Rjs And Bftp Winter 2010/11 Lrf

[The PIT]

If you don't understand the term bust look it up in a dictionary. :)

The first predicted storm for Dec 6th did actually show up for a while before being zapped by the models. So that's a miss. The 13th as Rodger says didn't go to plan which is also a miss. Cold spell over all correct. They maybe a hit on the 28th which all depends on how far the mild push gets.

Somebody else was on here was using the same method a year ago on here and had a very high hit rate. Can't remember who but they don't seem to post anymore.

The main problem is if you put in dates and the storms don't happen it does mean your predictions are wrong.

Overall they haven't done any worse than the met office who've also struggled to predict heavy snowfall despite making the call at short notice.

[sandstorm2]

Just got to say that your long LRF is fascinating and December has been very accurate. The methods you use appear to be as good as any 'scientific' based forecasts and there must be a place in meteorology for your more 'natural' based forecasts.

Looking forward to your January forecast.

I am almost on the verge of giving up on conventional meteorology for LRF after following your methods for quite some time.

SS2

[VillagePlank]

Not sure if this is any help?

[Magadan]

If you don't understand the term bust look it up in a dictionary. :)

The first predicted storm for Dec 6th did actually show up for a while before being zapped by the models. So that's a miss. The 13th as Rodger says didn't go to plan which is also a miss. Cold spell over all correct. They maybe a hit on the 28th which all depends on how far the mild push gets.

Somebody else was on here was using the same method a year ago on here and had a very high hit rate. Can't remember who but they don't seem to post anymore.

The main problem is if you put in dates and the storms don't happen it does mean your predictions are wrong.

Overall they haven't done any worse than the met office who've also struggled to predict heavy snowfall despite making the call at short notice.

You mean the Met Office was nowcasting whereas these chaps were predicting it weeks in advance.

Personally I don't give a toss about when a storm is meant to blow through. It's LRFs that are important, i.e., trends for the various seasons. Farmers & agribusinesses need the latter, wedding planners and Morris dancers need the former.

[sandstorm2]

You mean the Met Office was nowcasting whereas these chaps were predicting it weeks in advance.

Personally I don't give a toss about when a storm is meant to blow through. It's LRFs that are important, i.e., trends for the various seasons. Farmers & agribusinesses need the latter, wedding planners and Morris dancers need the former.

Very well put, couldn't have done it better myself.

SS2

[The PIT]

Forgot to mention they've got xmas right quiet and cold poss freezing fog. Now all eyes on the next date the 28th give or take a day. Met very unsure so maybe plenty of freezing rain then rain or snow if the the cold air sits. The other question is how far east it pushes.

[The PIT]

Well it did snow on the 28th albeit light and no sign of any more snow to the end of the month.

So Bulls eye xmas double forty for the cold treble sixty for the 28th missed the dartboard on the rest well the dart went in then bounced out. Overall not a bad effort.

All eyes to Jan more hits than missed or more misses than hits or bulls eyes all the way???

[Weather-history]

DECEMBER - CET 2.8C

There are indications of a very cold December with possible heavy snowfall events in central and northeast England, but also some snow from cold northwest flow patterns. We're predicting a rather low CET value of 2.8 C for December and some stretches of sub-freezing temperatures, most likely in the middle third of the month.

Moderate to heavy snowfalls are possible during this cold spell and most likely dates for these would be around the 6th and 13th (although 2-4 day periods of snowfall are likely). The southwest may escape this phase of the cold to some extent and see chilly rainfalls at least near sea level. With this early snow cover coming before the longest nights of the season, there is some chance of some notably cold overnight lows developing as high pressure swells up over northern central Europe in the days before the full moon on the 21st. An important storm event is likely to accompany this full moon, and it may produce another snowfall at least away from the south coast as the track of the event based on past analogues may be near to the Channel. The storm is likely to approach from WSW and is likely to slide under the block to the NE. If the trend happens to be milder than we're using as our base-line, this storm would bring mild, windy and rainy conditions but current indications for those conditions seem destined more for France and perhaps the coastal parts of Devon and Cornwall.

Following this storm system Christmas is likely to be dominated by HP and calm, cold and potentially freezing fog conditions may be prevalent.

Pressure is always higher to northern quadrant and the general flow remains North of E and W[indeed the set uop projected by models currently is a very good indicator of pressure pattern]. Further snowy conditions are possible shortly afterwards say 28th (all these snowfall forecasts are based on energy level peaks which are expected to form a fairly regular 7-day pulse in mid-winter, evolving into a 3.5-day cycle by February as two separate chains of events with different timing separate out from each other). We should note also that the energy level peaks are very strong this winter, so that parts of Europe will be battered by very strong winds and heavy precipitation. In general, look for these events around the new and full moons for simplicity.

Hi Roger and Blast.

You went for a very cold December and it turned out as such but given your techniques and analysts in determining your forecast, were you surprise that December was as cold as it turned out?

I have asked the strato guys this question because I think it is good to look at what causes extreme months and can anyone actually reasonably predict an extreme month.

[lorenzo]

I think Mr Data posts a good question, also is the outlook on this forecast still for a milder second period in January ? Cold seems to be more persistent now.

Another point of interest is in the ' Better than the models' thread MB has posted a presentation re: a great Storm for early February. Would be interested to know you guys thoughts on this and how it fits in with what you see is on the table for the beginning of February.

[BLAST FROM THE PAST]

Hi Roger and Blast.

You went for a very cold December and it turned out as such but given your techniques and analysts in determining your forecast, were you surprise that December was as cold as it turned out?

I have asked the strato guys this question because I think it is good to look at what causes extreme months and can anyone actually reasonably predict an extreme month.

Hi Kev

The main reason for a CET punt as we did is an overall view and alos took into view the background warming of previous decades so we were conservative with estimates, but also we highlighted that out CET might not tell the whole story. The depth of cold was not a surprise at all and indeed on the monthly CET RJS put 0.8c as his punt. Incidentally if Feb turns out as envisaged then a month to match Feb 86 in cold is in the offing....Roger and I have discussed Feb 1942 as a similar set up?

BFTP

[BLAST FROM THE PAST]

I think Mr Data posts a good question, also is the outlook on this forecast still for a milder second period in January ? Cold seems to be more persistent now.

Another point of interest is in the ' Better than the models' thread MB has posted a presentation re: a great Storm for early February. Would be interested to know you guys thoughts on this and how it fits in with what you see is on the table for the beginning of February.

There is New Moon on Feb 3rd but it isn't a particularly strong signal for storm potential. The general synoptic pattern doesn't look out of place though. I have had a look and I still don't see a particular 'notably strong event' signal for this period.

The forecast stands but 13-16 Jan is potential pivotal period. I still believe that Jan will be the less cold month of the winter.

BFTP

Edited January 4, 2011 by BLAST FROM THE PAST

[Roger J Smith]

Kevin (and others),

Our CET forecast numbers tend to the conservative side when we suspect extremes may be forthcoming. I think we are making progress towards a working system and as with any other time scale in forecasting, as progress is made, more confidence will attach to forecasts, ranges and outcomes. I think we made it really clear that we were expecting something that might rival Dec 1981 and in fact the temperatures surpassed even that, as to snowfall, that was perhaps more hit or miss than Dec 1981 from what little I know of that month through reading about it here. We were also shaping our own forecasts to lower numbers than 2.8 as we got closer to forecast period. This is very similar to the approach usually taken to major storms, you're more likely to hear 10-15 inches a week in advance of a 20-30 inch type snowstorm, even if the forecasters around the table are saying to each other, looks like 20-30. That's just how this business operates, and raw rookies often figure the only way to get noticed is to say 30-50 inches and hope for 20-30. (this is more east coast U.S. oriented, sadly Britain does not get drawn out of the hat for these 20-30 inch storms very often, except locally).

You'll recall that the forecast including the CET numbers was presented mid-November but was actually made in stages from September to late October and we were just held up posting it by our own work schedules, so not to make an excuse but 2.8 CET seemed pretty bold to me when I typed it here for the first time perhaps 25 October, I would have to look back on my messenger file with Fred. We were probably pretty confident of that verifying by publication time and certainly by about 23 Nov when the cold spell began, then we started to think in terms of just how extreme might this get.

Nothing much has changed in the forecast philosophy as a result, it's not a case of "spending the balance" of some imaginary budget of cold and now we might adjust upwards, it's more of an emphasis on reminding readers that we used conservative numbers for December, the 3.0 for January was based on that kind of regime being interrupted by milder weather mid-month, and the "2.0 or lower" that is quoted for February means just that, we're predicting sub-2.0. From the range of 4-5 months, that's as extreme as we feel anyone making a forecast likely to be consulted by others should go, there's a tendency maybe to punt something really extreme when consequences are lower, but we want to advance our position too, and so we're probably holding some back in reserve on the colder forecasts. If you want the inside scoop, I'm pretty sure the current hemispheric pattern can support near-record cold anywhere it delivers the goods, everyone over this side is quite ga-ga over the look of arctic air masses coming our way soon, and there's currently a Siberian high looking very healthy.

The numbers we use are also compromises for different probabilistic outcomes. When I do an index value forecast, I tend to get a mix of signals and it's very much like an ensemble forecast, we could maybe think about citing a range of outcomes and giving probabilities. Just for fun, this is what that might look like for the period 20 Jan to end of February:

CET 4.0 or higher ... 10% likely (51)

CET 3.0-3.9 ...... 15% likely (22)

CET 2.0-2.9 ...... 20% likely (14)

CET 1.0-1.9 ...... 25% likely (

CET 0.0-0.9 ...... 20% likely (4)

CET below 0.0 .... 10% likely (1)

The numbers in brackets represent random chance without regard to current situation, in other words climatology.

Looking at those probability stats, our "over under" forecast for the period would be 1.7 C. Now for Jan to verify at 3.0, assuming it stays rather cold 5th-10th then turns quite mild for 5 days and leaves the CET near 4.0 at some point around the 19th (possibly a bit high), the likely mean temp for 20-31 Jan would be about 1.0, so that leaves Feb at 2.0 to get the figure of 1.7 for the period to verify.

If you just want a pub-chat personal favourite number, I would say 0.5 C is in the mix for any period after that mid-month mild spell and another negative CET month would not surprise me greatly, I agree with Fred it's more likely to be February than January but mostly because of that mid-month spike, I figure it would have to average -3.7 17th-31st after a weak warm spell lasting only to the 17th, to place a negative CET in the mix for January. Seems remotely possible but not likely, I figure January could collapse down to about 1 degree CET if the Siberian high proves to be strong, but 1.8 C is my current guess, the earlier 3.0 is looking plausible today but I have the feeling it will go from 3.5 to 4 running CET down to near 1.5-2.0 in the final third.

I just need to remind myself as well as readers that we are dealing with a developing theoretical approach that at best will be giving right side of normal outcomes 60 to 70 per cent of the time. Looking back at the period 1 Nov to mid-January, and if you would indulge my idea that my Nov CET forecast (where I mentioned mid-month pattern change) can be used in this overall assessment, the current performance level is close to 80% if it turns milder than normal 12th to 19th and extending onward would stay at 80% if it turns colder than normal 20th to end of February with no interruptions. That warns me that we are running a surplus on expected accuracy and yet I see no structural reasons to say various periods 20 Jan to 28 Feb will be above normal, although surely some days could be (but then wasn't there a pattern very similar to what's on the models for days 6-10 around 16-20 Jan 1947? I think they then had almost 50 consecutive below normal days once the severe cold arrived.)

[ZONE 51]

RJS as always a very good post (above), interesting and well explained.

[Weather-history]

Thanks for your replies, Roger and Blast.

[BLAST FROM THE PAST]

Wanted it in the model discussion thread...post deleted

Edited January 6, 2011 by BLAST FROM THE PAST

[Fozfoster]

Interesting up date guys, Mascarpone,ham,mushroom and pasta melt with red wine a good post thats my fill !!:drinks:Thanks.

EDIT.....forgot to say and garlic bread.

Edited January 6, 2011 by fozfoster

[geoffw]

where is the post which got deleted in the model thread by bftp?

[damianslaw]

Reading this forecast today suggests this winter is panning out in a fashion consistent with the forecast alebit the cold in Dec was much more severe and the dates for the change to milder weather in January about a week out, with this in mind the quick return to cold weather as suggested in the forecast for Feb will occur a week to ten days earlier around 20th Jan just as I anticipate will occur looking at the upstream signals today. Can't ignore how accurate the forecast has been so far and looks like being..

[Catacol]

A very clear disagreement is emerging on the net - BFTP and Roger's forecast along with TWO's official forecast and Gavin P over on TWO - - all going for renewed cold. GP has said early spring; Joe B'di has now gone for no return of cold for N, NW or central Europe. Fascinating. 2011 and we are no nearer being able to read the long range signals effectively. One group is going to end up disappointed.

I'm still rooting for cold. My own expertise, after 8 years of lurking and learning, encouraged me to lean towards sustained cold before the winter started - the general alignment of the jet over the last couple of years plus the clear tendency for high latitude blocking over the same period was my rationale, ie using observations of recent past rather than any particular ability to predict the future by using sun or stratospheric signals for example.

If it does end up mild then I would see GP as being most likley to be right - HP centred in the atlantic allowing warm air over the top and the strengthening sun to melt early morning frost. Dry. But we shall see: I want to see snow one more time this winter at least, so I hope that mid atlantic block can set up further north and perhaps pressure might rise over Scandy as the respected pundits over on TWO suggested might happen in Feb in their long term forecasts.

Either way BFTP and Roger - I gather you are expecting the models to turn in your favour in the next 5 - 7 days. Do you still think it is all on track, or are doubts increasing as to any renewed cold spell in late Jan/Feb?

Alastair

Edited January 9, 2011 by Catacol_Highlander

[Interitus]

So Bulls eye xmas double forty for the cold treble sixty for the 28th missed the dartboard on the rest well the dart went in then bounced out. Overall not a bad effort.

Well I'm not sure what kind of dartboard you play on, but I agree with your sentiments.

If the forecast stuck to the headlines and general trends it would appear to be very good - "Winter 2010/11 - Cold with spells of severe wintry blasts"

"There are indications of a very cold December with possible heavy snowfall events in central and northeast England, but also some snow from cold northwest flow patterns"

It's when the actual detail is examined that the forecast falls down with the coldest and snowiest weather forecast for the middle third of the month which coincided with the milder interlude.

Most glaringly, the CET was well out. This might seem over-critical bearing in mind it would've been cold anyway if correct, but being purely objective the 3.5 degree discrepancy would be the same as a 1.75C below average cold month - ie. like that forecast - become a 1.75C above average mild one - would that honestly be considered an accurate forecast?

For BFTP to say that the depth of cold - the coldest for 120 years - was not a surprise...well maybe it should've been.

Once again I do not want to appear over-critical because the headline for December is impressive - on the face of it. Details are always prone to ridicule as esteemed bodies such as the Met Office know too well, and excessive scrutiny as MurcieBoy is receiving on his thread.

Btw, looks like the January forecast has gone tits up.

Edited January 9, 2011 by Interitus

[Catacol]

Btw, looks like the January forecast has gone tits up.

Hmm - that seems a bit premature. The forecast pointed to cold to start - and for much of northern Britain that is exactly what has happened. For the south continental HP stopped the trough digging as far south as it might, but the pattern was picked up correctly. We are now approaching the energy peak and rising temps that were also specified in the forecast. That looks correct again to me.

The jury is out on the final third - recently BFTP has made comments suggesting that the models will come back around to cold from the NE soon... I am not so sure about that, but Jan 20 (date picked out in the forecast as the end of the mild spell) is still out over T+240 so I think we should wait before stating that anything has gone awry.

Generally the forecast's ability to pick out the general pattern weeks in advance has been very very good. February is looking under pressure though - some esteemed voices gonig for mild - GP for one.

Alastair

Edited January 9, 2011 by Catacol_Highlander

[Weather-history]

We both factor in the lunar phases with full moon, new moon, perigee and apogee having affects on LP and HP development.

See this is where I have a problem. Lunar cycles are fixed, hence why the ancients could predict eclipses. So if the forecast goes wrong then this can't be used as an excuse.

I asked Roger this a couple of years ago, if his forecast doesn't go to plan you can hardly use excuses such as Mercury didn't retrograde enough or Jupiter didn't get close to Spica as I thought it would. These are fixed, the orbits are fixed, the timings are fixed. Therefore random elements must have come into play and for me, there are far too random elements.

[BLAST FROM THE PAST]

See this is where I have a problem. Lunar cycles are fixed, hence why the ancients could predict eclipses. So if the forecast goes wrong then this can't be used as an excuse.

I asked Roger this a couple of years ago, if his forecast doesn't go to plan you can hardly use excuses such as Mercury didn't retrograde enough or Jupiter didn't get close to Spica as I thought it would. These are fixed, the orbits are fixed, the timings are fixed. Therefore random elements must have come into play and for me, there are far too random elements.

Kev

Lunar only plays a part in the forecasts. For example new moon, perigee coinciding will have a stronger peak energy signal than on their own. No there won't be excuses like you suggest.

I don't think Jan has gone tits up as Interitus has put it. Plenty of time for further developments yet. Yes very much against what GP has forecast...but to say there is no change to the forecast. Joe B has blown already with the call for the core of the cold to be South and East Europe.

Re Interitus mentioning the CET well its in the forecast to not scrutinise too harshly....we were conservative and just really shows the overall pattern with Dec being very cold to the norm, Jan not so much and Feb to be coldest. Indeed Feb is under pressure to be the coldest as Dec was so so cold....but if the pattern sets out as anticipated building in from Jan then it is going to be in 'extreme' category again.

The current breakdown is NOT of concern and indeed is well within the anticipated pattern....let's see what models come Wed/Thurs/Fri show for following week....I suspect change afoot on the horizon for latter third of Jan.

BFTP

Edited January 9, 2011 by BLAST FROM THE PAST

[Great Plum]

Interesting ECM run this evening with heights to our east apparent from. T168... In line with your thoughts Blast and Roger?

[Tim Bland]

IMO so far your forecast has been almost spot on. The only difference appears to be that the milder spell has occoured a bit earlier than forecast. If the rest of your forecast transpires to be correct then you have out-performed all forecasts I have read!