climate change wa

It's getting warmer (more comfortable), not hotter

Contrary to popular belief nurtured by poor media research, Australia has enjoyed a decline in extreme temperatures since record-keeping began in the 1800s.

Analysis of daily raw temperature records maintained by the Bureau of Meteorology shows that most locations in Australia have less extreme cold nights and less extreme hot days than was the case a hundred years ago.

The evidence is strong of a reduction in the diurnal temperature range at a majority of Australian weather recording stations, with the greatest reduction in extreme cold nights, fewer scorching days and an increase in the average mean temperature.

Australia's weather has become slightly warmer but more docile with less extreme storms and fewer cold/hot extremes that are threatening to most plant and animal life.

The evidence can most easily be viewed in charts of all daily minima and maxima recorded at numerous weather stations around the country:

Daily maxima and minima for

Adelaide, Bathurst, Bourke, Brisbane, Bundaberg, Cobar, Dalby, Darwin, Deniliquin, Gayndah, Goondiwindi, Gunnedah, Inverell, Melbourne, Perth, Sydney, Tewantin, Wagga Wagga and Walgett.

adelaide minimum temperature history

adelaide maximum temperature history

Top of page

bathurst goal minimum temperature history

bathurst goal maximum temperature history

Top of page

bourke minimum temperature history

bourke maximum temperature history

Top of page

brisbane regional office minimum temperature history

brisbane regional office maximum temperature history

Brisbane had a Stevenson Screen from 1892 (Source)

Top of page

bundaberg minimum temperature history

bundaberg maximum temperature history

Bundaberg had a Stevenson Screen from 1892(Source)

Top of page

cobar minimum temperature history

cobar maximum temperature history

Top of page

dalby minimum temperature history

dalby maximum temperature history

Dalby had a Stevenson Screen from 1892(Source)

Top of page

darwin minimum temperature history

darwin maximum temperature history

Top of page

deniliquin minimum temperature history

deniliquin maximum temperature history

Top of page

gayndah minimum temperature history

gayndah maximum temperature history

Gayndah had a Stevenson Screen from 1892(Source)

Top of page

goondiwindi minimum temperature history

goondiwindi maximum temperature history

Goondiwindi had a Stevenson Screen from 1892(Source)

Top of page

gunnedah minimum temperature history

gunnedah maximum temperature history

Top of page

inverell minimum temperature history

inverell maximum temperature history

Top of page

melbourne regional office minimum temperature history

melbourne regional office maximum temperature history

Top of page

perth regional office minimum temperature history

perth regional office maximum temperature history

Top of page

sydney observatory minimum temperature history

sydney observatory maximum temperature history

Top of page

tewantin minimum temperature history

tewantin maximum temperature history

Top of page

wagga minimum temperature history

wagga maximum temperature history

Top of page

walgett minimum temperature history

walgett maximum temperature history

There is a notable decrease in the number of extreme cold days and extreme hot days at most locations, including capital cities where recordings can be biased by urban heat influence.

Although each stations is different, the early temperatures before 1910 may be influenced by warmer recordings in Glaisher and other thermometer stands before the introduction of Stevenson Screens.

Regardless, a majority of locations show fewer extreme hot days since uniform temperature screen recording was introduced.

Missing records

The fallacy of modern "extreme" heatwaves can be illustrated by examining temperature records from the western NSW town of Bourke in 1896 (download Word file).

An argument might be put that the Stevenson screen wasn't installed at Bourke until August 1908 so the maxima air temperature recordings of 1896 were higher than reality.

The accuracy of raw temperatures at Bourke in 1896 can be gauged through The development of a high quality historical temperature data base for Australia published in 1996 by Simon Torok from the University of Melbourne, which is the source document for BoM adjustments to raw temperature data leading to development of Australia's High Quality data series. The Bourke adjustment are below:

torok temperature adjustments to bourke

temperature adjustments to bourke

The maximum for 1896 in the top chart seems to have had no adjustment that year so it might be assumed that Bourke's weather station raw temperature recordings were considered accurate despite the absence of a Stevenson Screen.

These Torok charts for Bourke also highlight the disappearance of the hottest ever temperature ever recorded in Australia. The BoM nowadays claims the hottest ever day in Australia was 50.7C at Oodnadatta Airport, South Australia, in 1960. See here and here.

However and probably because the BoM will not accept the validity of temperatures recorded before 1910, Australia's record maximum of 125F (51.7C) at Bourke on 3 January, 1909, is discarded.

It is confirmed that a Stevenson Screen was used for Bourke temperature recordings in 1909. For example, the BoM recognised the validity of the 125F record in this Sydney Morning Herald newspaper report on 29 January 1932:

Bourke's temperature of 119 degrees was six degrees below the record of 125 degrees, made on January 3, 1909. Bourke is credited with having experienced a temperature of 127 degrees in 1877, says the State Meteorologist. Mr. Mares, but this is not accepted by the Weather Bureau because the thermometer was not enclosed in the wooden screen which is used as a standard throughout the world. These wooden screens containing the thermometer are usually placed about four feet from the ground, and permit the free circulation of air about the instrument so that the real air temperature is recorded. The screens have now been set up in most New South Wales country towns.

In other words, Australia's hottest ever temperature of 125F at Bourke was OK with the BoM in 1932 because it was a Stevenson screen recording. It is pointless trying to find the raw temperature for 3 January 1909 in the BoM temperature tables because it was a Sunday.

It's worth noting that on 6 Jan 1909, the Sydney Morning Herald reported Bourke's temperature on 4 January as 114F. That's 45.6C which, with a decimal adjustment to the rounded newspaper Fahrenheit, is the same as the BoM raw record of 45.3C on 4 January. Again, it can be assumed the Bourke recording equipment was considered satisfactory so no adjustment was needed for the raw temps.

In the adjusted maxima from the top chart above, the year 1909 has been adjusted up by almost 1C which suggests that, if anything, the raw temperature recording was considered less than the real air temperature.

So what exactly is wrong with Australia's hottest ever day of 125F (51.7C) at Bourke on Jan 3 1909? It was recorded in an almost brand new Stevenson box that was well sited and not worn by the weather.

How did the Sunday 3 January 125C get into newspapers, why was it cited in future years by the BoM, and where did it go?

A similar question might be asked about the hottest ever day recorded in Western Australia, which the BoM identifies as 50.5C (122.9F) at Mardie in 1998.

However, a maximum of 123.2F (50.7C) was recorded at the southern coastal weather station of Eucla on 22 January 1906. Note that the BoM acknowledges this recording but dismisses it, claiming it was recorded using non-standard instrumentation - i.e. not the Stevenson Screen.

Most WA weather stations were equipped with Stevenson Screens in the early 1890s and the BoM's claim about Eucla in 1906 is at odds with decades of newspaper reports. For example, The Western Mail newspaper reported the following in 1934:

eucla record temperature with stevenson

Other examples include The Daily News in 1932:

Yesterday at Port Augusta people sweltered, and they were not surprised to hear that the mercury had risen to a record height for that city - it stood at 119.4 at the hottest part of the day. This high figure recalls other hot days that have been experienced in various parts of the Commonwealth. The palm - a cool, shady tree - must go to Bourke, generally recognised as the hottest town in Australia. On one blistering day in January, 1909, Bourke scored the Australian record of 125 degrees. The highest recording in our own State was made at Eucla, now known principally as a residence for snakes. In January, 1906, the mercury climbed to 123.2 degrees. Marble Bar has a couple of rather warm days to its credit. In January, 1905, and again in January, 1922, 120.5 degrees were recorded in that pleasant little town.

The Mercury Hobart in 1939:

The claim by New South Wales that Ivanhoe, with an unofficial reading of 122 degrees, is the hottest place in Australia, is challenged by West Australia, in fact West Australia challenges all-comers as far as Australia is concerned. The 27 consecutive "centuries" of temperature claimed by Ivanhoe would represent a cool change for Marble Bar (W.A.). Marble Bar has the distinction of 106 consecutive "century" readings in the shade, made with tested instruments under regulation conditions. Marble Bar has an official record of 120.55 in January, 1905. Eucla has gone even better than that. In January, 1906, 123.2 was reached.

The Melbourne Argus in 1922:

Tho highest temperature ever officially registered in Australia is 125deg. Fahr. This was at Bourke. Eucla came second in 1906 with 123.2, and in 1908 with 122.

The West Australian in 1944:

Temperatures at many places in this State recently have approached record figures. A Morawa correspondent writes that on Saturday the centre sweltered under a maximum of 118deg, which was a record for the district. According to Weather Bureau records, the highest temperature ever officially recorded in this State was the 123.2deg reported from Eucla on January 22, 1906.

The Daily News in 1949:

Marble Bar's 11-day century heatwave reached its peak yesterday when the temperature was 111deg. It was the town's hottest day since last summer. Marble Bar's record run of readings exceeding the century is 160 days - from October 31, 1923, to April 7, 1924. Its hottest day was January 11, 1905, when the mercury soared to 120.5deg. This record was broken next year when the temperature in Eucla hit 123.2deg. on January 22, 1906.

These newspaper reports suggest the Eucla record temperature of 123.2C (50.7C) was from a Stevenson Screen and BoM spokesmen identify the station as such in ensuing years. What makes WA's hot temperature record at Eucla in 1906 invalid?

Different sources

Evidence of an increasingly benign climate can be found from many sources. For example, in Trends and variability in storminess over south-east Australia since the end of the 19th century, BoM meteorologist Blair Trewin examined the incidence of extreme storms in Australia since records began:

"Using geostrophic wind speeds derived from eight triangles of sub-daily mean sea level pressure observations, we calculated measures of storminess for the south-east Australian region from the end of the 19th century through to 2008 for all seasons. Storminess has reduced in almost all triangles and seasons across the region since the end of the 19th century but particularly in autumn and winter. Reductions are statistically significant at the 5% level in nearly all regions and seasons. Using data from all the station triangles to create a combined index for south-east Australia, our results indicate statistically significant declines in all seasons in both storm and severe storm activity. There is strong decadal variability in storm activity particularly in the early part of the record, with storminess peaking in the 1920s. Decadal variability in storm activity has reduced in more recent decades in all seasons.

The results show strong evidence for a significant reduction in westerly flow across south-east Australia over the past century, consistent with a southward movement of Southern Hemisphere storm tracks. While these changes can not fully explain the significant reductions that have been observed in rainfall in the region it is likely that reductions in storminess are exacerbating observed drought conditions."

Alternative evidence of a reduction in extreme hot days can be sourced to the CSIRO which in 2010 published An Investigation of Extreme Heatwave Events and Their Effects on Building and Infrastructure which has heatwave records tabulated for 548 Australian weather stations and reaches the following conclusions:

Figure 3.1 presents the changes in the yearly number of hot days evaluated given four different temperature thresholds by the use of observation at the Melbourne Regional Office station. It can be seen that whilst the numbers of hot days equal or more than 35°C and 40°C slightly decrease, the numbers of hot days equal or more than 25°C and 30°C has increased over the years.

In other words, the CSIRO finds an increase in duration of heatwaves above 25C but a decrease above 35C, which is in line with the daily raw data at so many stations above suggesting a reduction in the diurnal range with less "extreme" cold nights and less "extreme" hot days.

For long record regional stations, the CSIRO finds:

Meanwhile, the number of hot days with Tmax ≥ 30°C (Table 3.1) shows increasing trends at about half of the stations, and decreasing trends at the other half.

The tables show an increase at 21 of Australia's long-record stations since records began, and a decrease at 24.

Top of page

Bookmark and Share

This website produced by Scribeworks 2009/13

Cape Leeuwin
Cape Naturaliste
Halls Creek
Marble Bar
Mt Barker
Rottnest Island
Southern Cross

Back To Top