Dodgy data from the Bureau of Meteorology

Several chapters of Climate Science: The Facts raise really serious issues about the temperature records maintained by the Bureau of Meteorology (the Bureau). In addition to the adjustment (homogenisation) of records and the trashing of old files investigators met bureaucratic obstruction to keep the management processes out of sight.

One of Tony Abbott’s last moves as PM was to set in train an investigation into the operations of the Bureau but that initiative was aborted by the incoming PM Turnbull. This is understandable given the importance of fake science and fake news to sustain the push to destroy the electrical grid of the nation.

Joanne Nova described mysterious revisions to Australia’s long hot history that were located by a team of well qualified and experienced volunteer citizen scientists. The story really has to be read to get the full flavour. For example they found that thermometers accurate to a tenth of a degree were being adjusted by as much as two degrees. The records now indicate that the hottest day recorded in modern history was at Albany on the coast of WA rather than in the baked arid zones of Oodnadatta or Marble Bar. The temperature reported in Albany on 8 February 1933 was 44C and 8 decades this was adjusted by 7C to 51. This pipped the 50.7 that was recorded for Oodnadatta on 2 January 1960.

New electronic thermometers were widely installed in the 1990s and many ran alongside the old system which would permit comparison of the old and new systems but the data are not available to the public and are routinely deleted as a part of normal practice. It is likely that the new system generates a “step up” in recorded temperatures but it appears that the trashing of records has eliminated the possibility of a proper investigation.

Early explorers were trained to record temperatures and there are masses of newspaper records that indicate weather patterns around the nation with many accounts of temperatures around 125F “in the shade” translating into 52C. On 3 January 1909 an observer recorded 125F at Bourke. The handwritten entry was underlined as one might do for a remarkable reading. Decades later this was declared an error for two reasons: it was taken on Sunday when the observers didn’t normally work and the temperature at some other towns in midwestern NSW only hit 113F. On the first point, why would an observer NOT make an effort to record a potentially record-breaking temperature? On the second, a newspapers reported a figure of 123 at Brewarinna which is the nearest town to Bourke.

This team joined forces with Senator Bernardi to ask for an audit of the historical records from the Australian National Audit Office in 2011 but the Bureau announced a major revision of the data called the Australian Climate Observation Reference Network – Surface Air Temperature (ACORN-SAT) which sidestepped the audit because the request referred to the old data set, now unused and possibly deleted.

Jennifer Marohasy described the homogenisation of Rutherglen, a town in the Victorian wine country a little west of Albury. The temperatures were recorded at the agricultural research station since 1912 in a Stevenson screen, a wooden structure designed to standardise the immediate environment of the equipment. Very strange things happened in the course of homogenising the data for the new ACORN-SAT system. A cooling trend of 0.3C at the station is converted into a statistically significant warming of 1.6C per century, ignoring trends at nearby stations. The story is complex involving tables of data and helpful charts so I will not try to tell more of it here – just beg, borrow or steal the book if you are too poor or mean to buy it.

When Greg Hunt announced that the investigation of the Bureau would not proceed under the new management he explained to Tony Jones on Lateline that the Bureau used hard science, hard data, literally millions of data points of information through our satellite and local monitoring. Marohasy wrote that ACORN-SAT relies entirely on surface data from 112 stations, 109 of which are homogenised. Satellite readings are not involved which means that the acronym SAT is as deceptive as the information provided.

Marohasy and Dr Jaco Vlok wrote Chapter 10 “Moving in Unison: Maximum Temperatures from Victoria, Australia. Dr Vlok is apparently a refugee from South Africa, now engaged in research at the University of Tasmania (my alma mater) on building historical temperature records using artificial neural networks (don’t ask me).

The work in this chapter does not call for sophisticated mechanical or electronic analysis, merely the technique known as “eyeballing” in the technical language of the trade, using the neural network connected to your eyes. This involves more charts and I strongly suggest that you eyeball them. They simply show all the raw data for maximum temperature series in Victoria that are available, starting with some in 1860. To summarize, all the series move together suggesting that they are all reflecting the same reality, the same variations and changes over the many decades. “But there is no long-term trend. There are, however, cycles of warming and cooling with the warmest periods corresponding with times of drought.

This entry was posted in Global warming and climate change policy, Rafe. Bookmark the permalink.

25 Responses to Dodgy data from the Bureau of Meteorology

  1. H B Bear

    51 degrees in Albany? LOL Albany is like a little piece of Victoriastan that has drifted across to WA. Always jumper weather in February. Not a bad golf course though if you stay on the track.

  2. JMH

    Thank you, Rafe.

    Although not in the same class as Jennifer Marohasy et al, I have collected six months worth of daily minimum temperatures for one town from “the Bureau’s” website. This has been a fascinating exercise noting “disappearing” data some days and in one case six months after it was originally recorded. I can prove BoM recorded the data on the days it later mysteriously went missing. My intention is to deal with my information accordingly. There is something seriously smelly over at the BoM.

  3. stackja

    BOM it’s all Greek to me.

    mid 16th century (denoting any atmospheric phenomenon): from modern Latin meteorum, from Greek meteōron, neuter (used as a noun) of meteōros ‘lofty’.

  4. Genghis

    What about the outrageous manipulation of Arizona (USA) data for 1934? This was a well known really hot year.. This was adjusting temperature figures down by up to several degrees so later in the century the rise in temperatures are more dramatic.
    And they call themselves Scientists!

  5. Chris

    51 degrees in Albany? LOL Albany is like a little piece of Victoriastan that has drifted across to WA. Always jumper weather in February. Not a bad golf course though if you stay on the track.

    Albany at 51? U-huh.
    I remember a STINKING day it got to 112 deg F at the airport. That’s 44 deg C.

  6. Up The Workers!

    The retrospective “tinkering” with the records only seems odd, if you still think of the B.o.M. under its old name of: “Bureau of Meteorology”.

    Just think of the B.o.M. under its new name: “Bureau of Mendacity”, and it all makes perfect leftard sense.

  7. Bruce of Newcastle

    It’s very easy to see that such adjustments are crap.

    Just compare the GISStemp data with the snow cover record.

    Snow cover trend has been steady since about 1995. But in GISStemp, which most politicians like to point to, the temperature has supposedly gone up lots in the same period.

    Now unless that is all hot snow I suspect the GISStemp data is inferior to a patent medicine salesman’s pitch in accuracy.

    The satellite temperature data, however, fits the snow cover data pretty well. But that dataset is run by climate sceptics so you never see it in the MSM.

  8. Bemused

    They’re not even trying any more in Margaret River WA. A presumable fault has meant no minimum temperatures and no rainfall data since late May although they have conveniently rounded up maximum temperatures to the nearest whole number and quell surprise we are now above average temps for the month.

  9. JohnL

    There is something seriously smelly over at the BoM.

    Why is BoM bot being dismantled right at his moment?!?!
    What is the reason for having BoM?
    How many accurate weather forecasts there were from BoM (in comparison to wrong forecasts)?

  10. Tom

    There is something seriously smelly over at the BoM.

    It’s called corruption: I’ll doctor the data as long as the grants for doctored data keep flowing.

    And it’s always public money misappropriated in the government sector, where there are no ombudsmen or quality controllers working for taxpayers.

    It can’t be done if your pay cheque depends on untainted research.

    CAGW is the biggest scam in world history now worth around $US2 trillion a year based on the insurance industry’s 2015 estimate — bigger than the Australian economy.

  11. Dr Fred Lenin

    A day at the BOM, junior “scientist”to Boss “,these figures look nice and low Sir,” Boss “Low? low? Give me that paper ,where’s the whiteout?” “Here you are Boss” ,a minute later the Boss says “ that’s better ,lot higher now,who says there is no global warming eh ? Nothing whiteout and upward cirrection won’t fix

  12. Herodotus

    The satellite temperature data, however, fits the snow cover data pretty well. But that dataset is run by climate sceptics so you never see it in the MSM.

    And still there are some who say media doesn’t matter. Media is the core of the problem.

  13. Dr Faustus

    New electronic thermometers were widely installed in the 1990s and many ran alongside the old system which would permit comparison of the old and new systems but the data are not available to the public and are routinely deleted as a part of normal practice.

    Mercury-in-glass thermometers are accurate and repeatable and have been so since the mid 1800’s.

    In my industry, they have been replaced because electronic thermometers are easily integrated with digital data capture, are (usually) more robust, easier to read under field conditions – and, most importantly from a HSE perspective, don’t contain debbil-toxic mercury. But not because they are more ‘accurate’, or better calibrated (actually, electronic thermometers tend to lose calibration).

    There is likely to be a small systematic difference between new and old systems (perhaps +/-0.1C). But there should be no need to abandon, or make gross adjustments to careful readings taken back in the late 19th Century simply because they appear to be ‘low tech’. In fact, unless the comparison digital readings are very carefully calibrated it would lead to error.

  14. Os

    H B Bear, Albany was settled before Victoristan. And before Swan River as well, actually.

  15. Iggie

    This is a comparison of Bourke’s raw temp compared to the ‘adjusted’ ACORN temp for Jan 1939.
    Every temp above 30C has been adjusted down (by up to 1C) and the two below 30C have been adjusted up (by 0.1C).
    Bourke 1939
    Jan raw Jan ACORN
    1st 38.9 38.4
    2nd 40 39.1
    3rd 42.2 41.9
    4th 38.1 37.9
    5th 38.9 38.4
    6th 41.7 41.5
    7th 41.7 41.5
    8th 43.4 43
    9th 46.1 45.7
    10th 48.3 47.9
    11th 47.2 46.8
    12th 46.2 45.8
    13th 45.7 45.3
    14th 46.1 45.7
    15th 47.2 46.8
    16th 46.7 46.3
    17th 40 39.1
    18th 40.1 39.1
    19th 40 39.1
    20th 41.9 41.7
    21st 42.5 42.1
    22nd 44.2 43.8
    23rd 36.7 36.5
    24th 40.3 39.2
    25th 36.6 36.5
    26th 29.4 29.5
    27th 29.3 29.4
    28th 28.8 28.9
    29th 30.6 30.5
    30th 35.6 35.4
    31st 38.6 38.3

    Does that mean the SS was moved everyday? This is straight out adjustment for no possible reason.
    The BoM needs to come clean on, not just this one, but many other adjustments.

  16. This is straight out adjustment for no possible good reason.

  17. Bruce of Newcastle

    Just having a read of before dinner. The sheer number of climate crap papers is amazing.

    Here’s a list of headlines today.

    Study suggests buried Internet infrastructure at risk as sea levels rise
    Electric car batteries souped-up with fluorinated electrolytes for longer-range driving
    Thawing permafrost microbiomes fuel climate change
    Behavior-influencing policies are critical for mass market success of low carbon vehicles
    Reducing carbon emissions will limit sea level rise, study says
    Protecting tropical forest carbon stocks may not prevent large-scale species loss
    Flipping the switch: Making use of carbon price dollars for health and education
    Getting to know the microbes that drive climate change
    Poll shows consensus for climate policy remains strong
    A scientist’s final paper looks toward Earth’s future climate

    Another half dozen have a tangential link. But these take the cake: bullshit on the taxpayers’ dime.

  18. egg_

    Dodgy data from the Dodgy Brothers gummint?
    Strike me pink!

  19. Entrop

    One of the problems at BoM is that management is very keen on provision of services and modelling, not so keen on just being a data source. No money in that of course.

    A better value for taxpayers is for BoM to focus on recording keeping, and-this is very important-expanding rather than the current practice of shutting down physical observation points. This is the reverse of what they are doing to the extent WA and NSW governments have actually committed millions of dollars for new met stations and weather radar. Which, I would have thought, was the responsibility of the Australian Government and BoM.

  20. Entrop

    Electric car batteries souped-up with fluorinated electrolytes for longer-range driving

    What could go wrong?

  21. davefromweewaa

    Tried and failed to post earlier.
    Several funny things have happened to records for Bourke.
    Nothing before 1910 counts for the BOM’s “official” record.
    There has not been a hotter January since 1896 [when atmospheric CO2 was under 300 ppm.]
    Raw data for January 1939 had 17 days in a row of 40 degrees plus until it was “homogenised.”
    Now it is 11 days. Snap, Iggy @ 5:04.
    Lying liars who lie!

  22. Iggie

    As you say, 1896 was the hottest recorded Jan in Bourke and had 22 days of +40C. I’m fairly sure that Bourke had a SS then at the PO so temps would be reasonably accurate. From Trove:-
    ‘BOURKE, Jan 21.— Two deaths are reported from heat apoplexy, bringing the record to 35. The average heat for the month to date has been 112 in the shade, and for the past four days 118. A large number of people left Bourke by train this morning, and cheap excursion trains have been specially provided, and will be largely patronised to-morrow. The residents are panic-stricken, and hundreds are leaving for
    a cooler climate.’
    The average for the month was 43.4C. Nothing has come close in the past 20 years (40.6C in 2013).

  23. Entrop

    You can still find pre 1910 data on the internet. The Qld government provides it. Look up longpaddock and SILO. If you know how to get it, you can even pick out the unhomogenised point source data.

  24. Iggie

    I see that the current SILO system will be decommissioned at the end of this year to be replaced by a new SILO system. I notice that some data sets are sourced from the BoM.
    It will be interesting if the old and new systems match the historical record.
    When BoM replaced its High Quality Data set with ACORN, there were many changes.
    (Of course, HQ data had already been adjusted from the original raw data.)
    This BoM’s graph shows the adjustments from AWAP to ACORN. Note all the downward adjustments prior to 1965 and upward adjustments after 1990.

Comments are closed.