by Warwick Hughes
Jones et al 1986 looked at 86 Australian stations and rejected 46 (25 Short term – 21 long term). Of the 40 they used 27 were short term and 13 long term. Of the long term there were 5 large cities.
The 27 short term stations were mostly only quoted from 1951 onward – regardless of what data was available. It just so happens that the years just post WWII were not prominently warm in Australia so an “automatic” warming trend was reinforced into the CRU Australian component.
Here are 11 examples where Jones et al systematically truncated pre-1951 data or ignored more rural data around many small town Australian stations. These graphics and text have been extracted from a 1992 vintage Word doc that somehow survived the decades and how many HDD’s.
The aerodrome records 1951-80 shows a clear warming trend. Marble Bar, 150 kms south east, shows a similar trend over that period but a flat trend over 80 odd years.
For the period 1951-80 this trend is sharply upward, yet if the Longreach Post Office record is spliced to the aerodrome record (post 1940s) the trend becomes markedly flatter. When Longreach is compared to Isisford, a much smaller rural centre 80 kms south, the trend is closer to neutral over about 70 years.
The A.M.O. record 1951-80 shows a clearly steeper warming trend for Mackay when compared with St Lawrence and Pine Islet Lighthouse.
Compared to Bustard Head Lighthouse and St. Lawrence, Rockhampton shows a warming trend of about 0.5°C over 70 years. Rockhampton data was used only for the period 1951-1970. The two nearby more rural sites show a similar temperature pattern but a negligible temperature change over 70 years.
The Aerodrome record 1951-80 shows a clear strong warming trend. The small centre of Cue, 120 kms south west however has a flat trend over 90 years.
The trend for this station 1951-80 also shows a strong warming trend. Cunnamulla, a smaller centre approximately 170 kms south, shows a much flatter trend over about 80 years.
The aerodrome record 1941-80 shows a well defined warming. However, when Post Office records are spliced on, the trend is much closer to zero over 90 years. Looking at Southern Cross, a continuous Post Office record, approximately 200 kms west, the trend is very similar, flat over some 90 years.
This station 1951-80, shows a steeper warming trend than the nearby Rawlinna, where records go back to 1926.
A warming trend is seen over the 1951-80 period, yet the longer term and more remote Streaky Bay, where records are available back to 1925, shows a flatter trend.
Once again, this record shows the 1951-80 warming. Broken Hill, the nearest long term station in a similar climate, shows a cooling trend over a hundred year time span.
From 1951-80 this aerodrome station shows a strong warming trend. When the Post Office records are spliced on the trend 1860s to 1990, it is close to neutral.
It gets better, in Warwick’s blog comments, Warwick points this out:
The situation for Jones 1986 Sth Hem compilation – is that Sydney and Melbourne aside, there is not one station, long or short term, between Brisbane and Mt. Gambier. This area includes all of NSW and Victoria and contains the greatest concentration of long term recording stations in Australia. Must be one of the great and complete exclusions in the history of science.
He adds in another comment:
I have never been able to discover which stations contribute to their gridded data.
Maybe that will change now.