far more important than CO2 is man-made “pollution” called homogenisation.

When doubling the recorded trend makes “No difference”

Bear in mind – some adjustments are necessary because raw is not automatically right. Stations have moved. But the Australian Met bureau can’t explain why all these adjustments are necessary, and indeed still claims the adjustments make no difference to the trend when clearly they do. Ken Stewart heard Dr Vertussy, the Director of the Bureau of Met (BoM), claim on radio last week that the adjustments make “no difference at all”. We look forward to Dr Vertussy’s reply.

And then there is the bizarreness of the half a degree adjusted cooling that occurs in these historic records when modern screens are much more likely to be near tarmac, bitumen, jumbo jets and 15 storey apartment blocks. The Urban Heat Island effect means that modern temperatures are artificially elevated in city CBDs by as much as 5- 7 C, with tests also showing several degrees of UHI at regional weather stations. So we have the paradox that the old records near dirt roads and horse drawn carriages were apparently reading artificially warm compared to thermometers today near roads with 10,000 internal combustion engines passing daily.

Guest post by Chris Gillham

New historic temperatures suggest even more of Australia’s warming trend is due to adjustments

The earlier analysis of unadjusted temperatures in the 1953 Year Book found a 0.4C increase in mean raw temperatures from 1911-40 to 2000-14 at 44 weather stations across Australia, but it turns out that the 1954 Year Book adds another 40 to create a pretty large network of 84 stations.

These 84 stations suggest Australia’s mean temperature increased 0.3C from 1911-40 to 2000-14.

The Year Book network also suggests that the raw mean minimum has only increased 0.1C nationally and was static or cooled in NSW, Victoria and WA. When capital cities are removed and a comparison made of only regional weather stations, the average nightly minimum is nowadays cooler in those three states than when your grandparents were born – particularly in winter.

The 84 Year Book stations all existed in 1911 whereas many of the 112 stations in the ACORN network didn’t open till decades later. Nevertheless, 29 of the stations in the Year Book network eventually became a part of ACORN and these had a 0.1C cooling adjustment to maxima and a 0.5C cooling of minima in 1911-40.

The broader network of 84 unadjusted temperature stations reinforces the findings of the earlier 44 station Year Book analysis. In fact, it suggests that if the thermometers themselves recorded a 0.3C mean warming at an historic network of 84 stations, but ACORN recorded a 0.9C warming at 112 stations since 1911, about 0.6C or two thirds of the latter is probably due to adjustment of the early records and/or the addition of hotter stations to the network as the years rolled by.

As would be expected after 1910, the Year Book text suggests all temperature recordings in the 1911-40 climate period were in Stevenson Screens. So ACORN adjustments – particularly to minima – are presumably to compensate for station relocations since 1940, mostly to nearby airports, with a UHI adjustment that at least doubles the temperature increase itself.

Summary points 1911-40 to 2000-14 (National 84 stations)

 maxima up 0.5C

 minima up 0.1C

Excluding capital cities (National 67 stations)

maxima up 0.5C

minima up 0.1C

ACORN stations in the Year Book network  (National 29 stations )

raw maxima up 0.5C adjusted in ACORN up to 0.6C

raw minima up 0.4C adjusted in ACORN up to 0.9C

The biggest cooling adjustment was at Launceston where the 1911-40 raw minimum was cut from 7.4C to 4.8C in ACORN (adjusted -2.6C compared to maxima cooling of -0.9C). Launceston had a Stevenson Screen by 1895 but the airport is 160m higher in elevation than the city which results in lower temperatures, particularly minima. The airport’s raw mean minimum 2005-14 was 6.4C and the town of Launceston’s raw mean minimum 1885-1959 was 7.0C, a difference of 0.6C.

Australia’s Year Books can be an enlightening source of unadjusted temperatures compiled by our leading meteorologists in their day to most accurately represent Australia’s climate for a global audience. They can provide an accurate average of climatic conditions at various weather stations where the BoM has lost or not got around to providing temperatures in the first half of the 20th Century.

The 1970 Year Book provides a different network of regional weather station minima and maxima for January and July, again based on the 1911-40 climate period, and 49 of these also have 1911-40 temperatures in the BoM’s Climate Data Online. The 1970 Year Book and CDO agree that the average Jan/Jul minimum at these 49 stations across the country was 12.4C and the average maximum was 24.1C in 1911-40. As it turns out, when averaged together these hottest and coldest months also show a mean raw warming of 0.3C by 2000-14 (18.3C > 18.6C).
There are 64 weather stations in the 1970 Year Book that don’t all have CDO temperatures for 1911-40 but can be compared with 2000-14. They suggest 0.5C mean warming for January and July from 18.1C to 18.6C.

Among the 1970 Year Book stations, 22 have since become part of the ACORN network. These 22 show a 0.5C January and July raw mean increase from the Year Book 1911-40 temperatures to 2000-14. In ACORN and after a 0.5C cooling adjustment to 1911-40 Januarys and Julys, they show a 1.0C increase in mean temperature for those months.

The BoM’s ACORN review panel is unlikely to have paid any attention to the submission detailing an unadjusted mean temperature increase of 0.4C at 44 weather stations sourced from the 1953 Year Book. Let’s assume they know that adjustments for UHI and airport relocations have accurately caused a majority of Australia’s climate warming, so the historic documents aren’t worth considering.

But maybe Australia’s politicians and media should consider the fact that an alternative dataset of 84 stations shows the thermometers themselves have only warmed 0.3C over the past century.


The 1953 and 1954 Year Book spreadsheets are here and the 1970 Year Book spreadsheet is here.

Current temperature sources:

9.2 out of 10 based on 112 ratings