On Tuesday, March 4, 2014 5:05:06 PM UTC, Scott W wrote:
On Saturday, November 30, 2013 2:02:04 PM UTC, Scott W wrote:

I have posted a winter forecast on wordpress - the link of which is http://wp.me/p2VSmb-hY







I have tried to be as objective as possible, basing my findings on means and what has happened in the months of October and November.







Because it is based on data stretching back over 130 years to my local area it can only really by applied to this part of the south-east.







It is not a hopecast. As much as I like snow I am much less keen on the ever rising cost of heating fuel - so find I'm much more pragmatic about it. If it snows it snows.







For those that don't want to visit wordpress the body text follows:







Much has been printed in the tabloids over the past month or so that we are in for a severe winter. Not a week goes by without the Daily Express splashing that the 'Worst winter EVER is on the way'. Just this morning the same rag told us that three months of 'exceptionally cold' weather are due. On closer inspection the story elaborated the scene with quotes from James Madden of Exacta Weather, one of the 'experts' feeding these fantastical stories.







Quite how Mr Madden and other experts arrive at these forecasts is a bit of a mystery. The mystery has deepened further since I decided to crunch a few numbers and try to predict what is in store for the months ahead. Looking at data for this area stretching back over 130 years to 1881 I decided to calculate a seasonal average and arrived at a final figure using singularities - basically looking at the weather patterns we've had during October and November.







Many professionals would scoff at this method of pattern-matching, so I've incorporated a couple of other 'now' factors and taken on board current variables such as sea surface temperatures in the Atlantic and Pacific.







The figure I arrived at, taken as an average of the closest matching autumn periods, is a mean temperature of 4C with rainfall totalling 133mm over the months of December, January and February - that's about a degree colder than average and 90% of average rainfall. The probability of a winter with a mean temperature of between 4C and 5C is 37% - the most likely outcome.. With this in mind a winter in the form of 1986-87 is possible - though whether we would see the same extremes of temperature and snowfall that we experienced in January 1987 only time will tell.







Because this format can't cope with the formatting on usenet I have posted a full methodology of my forecast here.http://sdrv.ms/18sYo9N







Anyway, I hope some of you find it an interesting read. I'll be revisiting it on March 1. Best regards, Scott Whitehead



Apologies all but this is the first chance I've had to return to this...



A poor forecast, I'm sure you'll agree, which seemed doomed to failure as soon as the Atlantic cranked into action with such persistence in the middle of December.



Prediction: Mean: 4.0C Rainfall: 133mm

Outcome: Mean: 6.7C Rainfall: 272mm



Looking back it seems that I should have paid more attention to one or two signals that the winter would pan out overall mild - but I think I paid too much attention to broadly neutral ENSO and NAO signals which suggested winter would be 'average' at the very warmest. And perhaps was lulled into a sense of false security that we'd lurched back to a time of colder winters..



Considering the stats going back to 1881 the chance of an exceptional winter such as we've just had was just 5 per cent. And a repeat of 1963 was 1 per cent.



A review of the stats shows that out of my original shortlist of 19 winters that were preceded by an October and November similar to 2013, 4 of them were 6C or warmer - a not insignificant 21 per cent occurrence. However, none of them were anything like as wet. The wettest - 1924-25 - was still only 73 per cent as wet as this winter.



Is predicting a season ahead impossible? Perhaps. But since I made this forecast I've acquired yet more local data - back to 1797 - and will be back again in 9 months, hopefully wiser and with better signals at my disposal.



Cheers all - here's to a long and decent summer



The original methodology is here http://sdrv.ms/18sYo9N

Thanks Scott. A poor forecast, yes, but I really appreciate you returning to this forecast and explaining why you think it was so poor!