Met Office Winter forecast - Updated Feb 6th

[Big Bear]

I think it is a very interesting read, although obviously it doesn't really commit too much either way.

The most interesting aspect for me is that a 'wetter than average signal for Central and Southern Europe is favoured at this time'. That certainly suggests blocking further North, of course if correct it could transpire as blocking over the UK, like much of last winter or blocking to our North East, something which I think will play a big part in the coming winter at this stage.

What was that?! even Ian Brown going for Scandy high pressures in winter?! ...amazing.

I think it was around this time last year the Met Office started to use data in their winter updates which paid off with their accurate LRF. So heres hoping that nothing changes too much between now and the next update (unless its for a severe cold one of course)

[Dawlish]

how many on here remember the winters prior to 1950 let alone 1659. The arguments last winter were that the Met O should use the latest running 30 year mean. That is the official method.

Sure it masks the very long term average but its the recent past that people can recall.

John

The whole point is that very few people remember those winters. Manley knew that and spent most of a lifetime creating the baseline against which we can compare. We have to have that baseline in order to measure change.

You can't just move the baseline every 10 years, because few people remember winters 30 years ago! People's memories are notoriously unreliable. It is the hard statistics, built up over many years that we must compare to and being in the UK, we are luckier than anywhere else in the world in that respect. The NOAA don't run comparisons to a rolling 30 year mean - as Wilson rightly says, it raises difficulties - so why on earth should we?

The arguments last winter stretched much further than simply whether the Met Office should use a 30 year running mean. With respect, your last sentence there, John, is statistical nonsense and it would leave comparison to the past and judgement of how our climate is changing, impossible, if it is adopted. Fortunately it is not as widely adopted as you make out.

Paul

PS love the wild post in LOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOtons! You surprise us with your levity!

[MP-R]

I love the cold! Not that I would be whinging if we got good snow, but it would be really cool if another 28th Jan 2004 event could happen. I've just realised - today is the first day of autumn.

[Dawlish]

Also to note last winter that it was exactly average for the last 100!

BFTP

Spot on BFPT!

Paul

[johnholmes]

The whole point is that very few people remember those winters. Manley knew that and spent most of a lifetime creating the baseline against which we can compare. We have to have that baseline in order to measure change.

You can't just move the baseline every 10 years, because few people remember winters 30 years ago! People's memories are notoriously unreliable. It is the hard statistics, built up over many years that we must compare to and being in the UK, we are luckier than anywhere else in the world in that respect. The NOAA don't run comparisons to a rolling 30 year mean - as Wilson rightly says, it raises difficulties - so why on earth should we?

The arguments last winter stretched much further than simply whether the Met Office should use a 30 year running mean. With respect, your last sentence there, John, is statistical nonsense and it would leave comparison to the past and judgement of how our climate is changing, impossible, if it is adopted. Fortunately it is not as widely adopted as you make out.

Paul

PS love the wild post in LOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOtons! You surprise us with your levity!

each to their own point of view Paul I am a Manley fan, I did meet him once so feel very honoured. However, I am with the Met O on this and not as a past forecaster, simply because it makes sense. And your comment about it being statistically incorrect is incorrect.

John

[Cheviot]

Sky's report of said prediction issue:

http://news.sky.com/skynews/article/0,,30100-1234637,00.html

[johnholmes]

seems okay to me about both items

well done Sky

we await the Mirror etc !!

[Optimus Prime]

By the Metoffice....

Forecast for Winter 2006/7

Temperature

Over much of the European region, the situation is now finely balanced with approximately even chance that the winter will be colder or warmer than average.

For the UK, temperatures near the 1971-2000 average are slightly favoured for the winter season as a whole. However, later in the winter season, there is a signal for lower temperatures (relative to average) and an increase in the frequency of cold snaps.

Note that even an average winter is likely to include some notable cold snaps and snowfalls.

Precipitation

For this winter, wetter-than-average conditions are favoured over central and southern Europe, although for the UK, there is approximately even chance of a wetter-or drier-than-average winter.

Last winter saw much drier-than-average conditions across northern Europe and all parts of the UK - much as last autumn's forecast had suggested. In the south-east of England this was the continuation of a long period of dry weather starting in November 2004, continuing through to the summer of 2006.

[shuggee]

Daily Mail:

Averagest winter for 200yrs may affect house prices.

[Dawlish]

each to their own point of view Paul I am a Manley fan, I did meet him once so feel very honoured. However, I am with the Met O on this and not as a past forecaster, simply because it makes sense. And your comment about it being statistically incorrect is incorrect.

John

'fraid it's not incorrect, John. How can you monitor changing UK climate against a rolling 30 year mean?

Paul

Edited September 21, 2006 by Dawlish

[The PIT]

Okay I'm going to correct myself for an ealier comment no that I haven't got the phone buzzing every five seconds.

Temp wise they're going for an average winter in the UK. Pos chance of colder weather later.

Rainfall wise they're not saying anything it could go either way.

[johnholmes]

'fraid it's not incorrect, John. How can you monitor changing UK climate against a rolling 30 year mean?

Paul

quite simple

Met Office have been doing it for as long as I know. Why all the fuss now?

John

statistics simply have to have VERY clearly stated what they are about and on what they are based.

Edited September 21, 2006 by johnholmes

[VillagePlank]

'fraid it's not incorrect, John. How can you monitor changing UK climate against a rolling 30 year mean?

Paul

I don't think that that's quite the context. If you want a mean to measure against recent trends then the rolling mean will be just fine. It is , of course, summarily difficult to compare one winter to one which occured last year or even 50 years ago, because the means, by definition, will be different.

If you want to say (or, more formally, make a forecast statement) something like 'In the light of recent n years, then my forecast is this, then a rolling mean is probably a quite nice neat way of doing it.

The difficulty is that it's difficult to choose the period of the rolling mean. Is 30 years enough? Is it too much, or is it too little? But these arguments are mute when one considers a fixed mean suffers from the same injustices.

In order to talk to a layman intelligently about future weather states, I can see no problem in using a rolling mean. If you want to compare winters over vast tracts of time, then, as you rightly say Paul, this is nigh on impossible using a rolling mean - although it can be used to reflect, quantitively, the dynamics of a changing climate (how much it changes)

[Snowly does it]

Interesting forecast. It would seem to support the slightly milder December and also support probable blocking. Especially the comment about notable snowfall events possible in such conditions. I picture blocks and snow and mild/cold battles, the temperature going back towards average when the Atlantic breaks through and we go back to a period of SWerlies before the next block moves in.

Going to be a fn one this winter I think

[Anti-Mild]

It was part of my own forecast but the MetO have perhaps gave some positive reinforcement but I foresee a slightly milder than average December but no true zonality, a slightly cool January with blocking dominating and to end, a cold February, perhaps sub 2c, maybe even with a Jan 87 style cold spell (although not as cold as that).

I can see this winter being quite similar to last winter but with the Northern blocking being in a slightly more favourable position.

[kippure]

[Dawlish]

I don't think that that's quite the context. If you want a mean to measure against recent trends then the rolling mean will be just fine. It is , of course, summarily difficult to compare one winter to one which occured last year or even 50 years ago, because the means, by definition, will be different.

If you want to say (or, more formally, make a forecast statement) something like 'In the light of recent n years, then my forecast is this, then a rolling mean is probably a quite nice neat way of doing it.

The difficulty is that it's difficult to choose the period of the rolling mean. Is 30 years enough? Is it too much, or is it too little? But these arguments are mute when one considers a fixed mean suffers from the same injustices.

In order to talk to a layman intelligently about future weather states, I can see no problem in using a rolling mean. If you want to compare winters over vast tracts of time, then, as you rightly say Paul, this is nigh on impossible using a rolling mean - although it can be used to reflect, quantitively, the dynamics of a changing climate (how much it changes)

Agreed Wilson; well put.

Paul

It was part of my own forecast but the MetO have perhaps gave some positive reinforcement but I foresee a slightly milder than average December but no true zonality, a slightly cool January with blocking dominating and to end, a cold February, perhaps sub 2c, maybe even with a Jan 87 style cold spell (although not as cold as that).

I can see this winter being quite similar to last winter but with the Northern blocking being in a slightly more favourable position.

You could really take some virtual tipples off me, if you have the courage to back your position A-M! I think a sub 2.0C month in the coming winter is extremely unlikely!

Paul

[Michael Fishermans Friend]

When exactly did the METO release their LRF for last winter? and when will this be updated next?

Not alot can be read into this forecast, particularly as it will be updated nearer the time, one things for sure that the mid-Atlantic will have to cool down if this blocking is to occur, but as mentioned it is not like the METO to go from mild to sitting on the fence. I wonder why the slight reversal?

[moses]

Very good news for most of us!

---

Forecast for Winter 2006/7

N

Note that even an average winter is likely to include some notable cold snaps and snowfalls.

Precipitation

Is this on the basis that many of the recent winters have been well above average in terms of not snowing and warmth

[tinybill]

Sky's report of said prediction issue:

http://news.sky.com/skynews/article/0,,30100-1234637,00.html

all i say is bring it on lol snow and and the great east wind!!!!!!!!! cant wait

[Anti-Mild]

I know that a lot of the comments on here are tongue in cheek but I posted before the forecast was issued saying that whatever the MetO says does not nescessarily make it certain to happen.

Yes the signs are good, especially with the comment that subsequent forecasts may lean even more towards cold but they do get things wrong!

I feel it is my duty as SACRA founder to try to keep peoples feet on the ground (at least until October!!!).

[Shiny_Bottom_1]

All I will add is its a forecast, like any other. I guess it will be updated in time, again, like many others.

I personally cant see anything wrong with it, in fact, it is really a fair forecast.

It just goes to show how difficult it is to predict so far out, and how things can change.

I cant see any need for criticism, either now, during or after the event.

If they didnt release one, we'd only be complaining.

So lets look at its positives (and negatives) and take it on board.

[Dawlish]

I know that a lot of the comments on here are tongue in cheek but I posted before the forecast was issued saying that whatever the MetO says does not nescessarily make it certain to happen.

Yes the signs are good, especially with the comment that subsequent forecasts may lean even more towards cold but they do get things wrong!

I feel it is my duty as SACRA founder to try to keep peoples feet on the ground (at least until October!!!).

A very balanced post A-M. How many members do you have now? I have a feeling it could be a long winter!

Paul

[BLAST FROM THE PAST]

I think it is a very interesting read, although obviously it doesn't really commit too much either way.

or blocking to our North East, something which I think will play a big part in the coming winter at this stage.

Now that has made me sit up...our very own I "even larger teapot" B making such an early comment...interesting

BFTP

It was part of my own forecast but the MetO have perhaps gave some positive reinforcement but I foresee a slightly milder than average December but no true zonality, a slightly cool January with blocking dominating and to end, a cold February, perhaps sub 2c, maybe even with a Jan 87 style cold spell (although not as cold as that).

I can see this winter being quite similar to last winter but with the Northern blocking being in a slightly more favourable position.

AM

That is not bad at all...in fact I like it a lot inasmuch I support it. Currently though my method points to January being most interesting month with snowiest section. However, I am reviewing and signals are tentative but for Feb to be pretty cold. Your Jan 87 cold spell to me is set for Jan I am looking at 3.4 and 2.8 approx either way at present with Dec coming in at 6.8...all approx of course but one can get the gist.

BFTP

Edited September 21, 2006 by BLAST FROM THE PAST

[North-Easterly Blast]

Sounds a reasonably good forecast to me, probably indicating a winter that's likely to be close to last winter in terms of overall temperature which would be reasonable for recent years anyway. They mention about El Nino developing, and in 97-98 there was a strong El Nino and we all know what a disastrously mild winter in the UK that one turned into. At least they have not gone for a 97-98 disaster. Is this coming winter's El Nino likely to be a strong one or a weak one? They did issue a positive NAO forecast in July, but to me it suggests they are now expecting a more neutral NAO with it going either way at times.

The Met Office to me seem as though they are suggesting that a winter similar to 05-06 or even 00-01 is a likely outcome at this stage, but with an El Nino on the development, in my view increases the chance of the winter turning out a real stinker like 97-98.

I know winter 97-98 had a strong El Nino in place, but what other winters did. I do not know myself and neither do I know winters that had a weak El Nino in place.