The fires were severe on that Saturday because of a confluence of events in the course of a day. Yes, it was correct that the dangers were forecast several days before. But they would not have been forecast to the day, or even at all, on a monthly, annual, decadal or century timescale.

Most climate models attempt to do the latter, to some degree. I agree with David that there is little skill demonstrated in the DECR. My opinion is that it fails because we are dealing with events that are on a “sub-grid” scale if I could borrow a term from spatial analysis and use it for temporal. It is the Navier-Stokes problem with GCMs – to solve even approximately, you need grid cell dimensions so small that the biggest supercomputer ever could never do it. You need to be able to predict the confluence of all of the rare events that cause the feature and that is just not possible on annual scales or longer.

Not even if you are into crystal pyramids.

The fires were severe on that Saturday because of a confluence of events in the course of a day. Yes, it was correct that the dangers were forecast several days before. But they would not have been forecast to the day, or even at all, on a monthly, annual, decadal or century timescale.

Most climate models attempt to do the latter, to some degree. I agree with David that there is little skill demonstrated in the DECR. My opinion is that it fails because we are dealing with events that are on a “sub-grid” scale if I could borrow a term from spatial analysis and use it for temporal. It is the Navier-Stokes problem with GCMs – to solve even approximately, you need grid cell dimensions so small that the biggest supercomputer ever could never do it. You need to be able to predict the confluence of all of the rare events that cause the feature and that is just not possible on annual scales or longer.

Not even if you are into crystal pyramids.

The Melbourne station is ‘not high quality’, as you know, because it’s in the rain shadow of a building but even more because it’s surrounded by big roads with heavy traffic. Heat Island Central.

But the Melbourne ‘regional grid’ (144.50: 145.50E 37.50: 39.50S) doesn’t show 2009 as the coolest year in the past five either. 2005 is cooler (and 2004 cooler still).

I agree that the ‘official AGW’ views are not supported by this weather data.

Peter

The Melbourne station is ‘not high quality’, as you know, because it’s in the rain shadow of a building but even more because it’s surrounded by big roads with heavy traffic. Heat Island Central.

But the Melbourne ‘regional grid’ (144.50: 145.50E 37.50: 39.50S) doesn’t show 2009 as the coolest year in the past five either. 2005 is cooler (and 2004 cooler still).

I agree that the ‘official AGW’ views are not supported by this weather data.

Peter

2000 20.54
2001 23.44
2002 19.99
2003 21.57
2004 19.45
2005 21.23 Year 5
2006 22.49 Year 4
2007 22.50 Year 3
2008 22.75 Year 2
2009 22.25 Year 1

So my statement (based on average monthly) that January 2009 was the coldest for 5 years would be wrong and would be the coldest in 4 years. However, these are but small fluctuations and they have too many significant figures for the errors involved.

I restate that those few who are linking AGW with these terrible fires are not on entirely solid ground. It might be possible to find a station like Healesville that contradicts this table, but then we start discussing angels on pins as we go looking for a counter-counter example. I don’t play that game.

The “official” AGW contention that were shall see hot days more often and hotter, as the years pass, is also on non-solid ground. The 150 year trend around here is warming, some of it UHI, so I’d expect it to not suddenly reverse at the convenient time of bad fires. But I would expect it to reverse slowly at some time. Those who presume to guess when, are playing at being God. As for frequency, that’s little more than a sick joke by people who need the input of good statisticians and not spin doctors.

2000 20.54
2001 23.44
2002 19.99
2003 21.57
2004 19.45
2005 21.23 Year 5
2006 22.49 Year 4
2007 22.50 Year 3
2008 22.75 Year 2
2009 22.25 Year 1

So my statement (based on average monthly) that January 2009 was the coldest for 5 years would be wrong and would be the coldest in 4 years. However, these are but small fluctuations and they have too many significant figures for the errors involved.

I restate that those few who are linking AGW with these terrible fires are not on entirely solid ground. It might be possible to find a station like Healesville that contradicts this table, but then we start discussing angels on pins as we go looking for a counter-counter example. I don’t play that game.

The “official” AGW contention that were shall see hot days more often and hotter, as the years pass, is also on non-solid ground. The 150 year trend around here is warming, some of it UHI, so I’d expect it to not suddenly reverse at the convenient time of bad fires. But I would expect it to reverse slowly at some time. Those who presume to guess when, are playing at being God. As for frequency, that’s little more than a sick joke by people who need the input of good statisticians and not spin doctors.

My apologies. Figures are for Melbourne Central 86071, January arithmetic average of minimum daily temperature. I counted back for the past 5 years, the 6th year is a bonus, but makes my comment wrong for the wrong reason:

2000 15.3
2001 17.8
2002 14.8
2003 15.6
2004 14.3
2005 15.9 Year 5
2006 17.3 Year 4
2007 17.1 Year 3
2008 17.6 Year 2
2009 15.7 Year 1 Lowest in 5 years.

I can understand that you might have thought I meant the average of max and min, but I did not make my terms clear, nor did I calculate Tav.. So the statement, I think, still stands correct as shown in the table. But what the heck, it’s weather and not climate and my main point was allegedly selective reporting.

My apologies. Figures are for Melbourne Central 86071, January arithmetic average of minimum daily temperature. I counted back for the past 5 years, the 6th year is a bonus, but makes my comment wrong for the wrong reason:

2000 15.3
2001 17.8
2002 14.8
2003 15.6
2004 14.3
2005 15.9 Year 5
2006 17.3 Year 4
2007 17.1 Year 3
2008 17.6 Year 2
2009 15.7 Year 1 Lowest in 5 years.

I can understand that you might have thought I meant the average of max and min, but I did not make my terms clear, nor did I calculate Tav.. So the statement, I think, still stands correct as shown in the table. But what the heck, it’s weather and not climate and my main point was allegedly selective reporting.

I’d be interested to know how you reach the conclusion in your final paragraph.

The BOM data for Victoria and (as far as I can tell) Melbourne doesn’t seem to bear this out. See for example a data set corresponding to the grid for Victoria here: http://www.bom.gov.au/tmp/avetemp_355_1425_395_1505.txt

This shows 2004 had a much colder January.

Peter

I’d be interested to know how you reach the conclusion in your final paragraph.

The BOM data for Victoria and (as far as I can tell) Melbourne doesn’t seem to bear this out. See for example a data set corresponding to the grid for Victoria here: http://www.bom.gov.au/tmp/avetemp_355_1425_395_1505.txt

This shows 2004 had a much colder January.

Peter