(Update:- Oops! I’ve found a glitch in my calculations which I have corrected. This has the effect of changing and adding some predicted events- see below.)
Firstly, my apologies for not attending to this blog more regularly. Retirement certainly doesn’t mean having more time.
This is what I predicted back in April:
Meanwhile, refined predictions for April to early May (>90% certain) and long range predictions to early September (about these dates, plus several others), for weather changes (cold change, pressure drop, or rain):-
April:
3-4-5-6-7-8-9; 10-11-12; 13-14-15 (A succession of small changes)
17-18-19-20; 22-23-24
25-26–27-28
May:
April 30-1-2-3-4-5
7-8-9-10-11-12-13-14 (unsettled for the first 2 weeks).
(Pink is Tmax, dark red is Tmin, blue is rain, black is inverted mean sea level pressure.)
11 Correct, 1 Wrong. 91.6%. (I could be persuaded that there were only 10 major changes, which makes it 83%). Moreover, my prediction of a succession of small changes in early April and unsettled conditions for the first two weeks of May was definitely correct.
So far this year, my short term (up to 40 days) predictions fared as follows:
|
Predictions of changes |
Correct |
No change |
Unpredicted changes |
|
33 |
32 |
1 |
2 |
That’s 91.4 %.
160 day predictions (from 27 November and 2 January)
“April: 7, 9, 11, 13, 15, 19, 27.
May: 4, 9, 13, 19, 29”
up to today, were for 25 events so far this year, of which 23 occurred as predicted (later refinements removed one of the predicted events). 92% correct for the year, 80% for April and May so far. That is also 67.6% of all events.
I note the BOM forecast is for fine weather this weekend, whereas from 2 January I have been predicting a change for 19 May.
My predictions for sub-tropical Queensland for the rest of May and up to mid-October (temperature change, pressure drop, or rain) are as follows:
Unsettled with several successive events for the next several weeks:
May
17-18-19-20, 21-22-23, 24-25-26-27-28-29-30-31-1-2 June
June 5-6-7-8-9-10-11, 13-14-15-16-17, 20-21-22-23, 27-28-29-30
July
2-3-4-5-6-7, 9-10–11-12, 15-16-17-18-19-20-21, 25-26–27-28-29-30
August
1-2-3-4-5-6-7–8-9, 11-12–13-14, 15-16-17-18-19-20-21, 23-24-25, 27-28-29-30–31
September
1-2-3-4-5-6–7–8-9, 12-13–14-15, 18-19-20, 22-23-24-25-26-27-28
October
September 30-1-2, 8-9–10-11, 14-15-16.
Do you know of anyone else predicting a weather event (probably storms) for 15 October 2013?
I am still working on an objective test for my predictions, which I am close to finalising. One thing I have found is that the frequency of weather changes does vary from about 22 to 38 in a 90 day period, which means there are often lengthy periods of changes occurring at least once every three days, which would make accurate forecasting much more difficult. In a future post I will also show how my 160 day predictions are even more accurate for individual locations than when averaged over a whole region.
kenskingdom 