Taking Catalyst's Temperature

Consecutive temperatures Sea levels Tasmanian fish

Vineyards Melbourne temperatures Declining snowfall

The "landmark" ABC Catalyst production Taking Australia's Temperature broadcast on 15 November 2012 was researched and produced in collaboration with the Australian Bureau of Meteorology, which created four pages on its website to explain information prepared for the ABC Catalyst program.

The program's misleading claims about 330 consecutive months of above average temperatures since February 1985 have been illustrated at the Ken's Kingdom (here and here) and Jo Nova websites.

These established that the 330 consecutive months with above average mean temperatures were global averages published by the National Climatic Data Centre in the US, and were not representative of the Australian temperature record despite the program being titled Taking Australia's Temperature.

On either 19 or 20 November, the ABC altered the wording on its program transcript page. From 15 November the transcript read "... since February 1985, we have had... 330 months in a row of above-average temperatures" (as in the archived video 13" 35'). The text was altered on 19 or 20 November to read "... since February 1985, we have had... 330 months in a row of above-average (global) temperatures", adding the word "global".

This is a partial correction but not an acknowledgement of the error, which was in the broadcast program, will be ignored through lack of publicity and which was believed by viewers. The archived Catalyst video continues to broadcast the error.

Above average temperature mistakes

According to Catalyst, Australia had above average temperatures for 330 months in a row up to November 2012.

According to reality, Australia had above average temperatures for three months in a row up to November 2012.

The BoM monthly update Australia in July 2012 shows the national mean maximum temperature anomaly (average 1961-1990) was +0.07C and the minimum anomaly was -0.96C, a mean anomaly of -0.45C.

According to Catalyst, there was a 1 in 100,000 chance of this happening to Australia's climate in the absence of human influence.

June 2012: max -0.27C / min -0.94C / mean -0.6C

May 2012: max +0.3C / min -1.6C / mean -0.7C

Apr 2012: max +0.3C / min -0.4C / mean -0.05C

Mar 2012: max -1.6C / min -0.8C / mean -1.2C

Feb 2012: max -0.3C / min -0.72C / mean -0.5C

Jan 2012: max -0.43C / min +0.06C / mean -0.2C

Averaged over the 12 months to November 2012 when Taking Australia's Temperature was broadcast, Australia's annual mean temperature anomaly was 0.12C below the 1961-90 average.

The odds ratio of 330 consecutive above average months was a core statistical demonstration in this "landmark" program about Australia's temperature history and was presumably researched and checked by the BoM, but it was wrong in an Australian context.

The BoM's Catalyst Temperature page and the Catalyst production blog both source program evidence to the CSIRO's State of the Climate 2012, which contains the following graphic:

Ken's Kingdom presents other examples of how Australian mean temperatures have fallen below their average since 1985, but how/why did both departments script global statistics when their own source data showed it was an incorrect description of Australia's temperature record?

"We" in the transcript of a program titled Taking Australia's Temperature obviously means Australia, and it is remarkable that these incorrect facts and/or wording were not vetted by ABC or BoM fact-checking. The error was either an oversight or a bias.

Blogs have highlighted other shocking but dubious warming examples presented in Taking Australia's Temperature such as the irrelevance of birds dying in heatwaves, as they always have - 1896, 1899, 1913, 1924, 1929.

Top of page

Catalyst converter

According to the Catalyst narration, "Melbourne hit 46.5 degrees" in February 2009.

In fact, Melbourne Regional Office 86071 recorded 46.4C on 7 February 2009, according to both the BoM raw and ACORN temperature databases:

How can the ABC's premier science unit fail to accurately describe the hottest day ever recorded in Melbourne on a landmark national climate program produced in collaboration with the BoM, which presumably checked the program for accuracy prior to broadcast?

February 2009 had Melbourne's hottest ever day of 46.4C but it's worth looking at the city's top 50 hottest temperatures since Melbourne Regional Office began recording in 1855. The raw temperature data to the right shows an increase in extreme Melbourne temperatures since 2000 but little more than was recorded in the 1800s: 1850-1859 - 2 1860-1869 - 7 1870-1879 - 4 1880-1889 - 1 1890-1899 - 3 1900-1909 - 7 1910-1919 - 0 1920-1929 - 1 1930-1939 - 4 1940-1949 - 1 1950-1959 - 4 1960-1969 - 1 1970-1979 - 0 1980-1989 - 4 1990-1999 - 2 2000-2012 - 9 Although Melbourne Regional Office had the best meteorological equipment in Victoria at the time, a non-Stevenson screen was used before 1908. However, the office also wasn't enveloped by buildings and bitumen before 1908. A photograph of the Stevenson Screen position for Melbourne Regional Office (courtesy Watts Up With That), strongly suggests urban heat influence on the city's temperature record. The 2009 Melbourne heat wave was intense and maxima records were set across Victoria on a single day, 7 February 2009. However, it should be noted that the monthly mean maximum temperature at Melbourne Regional Office in February 2009 was 28.1C and the hottest mean February maximum ever was 30.2C in 1898. The monthly mean maximum temperature in January 2009 was 28.6C and the hottest mean January maximum ever at Melbourne Regional Office was 31C in 1908.

Catalyst correctly reported a record 48.8C at Hopetoun on 7 February 2009, although it is relevant that Hopetoun Airport 77010 has only been in operation since 2004. The average annual maximum at Hopetoun Airport has also dropped 1.15C since 2009:

Top of page

Rising sea levels

Catalyst also presented examples of rising Australian sea levels including the tidal gauge dating back to 1897 at Fremantle, one of the most important continuous sea record stations in the southern hemisphere, based on the following source chart:

The chart is ominous and suggests a sharp increase in sea level at Fremantle over the past 20 years.

However, Catalyst has overlooked a relevant point about Fremantle's sea level readings ... they've been flat since 1999.

The Bureau of Meteorology's Tide Gauge Metadata and Observed Monthly Sea Levels and Statistics provides monthly mean sea levels at Fremantle station 62230 from 1897 to 2010.

At first glance the Fremantle mean tide levels seem to confirm the rise illustrated in the BoM/Catalyst sea level chart.

There are numerous missing data points before 1944 but if the trend-line is considered accurate, is this claimed as evidence that there was enough C02 in the atmosphere in the first half of the 20th century to cause a 79mm sea level rise at Fremantle over 47 years (.6183 from 1897-1907 to .6969 from 1944-1954)?

This was 1.7mm per year. What was melting and why was the Arctic ice cover so large with little reported ice loss from 1900 to 1950?

Simultaneous estimation of global present-day water transport and glacial isostatic adjustment published in 2010 by Nature Geoscience estimates mass ice losses from 2002 to 2008 in Greenland, Alaska/Yukon and West Antarctica combined were 269 gigatonnes per year. Conversions show 360 gigatonnes of ice must melt to raise sea levels by one millimetre, meaning the 269 megatonnes per annum that melted during the hottest decade on record contributed 0.75mm per annum to the annual sea level rise from 2002 to 2008. How did ice melted by C02 warming contribute to sea levels at Fremantle rising by 1.7mm per year in the first half of the 20th century?

After 1944, sea levels at Fremantle were stable for several decades. The averaged decadal mean sea levels at Fremantle were:

1944-54 - .6969
1955-64 - .7275
1965-74 - .7176
1975-84 - .7243
1985-94 - .7079

The sea level at Fremantle rose significantly from 1995-2004 - .7800.

2000-2010 was said to be the warmest decade on record so this should correspond with melt, thermal expansion and rising sea levels. The highest averaged annual mean tide at Fremantle was in 1999:

The tidal record can also be viewed as annual averages with a linear rather than polynomial trend line:

Fremantle sea levels are at a higher decadal plateau since the 1990s so critics may argue this is cherry picking. However, the trend has effectively been flat to falling for a decade, with a rise in 2008/09. Global sea level indices have been flat or fallen sharply since 2010.

There is no BoM gauge for Port Arthur, also highlighted by Catalyst, but the five Tasmanian BoM tidal gauges have a mixed record with two suggesting higher sea levels in the 60s and 70s.

The Fort Denison tidal gauge in Sydney shows sea levels similar to Fremantle up to about 1950, flat from 1950 to the 1990s and still not matching the 1999 peak:

The Fremantle and Sydney tide gauge trends have similar plateaus during the same decades as temperature anomalies. Although there is no evidence land temperatures and sea levels are directly related, both the Fremantle and Sydney sea levels show a corresponding flat trend since the late 1990s similar to temperature anomalies as reported by the National Climatic Data Center.

The NCDC global anomalies used as source data in Taking Australia's Temperature also show temperatures have been flat since 1998, an acknowledged climate trend that throws into question the underlying premise of the program:

These tidal gauge records mirror the findings of Is There Evidence Yet of Acceleration in Mean Sea Level Rise around Mainland Australia?, a 2011 paper published in the Journal of Coastal Research by principal coastal specialist Phil Watson from the NSW Department of Environment.

In July 2012, Ocean and Coastal Management published a study by Alberto Boretti from the Missouri University of Science and Technology showing Australian sea level records do not indicate a looming catastrophe.

Top of page

Something fishy

Taking Australia's Temperature reveals mysterious new fish species showing up in Tasmanian waters, including "brand spanking new to Tasmania, it was a yellow-tailed kingfish".

According to redmap.org.au: "Often caught in Northern Tasmania the kingfish is anecdotally considered to be getting more common in southern Tasmania in recent years". More information here and here.

Catalyst narration: It's exciting times for Tasmanian fishermen. With so many new fish arriving, they've teamed up with scientists to plot them. They've seen leather jacks, green turtles, dusky morwong..."

Tasmanian fishermen may or may not be excited to learn that almost all species of leatherjacks inhabit Tasmanian waters, as do green turtles and dusky morwong.

Is Tasmanian fishing being affected by sea surface temperatures? According to Taking Australia's Temperature, since 1900 "sea temperatures here off Tasmania have risen an astounding 2.28 degrees. That's about four times the global ocean average."

According to the Australian Institute of Marine Science in 2009, the annual sea surface temperature in the Tasman Sea rose 0.08C per decade from 1950 to 2008, or 0.48C.

Charts below from Australian Fisheries Climate Change Program 2012 published by the Victorian Department of Primary Industries show an increase in Tasmanian sea surface temperatures as much as 1.5C since 1900 (1.2C since 1880), not 2.28C as claimed by Catalyst.

On Australia's west coast, Taking Australia's Temperature investigated the record sea surface temperatures at Rottnest Island, finding "Last year, on 28th February, the water in here hit 26.4 degrees. 26.4 degrees?! That's ridiculous."

It may be ridiculous that this landmark temperature analysis consistently failed to accurately report temperatures. According to The "marine heat wave" off Western Australia during the summer of 2010/11 published by the Department of Fisheries, the peak temperature off Rottnest on 28 February 2011 was 26.3C.

The error was small and the sea temperatures were at record levels, but why was it inaccurate and, like Melbourne, why was it an upward error?

Top of page

Tipsy temperatures

In keeping with its attraction to wine, Taking Australia's Temperature also shocked viewers with the following point: "The wine industry's detailed records show grapes in Australia's south are ripening, on average, 20 days earlier than in 1985."

The 20 days is in reference to a CSIRO/Melbourne University study published in 2011, the CSIRO case study noting:

For the period 1993–2009, 43 of 44 vineyard blocks displayed trend to earlier maturity. For 35 of these sites the trend was statistically significant. Only one Margaret River vineyard in Western Australia ripened later.

In 10 of these 44 vineyard sites longer records of observations, periods of up to 64 years (41 years on average), had been kept.

For these sites, the study revealed an increase in the rate of advance of maturity over more recent decades.

Over the more recent 1993–2009 period, the average advance was 1.7 days year, whereas for the period 1985–2009 the rate of advance was 0.8 days yr_1 on average.

Determining the underlying causes of these shifts was the next step, especially with earlier maturing potentially impacting wine-grape quality and regional branding.

This study, published in Nature Climate Change, set about to test the assumption that earlier ripening of winegrapes was due to observed regional warming.

What was found was that at these 10 vineyard sites only about a third of the shift was driven by regional warming.

Other factors were also affecting the timing of winegrape ripening. On average, over the period 1985-2009, early ripening of Australian wine grapes are equally attributable to climate warming, declines in soil water content, and lower crop yields.

An additional influence from changing management practices is also likely.

The 2000-2009 drought presumably affected the historical records since 1985 and a third of the earlier maturing of grapes was due to regional warming. In other words, the grapes are ripening 6.6 days earlier than in 1985 as a result of higher temperatures, not 20 days, and even this is probably an overstatement.

Newspaper reports suggest north-east Victorian grape-growers will enjoy a bountiful season in 2012/13 with the industry looking forward to a return of drier (and hotter) El Nino conditions.

Taking Australia's Temperature has accurately stated the CSIRO report finding that grapes matured 20 days earlier, but has strongly implied rising temperatures are the total reason by failing to report the underlying causes of the early ripening, as detailed in their source material.

Top of page

A snow job

Taking Australia's Temperature narration: "In 60 years, we've lost a third of our total snow cover."

Viewer interpretation: "In 60 years, Australia has lost a third of its total snow cover."

Catalyst source document The impact of climate change on snow conditions in mainland Australia: "Over the past 50 years, there is evidence of small increases in New South Wales alpine precipitation and small decreases in Victorian alpine precipitation. A small decline in maximum snow depths is evident at three of the four alpine sites in the years between 1957 and 2002."

Taking Australia's Temperature narration: "So, basically, since 1954, snow depth in July is much the same. When you reach September, it starts to drop off. So that by October it's noticeably less. Essentially, spring is coming earlier."

Viewer interpretation: "There's less snow in October so spring is coming earlier and temperatures are rising."

Catalyst source document The impact of climate change on snow conditions in mainland Australia: "A moderate decline in August and September snow depths is evident at three sites, possibly indicative of the tendency for mid to late season snow depth to be determined by temperature-dependent ablation (melt and evaporation), whereas the depth of early season snow is determined by precipitation."

The CSIRO study sourced by Catalyst was published in 2003 and does not take into account heavy snowfall seasons since then including 2012.

Taking Australia's Temperature contains numerous other misleading examples, such as 2011 Port Arthur sea flooding caused by coincidental strong winds and swell but blamed on sea levels, an unsubstantiated claim that Australia has nine times the number of flooding events for structures at sea level compared to 100 years ago, and the dry Jewel Caves in WA's south blamed on low rainfall but ignoring studies (PDF 17mb) suggesting land clearing and less bush fires have increased forest understorey growth and absorbant ground litter.

Closer inspection of the warming examples presented by Catalyst in Taking Australia's Temperature indicates misleading and poorly researched evidence to support the contention that Australia's climate is warming due to CO2.

This website produced by Scribeworks 2008-2020