Model Output Discussion - 1st February Onwards 12z--->

[loafer]

Pixel, the models are NOT fed with historical data. They are pure mathematical calculations from now.

At the risk of stating the obvious, the algorithms and calculations used in the model to predict the future will have been set based upon historical data.

[bluearmy]

At the risk of stating the obvious, the algorithms and calculations used in the model to predict the future will have been set based upon historical data.

I assumed the algs and calcs are based on physics?

[johnholmes]

I assumed the algs and calcs are based on physics?

 So am I although Fergie seems to suggest something else is also run at the same time.

[TEITS]

Following on from my post last night but even at this late stage we are seeing scatter at +96.

 

http://max.nwstatic.co.uk/ensimages/ens.20150204/00/t850Cambridgeshire.png

 

So some members are still indicating a very brief colder N,ly on the right flank of the HP. This is only being indicated though for the E as you can see on the ensembles for Wales.

 

http://max.nwstatic.co.uk/ensimages/ens.20150204/00/t850Abertawe~-~Swansea.png

 

Now I want to make something clear though. These much colder N,lys are looking more likely to remain to the E of the UK. Even if those colder ensembles did prove to be correct then any cold N,ly would be very brief indeed.

Edited February 4, 2015 by TEITS

[nick sussex]

Morning all

 

Not much change overnight although some differences still evident regarding the weekend. The ECM remains the coldest and brings the colder air closer to the east and se as the high sets up to the west.There is still a cluster of 5 GEFS that bring colder air back into the UK.

 

Its still a low probability this morning especially as the GFS and UKMO operational runs are against this idea. Before then a window of convection off the North Sea as some colder uppers get pulled in from the east/ne. The Euro 4 is quicker to bring in less colder uppers whilst the GFS keeps the -10 850 for longer. These situations are often down to nowcasting but something to keep an eye on Thursday into early Friday.

 

The ECM at T168 and T192hrs looks colder again, even though the uppers on the face of it look milder the flow moves round to the se and will  bring in some colder conditions from mainland Europe, this supported by the GFS.

[IDO]

The ECM diagnostic for the next 8 days seems right; HP easing in from the west from today and in situ for around 5 days before migrating east:

 

 

 

All the models have this general scenario with subtle differences to specifics. I would expect eight days of no precipitation for most (NW the closest to the Atlantic flow) with temps during the period average to just below. The core of the high around or to the west of Ireland.

 

From then there are some phasing differences between the model as there is a day or two of a slack mishmash before the next wedge of HP pushes in from the SW. GFS op as it's will brings in the Atlantic for a more changeable flow, the GEM tries to build heights in the Atlantic (normal bias) and ECM has a N/S split by D10. Although the GFS op is not an outlier it only has a few members support and it's mean rebuilds the UK high:

 

D10   D16: 

 

So overall little change just slight uncertainty to the transition from the upcoming HP to the next at around D8-10. There does not appear to be much developments in the NH with a static Russian/Siberian high, the Pacific ridge and the UK high, with the PV, in its seemingly slow wind down mode, happy to roll around in the gaps leftover. Without any forcing to the flow this period may take some time to disrupt into something more interesting? The GEFS MJO sits in phase 7 for the next 2 weeks so that does not really suggest a change is upcoming:

 

 

Only a couple of interesting GEFS from D12-15 so not really a background signal for any change at the moment.

[knocker]

 So am I although Fergie seems to suggest something else is also run at the same time.

 

Yes Fergie did suggest that but surely there is no dispute that they are based on physical principles.

 

A brief history and A Wiki article.

 

The origins of computer weather prediction and climate modeling

 

http://www.elsevierscitech.com/emails/physics/climate/the_origins_of_computer_weather_prediction.pdf

 

http://en.wikipedia.org/wiki/Numerical_weather_prediction

[snowking]

Possibly worth noting a couple of things with regards to the MJO.

First of all there are a number of people who believe the current RMM forecasts may not bare truth (and people who I would very much trust on these matters) and that in fact we will effectively end up back in the circle of doom *cue horror movie music*

However, it is worth bearing in mind that the last MJO p7 wave produced a cold spell shown via NWP in its mid-range at a time when the overal theme had been zonal from ensemble suites. Combine this with a lag period for the phasing to correlate with longwave patterns and in fact there is no reason to suspect that the current cold spell will be the last of this season

Conversely there is no guarantee of a return to cold of course, but if you would care to take a look through the archives you will see that MLB over the UK is not a bad holding pattern, especially with a dynamic upper vortex

SK

Edited February 4, 2015 by snowking

[Midlands Ice Age]

BBC (MO) are still reporting the onset of a bitterly cold easterly on thursday and lasting through the weekend on the Nat weather at 08:20 (and repeated at 08:50).

They stressed this signal last night, as I reported, and Knocker posted up a GFS chart showig a rapid extension eastward of the high off northern Ireland into Poland and the Baltic Sea.

This chart was again shown this morning by the gorgeous one (Carol) and she was indicating that it would last all weekend and the high would not be toppling quickly and maintaining the 'even colder' air into the southern/central porton of the UK, whilst stayig cold to the north.

They seem to have picked up this signal and it is not just a one dayer for thursday. as many people on here have been saying

PS This morning at 8:30 my local BBC forecast ( for the midlands) did not show this scenario with mildish temps and a light northerly. They have changed overnight.

Could it be that we have missed this extension to the cold whilst looking too much for the next cold spell in the 10 day period?

MIA

Edited February 4, 2015 by Midlands Ice Age

[knocker]

The GFS isn't running with the easterly fetch over the weekend. It does suggest an NE/E but within the circulation of the HP

http://www.atmos.albany.edu/student/abentley/realtime/atlantic.php

[Summer Sun]

Not a great deal to say this morning really the ECM ens sums it up pretty well with the high looking very stubborn to shift

 

Saturday 7th

 

 

Saturday 14th

 

[Dr. Astro]

BBC (MO) are still reporting the onset of a bitterly cold easterly on thursday and lasting through the weekend on the Nat weather at 08:20 (and repeated at 08:50).

They stressed this signal last night, as I reported, and Knocker posted up a GFS chart showig a rapid extension eastward of the high off northern Ireland into Poland and the Baltic Sea.

This chart was again shown this morning by the gorgeous one (Carol) and she was indicating that it would last all weekend and the high would not be toppling quickly and maintaining the 'even colder' air into the southern/central porton of the UK, whilst stayig cold to the north.

They seem to have picked up this signal and it is not just a one dayer for thursday. as many people on here have been saying

PS This morning at 8:30 my local BBC forecast ( for the midlands) did not show this scenario with mildish temps and a light northerly. They have changed overnight.

Could it be that we have missed this extension to the cold whilst looking too much for the next cold spell in the 10 day period?

MIA

 

Hi MIA,

 

You are not far wrong. I think people are just looking for something more widespread, but for the Southern contingent of the UK, the following charts highlight a pretty good snow risk tomorrow, not 10 days into the future, but tomorrow. I'm surprised this place isn't more busy.

 

Strictly for the South then...

 

We have the 528DAM line:

 

 

850's are good, some places tipping the famous -10C mark:

 

 

The flow is good from the ENE:

 

 

The min temps are good:

 

 

The precip seems to be light to moderate, but appears to be in streamer form across the South for several hours:

 

 

The precip type appears to be snow:

 

 

The "experimental" snow depths highlights the risk:

 

 

The Isotherm will be no problem:

 

 

And the UK Snow risk clearly highlights the opportunity:

 

 

All in a T24-36 hour window.

 

Albeit the FAX has it a little more marginal:

 

ECM singing from the same hymn sheet:

 

As is the GEM:

 

 

I'd day some in the South are going to see some snow tomorrow.

 

After that, it looks very dry and benign until around the 13th February.

 

Enjoy your snow, whoever gets it tomorrow.

Edited February 4, 2015 by Dr. Astro

[Gibby]

HERE IS MY LATEST ANALYSIS USING DATA SUPPLIED BY THE NWP OUTPUT COVERING 5 OF THE WORLDS MOST POWERFUL WEATHER COMPUTERS

THE CURRENT GENERAL SITUATION A cold Northerly flow will gradually veer to the NE tomorrow as a High pressure centre edges towards NW Britain tomorrow. A weak front over Scotland will decay as it moves slowly South.

MODELS-2 WEEK HEADLINE Becoming slowly less cold with a lot of dry and bright weather away from some rain in the North later. Some frosts at night still chiefly in the South and at first.

THE JET STREAM ENSEMBLE FORECAST The Jet Stream Forecast shows the cold trough over Europe lifting out over the coming days as the flow moves North to lie in a West to East flow pattern close to Scotland later next week, and continuing through Week 2 on the Northern periphery of High pressure close to Southern England and Northern France

GFS OPERATIONAL The GFS operational today splits into two halves with Week 1 illustrating dry and settled weather under a large UK based anticyclone with cold weather by night with sharp frosts especially in the South but with areas of cloud floating around too. Under this cloud temperatures should recover somewhat by day and night too should it prevail. In Week 2 the pattern changes to a more mixed scenario as the High drifts away to the East and SE however the week will still involve some High pressure based settled weather but also periods of cloudy and damp weather as troughs cross over from the West with a little rain in places.

THE GFS CONTROL  The GFS control run is very similar in style as the operational shows High pressure more resistant against leaving the South through Week 2 while the North sees the more changeable weather under a stronger Westerly flow.

THE GFS CLUSTERS The GFS Clusters show a 75% likelihood of High pressure over or to the SW of the UK in 14 days with benign conditions and temperatures just a little below average but with frosts by night in the South. The remaining 25% offer something rather more unsettled with stronger winds and rain at times in a NW flow.

UKMO UKMO this morning shows High pressure in total control of the UK weather through the weekend and start to next week as it drifts into Ireland and then the South of the UK at the end of the run. Typical Winter Anticyclonic weather is likely with some bright and sunny weather mixed with cloudier sells with sharp night frosts where skies stay clear of cloud overnight.

THE FAX CHARTS The Fax Charts show good support for the raw data with a few weak troughs drifting South increasing cloud cover and giving rise to a little rain, sleet or wintry showers in the SE for a time before all areas become settled and dry later.

GEM GEM takes the GFS route today of bringing the High across the South of the UK next week and away to the ESE. This sets up a fall of pressure with a more marked freshening of Atlantic winds driving fronts across the North and West in particular with rain at times before the end of next week.

NAVGEM NAVGEM holds High pressure across Southern Britain throughout next week with only the North seeing a Westerly breeze lifting temperatures near to normal here while the South sees occasional frosts still where skies clear with temperatures near or just a little below average.

ECM ECM this morning shows total support for the same general theme of High pressure moving across the UK at the weekend and into the start of next week, gradually relaxing the centre to the South and SE later and feeding ever milder air down from the North later on a strengthening Westerly breeze. Only the North would see rain later though as pressure holds High close to the South.

ECM 10 DAY MEAN The 10 Day Mean this morning indicates that High pressure will most likely lie close to the South of the UK next week and beyond. Conditions will slowly recover to fairly average temperatures and rather cloudy weather overall with patchy rain at times in the North.

NOTABLE TREND CHANGES FROM PREVIOUS RUNS The trend towards somewhat less cold but Anticyclonic conditions looks fully likely now with occasional rain more of a risk with time across the North.

MY THOUGHTS The end of the cold spell looks guaranteed now at the weekend as High pressure off the Atlantic makes inroads into the UK from the west by the weekend and start to next week. There looks to be a good infill of cloud with time as weak fronts cross around the Northern flank of the High and down across the UK, lifting temperatures both by day and night and making for some rather benign and uninteresting weather. Still as the High pressure moves over the UK some frosts are still possible where skies stay clear this most likely in the South where temperatures could stay just a little shy of average in the flat calm. Then we have to look to how the pattern develops beyond that through the second week with the general consesnsus shown by the models today of the High slipping to the SE but staying close to the South. This will feed stronger Westerly winds across the North especially with fronts close enough by to give rise to outbreaks of rain at times in average temperatures. However, there remains good support for High pressure to remain close to the South with light winds and fair weather with just variances in cloud amounts making conditions difficult to predict then but it should stay dry. So in summary after the next few days it should become less cold rather than mild with a long period of largely benign weather with light winds, variable cloud cover and a little sunshine and patchy night frosts, these most likely over the South next week. Then we may see some slightly less settled weather encroach into Northern areas later as Westerly winds freshen.

Issued at 08:30 Wednesday February 4th 2015

[Nouska]

 So am I although Fergie seems to suggest something else is also run at the same time.

 

NOAA are doing something similar with analogue calibrations.

 

 

http://www.esrl.noaa.gov/psd/forecasts/reforecast2/probabilities/index.html

 

A bit more meat on the bones of the GS5 and detail about the hindcasting to bias correct.

 

http://onlinelibrary.wiley.com/doi/10.1002/qj.2396/full

 

 

The hindcast set is used to bias correct all forecast members. Each week we complete a ‘hindcast week’ consisting of 14 years with three members per start date per year. The hindcasts are initialized from the following set of start dates: (1, 9, 17, and 25 in each calendar month). The same model is used for both hindcast and forecast members. Spread between members initialized on the same date is achieved through the use of a stochastic physics scheme. The number of years in the operational hindcast is sufficient for establishing the model climatology for bias correction (Arribas et al., 2011).

[Duane S.]

Very difficult to see true HLB forming in the next 2 weeks with the current profile (both trop and strat) but whilst a UK high can have the downside of being quite a persistent entity (in terms of meh dull benign etc) as the jet rides over the top pumping it up, once conditions allow again, it can very quickly turn it our favour from an MLB location to a more favourable one.

Just for fun though, before the 06z removes them altogether, still a few ''rogue' runs being produced by GEFS to make us drool...

Edited February 4, 2015 by s4lancia

[mushymanrob]

ive been banging on about the coming high being a 'dirty high'... i still think that, but, it does depend upon the exact position/centre of the high. previous runs have shown a lengthy enough draw off the atlantic in its circulation = moist air... of course IF the high taps into the cold, dry, continental air it would be clearer with pleasant wintry sun by day, sharp overnight frosts by night.

over the period of high pressure dominance we are likely to see a mix of both options, with location playing a part in who/where gets how much of which.

the anomaly charts suggest a sinking high, with by day 10/11 a mean upper flow from the west. so gradually getting less cold/average and a cloudier option more likely?

 

[Trom]

 So am I although Fergie seems to suggest something else is also run at the same time.

They must test the algorithms and adapt them based on historical data to prove the model one would assume.  If the model can't predict whats happened in the past then it would be no use in predicting the future.

 

So I would imagine some back testing using historic data must be carried out as a proof of the algorithms.

 

Edit: sorry mods I see this is correct from other posters - please feel free to delete.

Edited February 4, 2015 by Trom

[Day 10]

Very difficult to see true HLB forming in the next 2 weeks with the current profile (both trop and strat) but whilst a UK high can have the downside of being quite a persistent entity (in terms of meh dull benign etc) as the jet rides over the top pumping it up, once conditions allow again, it can very quickly turn it our favour from an MLB location to a more favourable one.

Just for fun though, before the 06z removes them altogether, still a few ''rogue' runs being produced by GEFS to make us drool...

image.jpg

In 2005 this:

 

 

Turned into this:

 

 

So yes agreed still time for things to flip.

[Shunter]

Looking for crumbs of comfort during the lull in the UK Winter proceedings..............  

 

Latest model discussion preferences from NOAA seem to favour yesterdays 12 ECENS over todays 00GEFS ENS ( and the 06z GFS OP looks similar too ).

 

...SECONDARY SYSTEM AFFECTING UPPER NEW ENGLAND BY FRIDAY...
...RESULTANT MID-LEVEL VORTEX SETTLING INTO QUEBEC BY THE
WEEKEND...

FINAL PREFERENCE: 12Z ECMWF ENSEMBLE MEAN
FORECAST CONFIDENCE: SLIGHTLY BELOW AVERAGE

AS THE INITIAL UPPER TROF AMPLIFIES OVER NEW ENGLAND...A SECONDARY
SHORTWAVE IS FORECAST TO DIVE DOWN FROM SOUTH OF HUDSON BAY TOWARD
THE INTERIOR NORTHEAST BY FRIDAY. WHILE MODEL SPREAD IS NOT
TERRIBLY SIGNIFICANT WITH THE INITIAL IMPULSE APPROACHING NEW
ENGLAND...THERE IS GREATER UNCERTAINTY WITH THE PLACEMENT OF A
BROAD VORTEX DIPPING DOWN ACROSS QUEBEC. THE 00Z CMC/UKMET DID
BOTH MOVE NORTHWARD AWAY FROM THEIR PREVIOUS FORECASTS WHICH DID
NOT MATCH ANY OTHER PIECE OF GUIDANCE. IT LOOKS LIKE MOST OF THE
DETERMINISTIC SOLUTIONS...AT LEAST AMPLITUDE-WISE...ARE MOVING
TOWARD THE 12Z ECMWF ENSEMBLE MEAN. THE FLATTER 00Z GEFS/21Z SREF
MEANS ARE NOT SUPPORTED AS MUCH AT THIS TIME. WILL RECOMMEND THE
12Z ECMWF ENSEMBLE MEAN HERE WHILE SLIGHTLY RAISING CONFIDENCE.

 

IIRC the 12 ECENS held the MLB further North West of the UK which would extend any colder Eastern influence maybe a little longer    
 

Edited February 4, 2015 by Shunter

[legritter]

They must test the algorithms and adapt them based on historical data to prove the model one would assume.  If the model can't predict whats happened in the past then it would be no use in predicting the future.

 

So I would imagine some back testing using historic data must be carried out as a proof of the algorithms.

 

Edit: sorry mods I see this is correct from other posters - please feel free to delete.

why delete ,an interesting post ,would be great if we had upper air profiles from past history ,how strong was the P Vortex ,MJO  Strat etc ,of course IF we did we would be further down the line with long range forecasting wouldn,t we ?,Well we still have time apart from present for some High level blocking ,Let the hunt continue .  

[492]

http://mp1.met.psu.edu/~fxg1/ECMWF_0z/hgtcomp.html

Latest 8-10 day 500mb charts continue to show low heights in the eastern Med and Azores preventing the Azores high from assuming its normal position. ECM has the axis of the ridge extending from mid Atlantic to Russia slightly further north with respect to the UK compared to yesterday. GFS has the upper high cell over Scandinavia further to the NE than yesterday. The upper ridge and associated surface high will have difficulty sinking too far south while low heights continue to be modelled from the Med to the Azores.  
 

[johnholmes]

Interesting inputs above about testing forecast data against historic data. I wonder if someone, fergie, Jo might be able to tell us all if this part of the outputs is actually used in the synoptic models we see/ Is this used in a forecast from UK Met? really interested to know just what is done with this additional output-help please?

[SE Blizzards]

Not much to cheer in the charts for snow lovers at present, so just thought i'd post the latest CFS run for March just for fun, and also try to stop the tumbleweed from blowing across the thread.

 

2013 revisited?

 

 

 

 

 

We live in hope.

 

 

 

 

[Cheshire Freeze]

Not much to cheer in the charts for snow lovers at present, so just thought i'd post the latest CFS run for March just for fun, and also try to stop the tumbleweed from blowing across the thread.

 

2013 revisited?

 

 

 

 

 

We live in hope.

Yep getting a 2013 vibe. Been showing this scenario very regularly recently.

[Mucka]

Based purely on potential and interest in the output then the upcoming period looks to be the most dire outlook of the winter for cold fans.

A long lasting UK high slowly sinking South with the Azores high then repeatedly ridging in. That is about as  desperate as it gets for long term prospects but it is marginally better than being stuck in the fast zonal setup of last winter.

 

GFS ensembles have positive 850's from the 8th right out to the final third of Feb and likely far beyond if they went that far out. It really has been a disappointing winter given the possibilities although some have done okay which will give them a different perception.

Here in the NW I have had a couple of hours of wet snow and slush on the ground that was gone a few hours later - that was the high point of the winter.

Given this winter was still a big improvement on last winter it shows how dire our winter weather has been since we had those wonderful cold snowy interludes a few years ago.

At least Southerners had that late winter Easterly I guess.

 

Normally we can at least see the hint of a signal for a small chance of something colder in deep FI but I can't even find that so I am writing off the next 10 days for any snow chances in my region though the SE have remote chance before the high settles in proper.

In reality though it looks as though we can write off up to the final 3rd of Feb.

 

This can be moved to model moans or whatever I guess and at least it offers a nice break from model watching and snow fixation.

Enjoy the dry weather, hope we get some sunshine.

Edited February 4, 2015 by Mucka