climate change wa

RAW temperatures disagree with ACORN anomalies


Bureau of Meteorology temperature anomalies in ACORN 2 show Australia's average annual maximum in 2019 was 0.54C warmer than 2018 even though maxima observations at the ACORN stations rose 0.27C from 2018 to 2019, and the average annual ACORN minimum in 2019 was 0.22C warmer than 2018 even though minima at the ACORN stations cooled 0.03C from 2018 to 2019.

Such incongruities between absolute temperatures and ACORN anomalies are not uncommon, but are worth questioning in light of the BoM claim that its anomaly adjustments show 2019 to have the hottest annual average maximum recorded in Australia since 1910.

The ACORN maximum anomalies were 1.55C in 2018 and 2.09C in 2019, suggesting averages increased 0.54C. ACORN minimum anomalies were 0.73C in 2018 and 0.95C in 2019, suggesting averages increased 0.22C.

The comparison: RAW temperatures v ACORN anomalies

The Australian Climate Observation Reference Network (ACORN) began in 2011 and its annual average temperature estimates were recalculated in early 2019 to create what is known as ACORN 2. ACORN is the national station network used by the BoM to average temperatures and inform the public what was the temperature anomaly each year (compared to the 1961-90 average).

There are 112 ACORN weather stations across Australia but only 104 are used by the BoM to calculate national averages as eight stations are deemed urban and affected by artificial warmth, and are thus unsuitable for inclusion (Adelaide, Hobart, Laverton, Melbourne, Richmond NSW, Rockhampton, Sydney and Townsville).

Among these 104 stations, the most recent start year was Learmonth Airport in 1976, meaning that earlier averages are of smaller station numbers back to the 57 long-term stations that opened in 1910. Thus, the following tables begin in 1976 so that station numbers are uniform throughout.

Below are tables of averaged original RAW annual maxima and minima at the 104 stations compared to adjusted ACORN 2 temperatures, along with anomalies for each based on their 1961-90 averages as well as the BoM's annual average rainfall anomalies as a guide to environmental conditions affecting temperatures each year (note - it's not necessary to read or interpret the many years of data tabulated, unless interested for research purposes, as there are summaries at the end of each table :

104 station maxima 1976-2019

australian maximum temperatures anomalies for 104 stations

* RAW anomalies based on 1961-90 average temperature at the 104 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

Observations in both the RAW and ACORN datasets suggest 2019 had the warmest maxima on record across Australia, likely contributors being:

  • The lowest national rainfall since 1910.
  • Exacerbating the low rainfall was delayed monsoonal rain arriving in northern Australia, coupled with a sudden stratospheric warming above Antarctica (strongest since 2002), inducing a negative phase of the Southern Annular Mode which suppressed rainfall and increased maxima in eastern Australia during October, November and December.
  • A very strong and long-lived Indian Ocean Dipole starting positive in May 2019, the IOD index peaking in spring at the highest weekly value observed since records began in 2001.
  • The influence of small screen automatic weather stations with one second observations which record extreme temperatures that could not be detected by liquid-in-glass thermometers with slower response times prior to the 1990s.

104 station minima 1976-2019

australian minimum temperatures anomalies for 104 stations

* RAW anomalies based on 1961-90 average temperature at the 104 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

The BoM states that out of 110 years from 1910 to 2019, Australia's 2019 national minimum was ranked at 105 for averaged warmth. However, the RAW data above shows 2019 ranked 14th warmest since 1976 (from warmest to coolest, 2016, 2013, 1988, 2014, 2005, 1983, 1998, 2015, 2007, 2017, 2010, 2009 and 2018 had warmer minima).

104 station mean temperature 1976-2019

australian mean temperatures anomalies for 104 stations

* RAW anomalies based on 1961-90 average temperature at the 104 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

However, it should be noted that four ACORN stations had no observations in different months during 2019, creating inaccurate annual averages for that year (Boulia max and min July, August, September / Inverell min June, July, August / Tibooburra min November and December / Wittenoom as of September 2019 when the station was closed).

Below are tables of averaged original RAW annual maxima and minima at the 100 stations when these inaccurate stations are removed, compared to adjusted ACORN 2 temperatures, along with anomalies for each based on their 1961-90 averages as well as the BoM's annual average rainfall anomalies as a guide to environmental conditions affecting temperatures each year :

100 station maxima 1976-2019

australian maximum temperatures anomalies for 100 stations

* RAW anomalies based on 1961-90 average temperature at the 100 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

100 station minima 1976-2019

australian minimum temperatures anomalies for 100 stations

* RAW anomalies based on 1961-90 average temperature at the 100 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

100 station mean temperature 1976-2019

australian mean temperatures anomalies for 100 stations

* RAW anomalies based on 1961-90 average temperature at the 100 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

Below are tables of averaged original RAW annual maxima and minima at all 112 stations including the urban and inaccurate stations, compared to adjusted ACORN 2 temperatures, along with anomalies for each based on their 1961-90 averages as well as the BoM's annual average rainfall anomalies as a guide to environmental conditions affecting temperatures each year :

112 station maxima 1976-2019

australian maximum temperatures anomalies for 112 stations

* RAW anomalies based on 1961-90 average temperature at the 112 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

112 station minima 1976-2019

australian minimum temperatures anomalies for 112 stations

* RAW anomalies based on 1961-90 average temperature at the 112 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

112 station mean temperature 1976-2019

australian mean temperatures anomalies for 112 stations

* RAW anomalies based on 1961-90 average temperature at the 112 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

The ACORN maximum anomaly at all Australian stations increased 0.54C from 2018 to 2019 but at the 54 ACORN stations with a start year of 1910 (excluding Boulia, Inverell and Tibooburra) the absolute RAW maximum increased 0.32C from 2018 to 2019.

The ACORN minimum anomaly at all Australian stations increased 0.22C from 2018 to 2019 but at the 54 ACORN stations with a start year of 1910 (excluding Boulia, Inverell and Tibooburra) the absolute RAW minimum increased 0.01C from 2018 to 2019.

Below are tables of averaged original RAW annual maxima and minima including the urban locations, compared to adjusted ACORN 2 temperatures, along with anomalies for each based on their 1961-90 averages as well as the BoM's annual average rainfall anomalies as a guide to environmental conditions affecting temperatures each year :

54 long-term station maxima 1910-2019

australian maximum temperatures anomalies for 54 stations

* RAW anomalies based on 1961-90 average temperature at the 54 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

54 long-term station minima 1910-2019

australian minimum temperatures anomalies for 54 stations

* RAW anomalies based on 1961-90 average temperature at the 54 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

54 long-term station mean temperature 1910-2019

australian mean temperatures anomalies for 54 stations

* RAW anomalies based on 1961-90 average temperature at the 54 stations - this average is not the same as the adjusted ACORN 1961-90 average but can be considered a static comparator

Annual absolute temperatures and ACORN anomalies in the tables above are in disagreement for all years before 2018. The differences are not unique to 2018 and 2019, except that 2019 is considered Australia's warmest year on record.

For example and as per the tables above, the actual averaged RAW minimum at the 104 non-urban ACORN stations was 13.92C in 2014 (13.91C in ACORN) and 13.80C in 2015 - a 0.12C cooling - yet the ACORN minimum anomaly was 0.78C in 2018 and 0.81C in 2019, a 0.03C warming. Alternatively and further back in time, the actual averaged RAW maximum at the 104 non-urban ACORN stations was 25.40C in 1985 (25.22C in ACORN) and 25.06C in 1986 (24.79C in ACORN) - a 0.34C cooling in RAW or 0.43C cooling in ACORN - yet the ACORN minimum anomaly was 0.30C in 1985 and 0.17C in 1986, a 0.13C cooling.

Alternatively, the ACORN average annual maximum at the 104 non-urban stations was 25.16C in both 2000 and 2011 (25.20C and 25.18C respectively in RAW), yet the ACORN anomalies were -033C in 2000 and -0.10C in 2011.

Why don't the anomalies, which are deviations from a static 1961-90 average at the 104 stations, agree with annual differences to the actual temperatures observed?

The BoM is likely to attribute these many differences to what is known as area averaging, wherein ACORN adjusts each station's annual temperature through reference to neighbouring station temperatures in corresponding months and/or years - as well as changes to station equipment and locations.

However, the number of stations, their locations, equipment and neighbouring stations were unchanged from 2018 to 2019, raising questions as to why the ACORN anomalies (based on an unchanging reference to 1961-90 temperatures) are warmer than actual temperature changes in minima and maxima from 2018 to 2019.

Note: All tables presented above can be downloaded here (Excel 571kb).




Top of page
Website home


Bookmark and Share


This website produced by Scribeworks 2009-2020

Home