Met Office Winter forecast - Updated Feb 6th

[Gray-Wolf]

Am I correct in assuming that what has been said over the past couple of pages is that conditions that indicate possible future trends are still, at present, either in flux or not at a point where the data is useful.

If so does this means that in 6 to 8 weeks people will start to firm up their ideas of coming winter trends ?

This would mean that the MetO release was just a kind of 'heads up' to advise that their winter trends forcast has changed due to alterations in their data pool.

EDIT: I would tend towards what Dawlish is saying in his post above. The overall trend has to be towards warmer winters but that doesn't exclude events like those in Eastern Germany/Poland or Japan last year with very extreme winter weather events in amongst the 'more of the same' type winter weather. I think we were very unfortunate/fortunate not to cop for a countywide 3 to 4 day snow event last year and hope that this year one of those potent Northerly outbreaks will bring us just that (I don't know about a deep cold pool to our East this year though)

[Dawlish]

Am I correct in assuming that what has been said over the past couple of pages is that conditions that indicate possible future trends are still, at present, either in flux or not at a point where the data is useful.

If so does this means that in 6 to 8 weeks people will start to firm up their ideas of coming winter trends ?

This would mean that the MetO release was just a kind of 'heads up' to advise that their winter trends forcast has changed due to alterations in their data pool.

1. Definitely. as they always are this far out.

2. In 2 months, they'll still be guessing.

3. Yes. Still has only a reasonable chance of happening.

Paul

[Snowly does it]

1. Definitely. as they always are this far out.

2. In 2 months, they'll still be guessing.

3. Yes. Still has only a reasonable chance of happening.

Paul

The data is ever changing certainly, but the MetO clearly feel there is value in longer range forecasting and seem to have some success with it. I think to say they will be 'guessing' in 2 months time is rather disingenuous, they are not 'guessing' now, they are presenting a long range forecast based on the data to hand and incorporating their experience in meteorology which cannot be denied, is extensive.

The MetO release was not a 'heads-up', they had planned to release their first formal winter forecast at this point, and did so.

In regards to the earlier point about synoptic change, I think there is in fact a probability that this will occur as a direct result of GW. If one considers the meandering of the Azores High northward in recent times then clearly a warmer world has an effect. I would anticipate this leading to anomolously cool periods (in winter rather than summer) and much warmer ones as things continue to alter (relative to the UK)

[TEITS]

For me, frozen, they are the best in the world and the most experienced, in forecasting our weather in the UK. Pound for pound, over the years, I feel they have been better than the American agencies and they still are.

No-one's perfect, but UKMO do us proud, for a large percentage of the time.

Paul

Im a little confused Dawlish because on the one hand you say the above, and yet on one of your earlier post's you question if the Met O take into account GW when producing a LRF!. This TBH honest is a silly statement especially when you take into account the experience & expertise that goes into their LRF's and all the data they use to come to these conclusions.

Even though the Met O LRF is a little vague I would much prefer an honest forecast than a wrong one. Im sure as time progresses the Met O with firm up the details of this coming winter as more data comes in.

[Guest Viking141]

Quite so TEITS they got it pretty spot on last year.

[Dawlish]

Im a little confused Dawlish because on the one hand you say the above, and yet on one of your earlier post's you question if the Met O take into account GW when producing a LRF!. This TBH honest is a silly statement especially when you take into account the experience & expertise that goes into their LRF's and all the data they use to come to these conclusions.

Even though the Met O LRF is a little vague I would much prefer an honest forecast than a wrong one. Im sure as time progresses the Met O with firm up the details of this coming winter as more data comes in.

No, I've always been consistent about that. The Met Office describe their NAO forecast as "experimental" and I class all and every long range winter forecast as little better than a guess at this range. The stats prove it. I've also always supported the Met Office, but no matter how much expertise and experience goes into their long-range forecasting, it is still far from being reliable. In the UK, they are the best, because of their experience and expertise; but their best does not produce anything approaching a good percentage success rate.

Silly? Nah. <_<

Paul

[TEITS]

No, I've always been consistent about that. The Met Office describe their NAO forecast as "experimental" and I class all and every long range winter forecast as little better than a guess at this range. The stats prove it. I've also always supported the Met Office, but no matter how much expertise and experience goes into their long-range forecasting, it is still far from being reliable. In the UK, they are the best, because of their experience and expertise; but their best does not produce anything approaching a good percentage success rate.

Silly? Nah. <_<

Paul

Im not talking about how reliable their forecast's are my point was in reference to your comment about them not taking into account GW when producing their LRF. I still maintain this was not just silly but a stupid comment.

[guitarnutter]

Why do they bother? any weather forecast is rarely right past about 3 days, so whats the point forcasting for more than about 5 days away, and especially months off.

[Snowly does it]

No, I've always been consistent about that. The Met Office describe their NAO forecast as "experimental" and I class all and every long range winter forecast as little better than a guess at this range. The stats prove it. I've also always supported the Met Office, but no matter how much expertise and experience goes into their long-range forecasting, it is still far from being reliable. In the UK, they are the best, because of their experience and expertise; but their best does not produce anything approaching a good percentage success rate.

Silly? Nah. <_<

Paul

OK, look at this another way, if you will.

In terms of what they put out, the success rates are not eye-opening at this stage. However that is far from saying that the forecast is a 'guess'. That imples an irresponsibility on the part of the MetO that I do not feel is warranted. The forecast as delivered is based upon the best possible LRF techniques available from the MetO perspective and is deliberated carefully. I have no contention over the actual success rates, more over the terminology in play here.

I mean, percentage based systems are guesswork right? 70% chance of being warmer, well I guess it will be warmer... etc etc. Trend noted, forecast unavailable.

It is important to keep faith in research into LRF because let us assume we just go for 70/20/10 etc. How long into a change do we have to get before changing that number system? In a fantasy world of a sudden cooldown, many years, by then the 70/20/10 has been shown to be woefully inadequate. LRF is proactive, probability is reactive. Longer term probability will probably be more effective purely on numbers, but it makes no specific forecast, so for forward planning or anything else similar, it is of no use.

Why do they bother? any weather forecast is rarely right past about 3 days, so whats the point forcasting for more than about 5 days away, and especially months off.

It allows local authorities to make provisional plans for whatever season is in question.

[disco-barry]

Well i hope it is colder than expected but if it isnt it isnt it's beyond there control all they can do is anticipate what will happen by reading there computers and sticking there wet finger in the air.....

[Dawlish]

Im not talking about how reliable their forecast's are my point was in reference to your comment about them not taking into account GW when producing their LRF. I still maintain this was not just silly but a stupid comment.

Perhaps I should have explained in more detail. I was making an inference, that I have explained before, about the set up of their forecasting model, which "takes into account a range of factors". It doesn't, in my opinion, have enough bias towards a warming trend in the perameters they use, Eye. There will be prioritisation built into that model, with particular features having more weighting than others. For example, I would imagine that their NAO forecast has quite a high weighting. I feel their success rate would be better if they weighted the warming trend more heavily than they presently do. A simple technique of predicting the % possibility of a cooler than average winter, from the percentages over the past 16 years, then extrapolating this, produces a better % accuracy than the Met Office achieves.

The Met Office is under pressure to predict every winter, however, not just to give an indication of the chances. netweather is under the same pressure, as is TWO, metcheck, UKww and other Internet agencies. Their customers expect it. Unfortunately, the expectations of the customers are not yet matched by the capacity of the forecasting agencies to meet them. In other words; Long Range forecasting is in its infancy, whereas the public and commercial customers expect something approaching perfection.

That's why i feel the Met Office is between a rock and a hard place when it comes to LRF. They are damned if they do, because an unacceptable (in the eyes of the general public) success rate means they will get a proportion of their forecasts badly wrong and they are damned if they don't, because who wants forecasters who wont forecasts the coming season.

There's a lot of science behind that TEIS, It's really not silly, or stupid, honest!

Paul

[mcweather]

Perhaps I should have explained in more detail. I was making an inference, that I have explained before, about the set up of their forecasting model, which "takes into account a range of factors". It doesn't, in my opinion, have enough bias towards a warming trend in the perameters they use, Eye. There will be prioritisation built into that model, with particular features having more weighting than others. For example, I would imagine that their NAO forecast has quite a high weighting. I feel their success rate would be better if they weighted the warming trend more heavily than they presently do. A simple technique of predicting the % possibility of a cooler than average winter, from the percentages over the past 16 years, then extrapolating this, produces a better % accuracy than the Met Office achieves.

The Met Office is under pressure to predict every winter, however, not just to give an indication of the chances. netweather is under the same pressure, as is TWO, metcheck, UKww and other Internet agencies. Their customers expect it. Unfortunately, the expectations of the customers are not yet matched by the capacity of the forecasting agencies to meet them. In other words; Long Range forecasting is in its infancy, whereas the public and commercial customers expect something approaching perfection.

That's why i feel the Met Office is between a rock and a hard place when it comes to LRF. They are damned if they do, because an unacceptable (in the eyes of the general public) success rate means they will get a proportion of their forecasts badly wrong and they are damned if they don't, because who wants forecasters who wont forecasts the coming season.

There's a lot of science behind that TEIS, It's really not silly, or stupid, honest!

Paul

I agree with you about the rock and the hard place Dawlish. However if the METO put more weighting into the supposed warming trend they will almost certainly never forecast the next worthwhile winter that we do get with any accuaracy and imagine the outcry then!

[Snowly does it]

I agree with you about the rock and the hard place Dawlish. However if the METO put more weighting into the supposed warming trend they will almost certainly never forecast the next worthwhile winter that we do get with any accuaracy and imagine the outcry then!

edited

They could merely shift their baseline in accordance with continued warming. Cold/Average/Mild will always be related by most people to the recent trend and experience in any case.

[SMU]

Good evening everyone

I must be quite frank and say a lot of what you are trying to suggest Dawlish is wholly incorrect. The long term trend is for the climate to warm, but this does not mean that within a trend there cannot be shorter term periods of cooling. A forecasting organisation can only forecast what the data in front of them presents, climatic discussion and long term climate change has little or nothing to do with that fact. If every long range forecasting model or methodology were to reflect this continual warming then every forecast since the 1990's would have to continue the trend of record breakingly warm winters. This though has not been the case.

The forecast they have given is an indication. It is based on quantifiable data, so is certainly not a guess even if the conclusions they are drawing from such data may be experimental.

Your suggestion that forecasting centres and independant organisations are under pressure to give long range forecasts by customerseven though they are yet to provide any reasonable information is also untrue. Customers of these organisations require useful forecasts, they would not be pressuring to see a forecast that they have no belief in and have not found of use previously, so we have to assume that previous years and seasons lrf's have proved to be of some use to them.

The world is not black and white, even with an early indication of weather for the coming season forward planning becomes possible. Your assertion that there is no value in this and that it is purely guesswork is more an indication of your lack of knowledge on the subject. The one point you are partially correct about is the infancy of the climatic modelling that is now becoming more prevelant but to dismiss it because of this is quite ridiculous.

Ditto-

The usual high level of contribution I expect-

Global Warming never negates local cooling, The preliminaries the Meto present are consistent with what most Seasoned LRFers I talk to are expecting-

In the November update I expect more of a swing away from the mild type winter scenario-

Steve

[summer blizzard]

I agree Steve, but one question i would like to ask is whether current indications favour a wet or dry winter, my own reasearch is inconclusive and matches the Met Office preliminary at the moment, with an even spread of wet and dry anologues???

Weather Forecaster, mobile does not always mean mild, take the winters of 1947 and 1979 as an example.

[johnholmes]

Good evening everyone

I must be quite frank and say a lot of what you are trying to suggest Dawlish is wholly incorrect. The long term trend is for the climate to warm, but this does not mean that within a trend there cannot be shorter term periods of cooling. A forecasting organisation can only forecast what the data in front of them presents, climatic discussion and long term climate change has little or nothing to do with that fact. If every long range forecasting model or methodology were to reflect this continual warming then every forecast since the 1990's would have to continue the trend of record breakingly warm winters. This though has not been the case.

The forecast they have given is an indication. It is based on quantifiable data, so is certainly not a guess even if the conclusions they are drawing from such data may be experimental.

Your suggestion that forecasting centres and independant organisations are under pressure to give long range forecasts by customerseven though they are yet to provide any reasonable information is also untrue. Customers of these organisations require useful forecasts, they would not be pressuring to see a forecast that they have no belief in and have not found of use previously, so we have to assume that previous years and seasons lrf's have proved to be of some use to them.

The world is not black and white, even with an early indication of weather for the coming season forward planning becomes possible. Your assertion that there is no value in this and that it is purely guesswork is more an indication of your lack of knowledge on the subject. The one point you are partially correct about is the infancy of the climatic modelling that is now becoming more prevelant but to dismiss it because of this is quite ridiculous.

nice to see you posting WF and I have to agree with almost every word.

John

[SMU]

I agree Steve, but one question i would like to ask is whether current indications favour a wet or dry winter, my own reasearch is inconclusive and matches the Met Office preliminary at the moment, with an even spread of wet and dry anologues???

Weather Forecaster, mobile does not always mean mild, take the winters of 1947 and 1979 as an example.

I would say Dry- With plenty of Split flow, & an anamolous storm track to boot- ( Versus the norm)

Anyway- this topic will have to be covered in the Autumn & Winter discussion-

We have a full 6 weeks before the next METO update so lets wait & See.........

S

[Dawlish]

Good evening everyone

I must be quite frank and say a lot of what you are trying to suggest Dawlish is wholly incorrect. The long term trend is for the climate to warm, but this does not mean that within a trend there cannot be shorter term periods of cooling. A forecasting organisation can only forecast what the data in front of them presents, climatic discussion and long term climate change has little or nothing to do with that fact. If every long range forecasting model or methodology were to reflect this continual warming then every forecast since the 1990's would have to continue the trend of record breakingly warm winters. This though has not been the case.

The forecast they have given is an indication. It is based on quantifiable data, so is certainly not a guess even if the conclusions they are drawing from such data may be experimental.

Your suggestion that forecasting centres and independant organisations are under pressure to give long range forecasts by customerseven though they are yet to provide any reasonable information is also untrue. Customers of these organisations require useful forecasts, they would not be pressuring to see a forecast that they have no belief in and have not found of use previously, so we have to assume that previous years and seasons lrf's have proved to be of some use to them.

The world is not black and white, even with an early indication of weather for the coming season forward planning becomes possible. Your assertion that there is no value in this and that it is purely guesswork is more an indication of your lack of knowledge on the subject. The one point you are partially correct about is the infancy of the climatic modelling that is now becoming more prevelant but to dismiss it because of this is quite ridiculous.

Good Evening WF. I'm afraid the proof of the pudding is in the eating. "Useful forecasts", by definition, must have a good level of accuracy, or at least a level of accuracy acceptable to their customers - what is their actual percentage accuracy and what does the general public really think of long-range forecasts in general?

There is no dismissal, rather an assessment of the current state of LRF. It is in it's infancy and not a single organisation, that I know of, has produced a verifiable accuracy above 70%, Even 70% is less than acceptable. In the future, I'm sure it will be better and research is both necessary and continuing, but at this moment, LRF is just not good enough to rely on.

The warming trend cannot, and never will be, linear. Since 1990, there have been cooler than average winters and there will be cooler than average winters in the future. Anyone who thinks otherwise would be wrong. Of course there can be cooler interludes in a warming trend. In a tiny area (UK), within a chaotic, complex, global system, it is bound to happen; that, however, changes neither the trend, not the odds on any winter being cooler, or warmer, than average.

Whatever this winter turns out to be, it will only be a single point on a developing graph. I maintain that there is not a single individual, or a single agency that can guarantee a reasonable seasonal forecasting accuracy, though somebody will probably get this winter spot on. That is why the Met office uses a different formula to other Internet forecasting agencies and why they are all still searching for an accurate method, which has not yet been found.

A good agument, well put, but to describe a lot of what I say as "wholly incorrect", is wholly incorrect.

Regards, Paul

[BLAST FROM THE PAST]

Blimey guys, some very interesting posts tonight Paul within your post it suggests that UK is affected almost beyond repair...the goal posts have not moved that much if indeed they have. Without a doubt in my mind a synoptic pattern shift has occurred since Feb 05 and this autumn/winter will go on to confirm that we will see quite a variable winter and a quite often southerly jetstream.

Without going on 77/78 'style' winter beckons

BFTP

[Dawlish]

Blimey guys, some very interesting posts tonight Paul within your post it suggests that UK is affected almost beyond repair...the goal posts have not moved that much if indeed they have. Without a doubt in my mind a synoptic pattern shift has occurred since Feb 05 and this autumn/winter will go on to confirm that we will see quite a variable winter and a quite often southerly jetstream.

Without going on 77/78 'style' winter beckons

BFTP

Not "beyond repair", BFTP, but I would say that the UK has been affected significantly, by warming. As snowmaiden has said (I think I'm quoting you correctly, sorry If I've botched it, memory and all.....) there is likely to be a pattern change with increasing warmth and I agree with that, but that would take a chunk of time and a longer to become apparant. Whichever pattern shift occurred would be one which is more likely to shepherd in warmer conditions, over the whole year, than colder ones.

Paul

[BrickFielder]

My understanding about climate models is that they are tested in their ability to accurately predict previous years weather. By that I mean the model should take the conditions from the summer of 1976 and be able to predict the conditions for the following winter, verifyable by what actually occured. What tends to happen is that the models have a built in bias towards what occured over the last few years because better accuracy is obtained that way. Where sudden and dramatic changes occur then that bias can make the forecast inaccurate and there have been some notable occasions in recent years.

As to the accuracy of the climate models then there is a mixed bunch of results. HADGEM2 which I think is one of the latest by the Hadley center has been shown to get an accuracy of about 90% predicting the NAO for years gone by, based on Sea Surface Temperatures ,Sea Ice extents and Snow cover. This shows significant improvements over the last few years, but it does have limitations in terms of accuracy.

The average persons memory (probably not a netweather posters) about seasons does not extend back very far and winter tends to be rated in comparison to the previous winter. In this way a winter is rated warm or cold by many people almost in isolation to any underlying trends.

For all the limitations of long range forecasts they become important for those who try to plan gas and electricity supplies and for those who need to buy grit for roads and organise other winter based requirements. These are areas where people work in percentages and damage limitation and where a 60/40 bias can tip a decision one way or another. Providing these decision makers can show their decisions were based on reasoned arguments they can feel reasonably guilt free if things go wrong. This may not be how things should work, but the large sums of money involved mean long range forecast are likely to continue whether they are accurate or not.

[VillagePlank]

On the very basis that there is this discussion then I’m afraid I will have to agree that LRF is still at an unacceptably low level. If one can point me to this study or that which validates and verifies a LRF technique then, we can bring this discussion to an end.

As no one has I can only assume either that LRF is still considered unreliable, or that someone is trying to hide the evidence that it’s not from me. I reject the second on the basis that it is entirely improbable that the complete membership of netweather are conspiring against me . . .

I am, therefore, left with the only conclusion which accounts for the very existence of this debate; and that is that long-range forecasting is unreliable.

[Rustynailer]

Long range forecasts are interesting and long may people continue making them 70% is better than nothing......

[Dawlish]

Good morning Dawlish

As I suggested earlier I really do feel that you are trying to mask your lack of knowledge on this subject with statements presented as fact which are in fact untrue.

That would be disingenuous and, to use your own words....wholly incorrect. I'm afraid you are not in a position to state that and I think you may need to scroll back and read my other posts on and around this subject to determine a better judgement.

Paul

[Dawlish]

My understanding about climate models is that they are tested in their ability to accurately predict previous years weather. By that I mean the model should take the conditions from the summer of 1976 and be able to predict the conditions for the following winter, verifyable by what actually occured. What tends to happen is that the models have a built in bias towards what occured over the last few years because better accuracy is obtained that way. Where sudden and dramatic changes occur then that bias can make the forecast inaccurate and there have been some notable occasions in recent years.

As to the accuracy of the climate models then there is a mixed bunch of results. HADGEM2 which I think is one of the latest by the Hadley center has been shown to get an accuracy of about 90% predicting the NAO for years gone by, based on Sea Surface Temperatures ,Sea Ice extents and Snow cover. This shows significant improvements over the last few years, but it does have limitations in terms of accuracy.

The average persons memory (probably not a netweather posters) about seasons does not extend back very far and winter tends to be rated in comparison to the previous winter. In this way a winter is rated warm or cold by many people almost in isolation to any underlying trends.

For all the limitations of long range forecasts they become important for those who try to plan gas and electricity supplies and for those who need to buy grit for roads and organise other winter based requirements. These are areas where people work in percentages and damage limitation and where a 60/40 bias can tip a decision one way or another. Providing these decision makers can show their decisions were based on reasoned arguments they can feel reasonably guilt free if things go wrong. This may not be how things should work, but the large sums of money involved mean long range forecast are likely to continue whether they are accurate or not.

I agree entirely with your last sentence Steve and much of what you say and I'm glad my points have raised the discussion. The more research, the better. There is a key in there, it just hasn't yet been found.

Your 90% success rate, quoted there and just so nobody comes back with this, as figures like that tend to stick, is not a prediction success rate. I do appreciate it could bode well for the future.

Again, I mentioned that no-one, to my knowledge has achieved a 70% success rate at predicting seasonal weather accurately. That actually means that the success rates are less than 70%, maybe 2 out of 3. The Met Office and others, have a number of paying customers and some large utilities either increase, or run down their stocks, based upon the LRF they pay for, that is true; last year's leak of the "colder than average winter" came from an ill-advised article by a UNISON member, following a Met Office presentation to their commercial customers, on exactly that topic, but only 2/3 is one hell of a gamble and I'm sure they know it. It would not be hard to imagine the furore when a cold winter does come along (and it will) and the 1 in 3 chance of the forecast being wrong drops on that year (which, statistically, it will). It is that which will be remembered.

The number of commercial customers has grown in recent years and their expectation of getting their money's worth, with a correct forecast has grown with it. I only hope that the advances in forecasting techniques improve before the combination of a cold winter and a less than 70% success rate collide.

Paul

PS If I'm wrong about the <70% and an organisatiuon has demonstrated a >70% accuracy, whose dataset is statistically significant at the 95% level (ie is long enough to demonstrate reliability), tell us, anyone, please. WF? You seem to have some knowledge here? Now that would be a forecast worth investigating. I know Piers Corbyn claims a success rate greater than that, but he has never made his research available to the scientific community, claiming (perhaps rightly) commercial confidentiality.