Thursday, January 23, 2014

Heat Wave in Victoria

Update - I've extended the table of temperature extremes below
Tennis fans everywhere will know that we had a pretty torrid heat wave last week, the first week of the Australian Open. It wasn't our hottest, but it went for four days. In Melbourne max temperatures were were 42.8,41.7, 43.9 and 43.9°C, or, if you prefer,109, 107, 111, 111°F. This forecast from the BoM explains what was happening; it happened as predicted, a bit warmer at the end. MikeH in comments noted this more complete report from B0M.

There's the usual argument about AGW. WUWT doesn't think so. In fact, they won't much concede that it was hot, and say it happens all the time here. So I thought I'd review in this post our summer temperature history, and also refute some notions that it's all happened before but been hushed up by "adjustments".

Monkey Mia, WA, From ABC


The Bureau of Meteorology here has online daily data which is unadjusted. It's convenient to use, but not to download. For that I used GHCN Daily files which can be found here. They are unadjusted historic temperatures, derived from (and apparently identical to) BoM. Melbourne’s file is ASN00086071.dly; it has good coverage from June 1855 to Dec 2013. I supplemented it with current online data to date.. I extracted them to a CSV file which I've placed here.


I've plotted each summer (DJF) max temperatures from 1856 to 2014 (counting by the JF year).



You can switch from year to year using the top buttons, and go to any decade year with the radio buttons.

Our summers are made up of many pleasant days, with southerly winds, with heat when the wind blows from the North. Notably hot summers were in 1908, 1939, 1983, 2009. The last three had devastating wildfires on the hottest days, which I've linked. Big heat has bad consequences here. We lose towns, homes and people.

Table of extreme temperatures

Originally, I had here a list of the twenty hottest days in Melbourne since 1855. Following a suggestion in comments by SC M, I have updated to inclide tables of the 20 highest daily max, highest min, lowest max and lowest min. I've marked those occurring in the last five years in red. For highest max, there are seven. Click the buttons.

High Max  Low Max  High Min  Low Min


In the WUWT thread I encountered over and over claims that BoM had rigged the records by adjusting old readings downwards. It's very hard to get specifics (typically the complainant just gives a link, which is often to a rant about GHCN, or GISS). Eventually it crystallised to complaints about Acorn. This is a recent homogenised set of data for 112 chosen stations, starting in 1910. The lead post in fact claimed that the BoM had discarded all data before 1910, which puzzled me.

But of course the data is not discarded. All the old unadjusted data is available from BoM, and also from GHCN. And it is what the BoM uses when talking about station extreme temperatures. If you go to the Melbourne climate page, for example, it will tell you that the hottest December day was 15 December 1876. That's not Acorn.

So how do we know these are really unadjusted? I've known since boyhood that Melb reached 114.1°F on Jan 13, 1939. That's what the record still says. But I checked others. The National Library has an old Melb (and others) newspapers here. I've checked a few of the reportewd high temps; the GHCN record linked here agrees with the contemporary reports. I think it's up to claimants of adjustment to give one example from this dataset.


  1. The red style is not showing on six of the table entries, Nick.

  2. "the seven occurring in the last **20** years"

    1. Sorry Nick. I initially completely missed your point and thought that it was a typo. So 7 of the top 20 records occurred in the last five years and 8 records in the last 20.

      What would that look like with the ACORN data? Is that available for that station?

  3. From the BOM "Special Climate Statement 48 – one of southeast Australia’s most significant heatwaves"

    "The heatwave was more notable **for persistent heat than for individual extreme hot days**, but some locations still had their hottest day on record, particularly in the southeast of South Australia, and around and to the west of the Snowy Mountains in New South Wales."

    "The daily mean temperature of 35.45°C on the 16th was Melbourne’s highest on record, just surpassing the 35.4°C observed on 30 January 2009. All four instances of daily mean temperature of 35°C or above in Melbourne have occurred since 2009, two in 2009 and two in 2014. "

  4. I guess if it's especially hot in a region of the earth, the alarmists are going to blame it on AGW. That's typical of them.


    1. I guess if it's especially hot in a region of the earth, the scientists are going to want to know why. That's typical of them.

    2. I was just kidding. I was in Australia in the late 1990's. It's beautiful from end to end.


  5. Australia makes up 1% of the globe doesn't it?

    Is that why its having no impact on temperatures and the pause has now lasted 16 years?

  6. Nike should know that he is talking about temperature extremes when he is discussing heat waves.
    There cause and exceptional nature are not necessarily driven by the same physical phenomenon
    as those that drive the mean summer time temperatures.

    This recent heat wave was exceptional compared to past events, however, the factors
    that produce these terrible events are dependent on a chance combination of phenomenon
    that is relatively rare.

    Extended heat waves in the South and South-Eastern parts of Australia require a number
    of things to occur simultaneously.

    [Note: the following is only a suggested sequence of events and it is not to be
    treated as a definitive explanation of what happens in any given heat-wave.]

    1. The heat wave is usually proceeded by a build up of heat in Australia' central deserts.
    This could be bought about by an extended dry period in central Australia - usually leading
    to a slow build of heat. Alternatively, the heat could quickly build up in the southern deserts
    because of a powerful rain event(*) along the inter-tropical convergence zone that lingers along
    Australia's north and western coasts during the Southern summer. The rising air from this
    tropical rain event moves high up into the upper troposphere before moving towards the south
    and east. It then descends onto the Southern and South Eastern parts of Australia producing
    a building high pressure dome of extremely dry and hot compressing air that provides the
    heat for the coming heat-wave.

    (*) Not this could be due a rapid onset the Monsoon or even a large than normal
    Madden-Julian Oscillation event in far-northern Australia.

    2. A large blocking high in the South Tasman. This is usually formed by the reinforcement of the
    semi-permanent high that is located in the southern Tasman Sea. During the Summer months
    High pressure systems move from west to east at latitudes that can (on occasion) reach as
    far south as Southern Tasmania.

    3. The blocking high in the Southern Tasman determines the way in which the
    building heat in Australia's southern deserts is vented off the Australia continent
    and into the surrounding oceans. With the blocking high firmly ensconced in the
    Southern Tasman, this means that building heat in the central and southern deserts
    is pushed south over the states of South Australia and Victoria [which contain
    the large cities of Adelaide and Melbourne].

    4. And last but not least, the presence of a sharp cold front that extends diagonally
    from the about + 25 - 30 degrees south on the Western Australia coast all the
    way down deep into the Southern Ocean below South Australia. This cold front
    is slowly moving towards the east. As it moves, it acts like a large diagonal
    brick wall of cold dense air, that diverts the southerly moving high temperature
    air coming off the deserts, funneling the super-heated air towards the southern
    state capitals of Adelaide and Melbourne.

    5. Of course, the whole shebang is helped if the summer time maximum's
    in Southern and South-Eastern Australia are higher than normal.

    [Note: You do not need all of these ingredients to get a heat wave but at
    least a couple of these phenomenon have to occur to get the whole
    show on the road.]

    1. "Nike should know that he is talking about temperature extremes when he is discussing heat waves. There cause and exceptional nature are not necessarily driven by the same physical phenomenon as those that drive the mean summer time temperatures."

      Yes, heat waves are not climate. It seems we've always had them. But there is some indication of a pattern that the very hot days are getting hotter. That's why I drew attention to the 7/20 hottest days in the last five years. It's not proof, and more evidence is needed to affirm the pattern.

    2. "Our recent research in the internationally peer-reviewed Journal of Climate shows that there has been a significant increase in the number of heat wave days for most of the country from 1951-2008. "


    3. Anonymous (Mike H), You cannot be serious, can you? The selection of dates "......from 1951-2008......" covers a period from the bottom of a cooling period [1945 - 1975] to the peak of a warming period [1975 - 2005]. This is cherry picking of dates at its finest. Why have you left out the earlier dates? Could it be because they do not support your case?

    4. I see you have immediately gone for the conspiracy theory rebuttal of the Perkins & Lewis paper on heatwaves.

      You may find this article useful.

    5. Ninderthana would like to think than the period 1945 - 1975 covers a cooling period.

      Not so in Melbourne:

      Facts are boring.

  7. Sounds like a reasonable idea Nick - However, all it proves is that temperatures have
    been going up over the last few hundred 100 years and no one disputes this. In addition,
    it says nothing about what is causing that increase.

  8. I would be interested to see a similar 20 hottest list for overnight minimum temperatures. I recall somewhere that minima are more strongly affected by AGW than maxima - though data for one location can't hope to be definitive of course.

    1. SCM,
      Yes. I can do that. My current project is preparing similar graphics for a number of cities and sites around the world. The graphic would show daily max and min. I can collect top 20 hot max, cold max, hot min, cold min.

  9. FWIW four consecutive daily maxima over 40 is without precedent and one year 2013 1014 has that. Very few decades have four days over 40 daily maxima, and none consecutively. This year WAS a record breaker.

  10. I've done something similar for Adelaide
    The older temperatures were in park land site that was closed in 1979. The modern site has setting sun reflecting on the Stevenson Screen in a built up location.

    The old Melbourne observations were taken in the botanic gardens (shifted prior to the shrine of remembrance being built).

    I came up with a way of comparing heat waves more easily. Rather than adding up the degrees over a certain temperature, I added together the squared of the degrees above 32°C (F). You might not agree with it but it weights the hot days as 22 days of 35°C is the same as 2 days at 44°C and one day at 46°C.

    Its not up to date but here are the the plots for Adelaide
    and Melbourne. It has not been updated in the past two weeks.

    1. Thanks for that. Yes, the Melbourne site shifted in 1908 from the Observatory to the BoM Office. This was initially in Drummond St (nearby) and then Latrobe St. Details here.

      Obviously there is some arbitrariness in quantifying heat waves - your formula sounds reasonable. The last few days here will have bumped 2014 up substantially.

  11. Curiously enough, UAH is showing a drop from +0.260C to +0.196C for the SH for January 2014.

    I suspect the phenomenon of Australia warming faster that the rest of the globe has something to do with it's unique geography.

    For comparison, the US had an extremely record cold month, by the NH showed a moderate increase from +0.266C to +0.291C.

    1. (strike the word "record" in "US had an extremely record cold month", I don't think it's any record.)

    2. Carrick,
      There's a more detailed (spherical) picture here. Yes, in Australia the heat was mainly in the SE (plus W NSW and Qld). There were complaints about heat in SE Brazil and Argentina, which show up on the map.

      As for US, the East of course was very cold. In anomaly terms, Alaska was very warm, and the West quite warm too.

      Elsewhere, N Siberia was cold; Central Asia and Europe quite warm.

      As for Australia's geography, a lot of our warmth in the last 18 months seems to be due to warm surrounding seas. I believe this heat wave was due to a puff of warm air from the Indian Ocean NW of WA. The phenomenon may be related to recent claims that trade winds in the Pacific have strengthened, causing a cool E Pafific and global "pause". That would blow the heat our way.