Model output discussion - cold spell approaching?

[Battleground Snow]

4 minutes ago, Battleground Snow said:

It's needs not to "pop out* out later into the run... If it's shallower it may track further south

It's does pop out later and is a little shallower, this allows a little more WAA into Greenland.. probably won't be enough this time, but a little step in the right direction.

 

[sebastiaan1973]

 

52 minutes ago, Cold Winter Night said:

That same page also shows the wind directions. The control is indeed mostly SW (225 degr), as is the majority of members. Just a handful that goes NE (45 degr), those must be the same ones that still show a minor grouping of ice days in De Bilt later on.

Indeed. Control shows a (south) westerly flow.

Model charts for Europe (Geopotential height, 500hPa) | ECMWF IFS HRES (10 days)

METEOLOGIX.COM

---

ECMWF IFS HRES (10 days) - Current forecast valid for 01/26/2021, 07:00am of parameter "Geopotential height, 500hPa", model chart for map "Europe"

 

[That ECM]

3 minutes ago, mushymanrob said:

Make of this what you will

 

Huge difference to the south not as big for the north. West for the south nw for the north. Now 2010 was much more from north east.

[DavidS]

4 minutes ago, mushymanrob said:

Make of this what you will

 

Looks similar to me. Trough not as deep

[Phil Blake]

4 minutes ago, mushymanrob said:

Make of this what you will

 

Subtle differences in the flow over the UK? Today is north going westerly over the UK. 2010 is north north east going westerly south of the UK? If I'm reading it right that is

[Mike Poole]

Big set of 12z runs tonight.  The short term snow question aside, we need to get a better handle on the longer term evolution and there is huge uncertainty about this.  I was concerned about some of yesterdays 12z runs with the pattern being too far west, but was not expecting to see the whole lot go up in smoke this morning.  We need it to come down again, fast!

[If Wishes made Weather]

Does anyone know if the models either add new data or use different calculations to produce the different frames in any particular run, or are the different frames simply a reiteration of the same calculations based upon the  outputs that produced the previous frames?  I suspect a blend of both, but would like to know as it would explain one or two things about how much predictive value any particular chart can have. Many thanks

[Kasim Awan]

2 hours ago, Paul said:

What you're saying is possible, but equally so is Ninman's view - it's all on a knife edge. I'm not sure what you're meaning when you say there's a consensus on the snowier outcome, as tbh I'm not seeing that. Some models are more bullish than others, but I'd say that the most likely outcome isn't for widespread lower level snow. 

I don't know if this is applicable in this instance, but a lot of people seem to be taking some of the snow accumulation maps at face value from models such as the Arpege, GFS etc. But the issue with these is that many of them are showing a water equivalent snow accumulation, based solely the precipitation rates and not taking melt or the moisture content of any snow into account.  Typically on those charts the water equivalent snow amount is multiplied by ten to get a depth, but that multiplication is what dry snow would bring. On this occasion in particular, even assuming everything was snow, it's not going to be dry, so immediately, even if all of it settled you could be looking at needing to perhaps halve the values shown on those maps. And on top of that, you also have to take into account the snow (and maybe other frozen/semi-frozen precip types) which fall but don't settle.

I'm not saying lower level snow isn't possible, I'm not even saying that the accumulations you're describing aren't possible, but equally they're not a certainty, and I think the uncertainty around all this is where the true consensus is right now. 

Yes, you effectively covered my point about the uncertainty as all the unorganized precip will do is prevent the ground environment below 200 metres (which is where there is marginality) from being conducive.

Given the projection of a well defined area of precipitation tomorrow, the outcome of some snow to lower levels producing a lowland area of snow maybe no greater than 5-25,000km^2 (up to 10% of the UK) with accumulations of 1-6cm in the favored stretch is fairly high in confidence. Of course, there are mechanisms which increase snow with altitude (orographic lift, lower temperatures, lower ground temperatures), and this means locally 20cm+ is possible for somewhere like Queensbury tomorrow. This is most favoured to occur across West Yorkshire, South Yorkshire, Notts, Derbs, into the North & East Midlands before moving towards the home counties. Even in this situation, there will be marginality on the north, south, west and eastern side of the snow area which will all result from slightly different thermodynamic mechanisms. North & east - lack of evap cooling. West -Fohen effect, lack of evap cooling. South - lack of cold air and low dew point injection from east. Defining which areas will see this is very uncertain and is down to nowcasting, however this does not take away or limit the main precip area evap cooling derived snow from verifying.

A strong consensus (70%) of weather models show this area verifying, the uncertainty is more to do with placement further south. The outcome of snow to lower levels in the Northern areas is not a certainty infact far from it however, the probabilistic forecasting method is always the way to go with consensus whilst leaving room for other possibilities which is what I've done, and you have covered here. Other outcomes include a more northerly track (some of Harmonie runs) and a more unorganized precipitation regime. However, I feel at this moment it would be unrealistic to go with 50/50 or even say "too close to call" for the precip further north at least, given the data for the triple point system which I am looking at. This does not discount a more unorganized precip area from occurring, just gives it a chance of say 25% whilst remaining aware that the higher confidence of snow for some areas (S Yorkshire, North & East Midlands) means imo a warning is valid from both an impact and likelihood perspective. Further south there is a risk of some snow too which of course will be bounded by the same spatial marginality mentioned above. Uncertainty is always the case with snow it's about doing the best one can with the data provided and forecasting knowledge to warn the public beforehand.

Edited January 13, 2021 by Kasim Awan

[northwestsnow]

Icon swings back to a stronger Greeny block...

[southbank]

Just now, northwestsnow said:

Icon swings back to a stronger Greeny block...

is it a greeny block or Atlantic ridge ------------ ok i get my coat 

hoping mood lift here with better charts after this morning kick

[Paul]

17 minutes ago, Ben Sainsbury said:

Big improvement on the UKV 12z compared to its UKV 9z run for tomorrow's snow event:

 

Keep an eye out for the 15z as we've made a change to our prec type algorithm on there which'll come into effect then (the prec type isn't a direct data feed from the UKV, it takes various data params from it and then our algorithm takes over to produce the output). The reason for the change is that in this type of situation we felt that the snow was a bit underdone in places, so we've put a correction in. 

For those expecting it to go nuts and put snow everywhere, that's not going to happen I'm afraid but there will likely be some more pink on there than there is right now (unless the raw data on the 15z has a massive change of direction of course!). 

Edited January 13, 2021 by Paul

[sebastiaan1973]

1 minute ago, northwestsnow said:

Icon swings back to a stronger Greeny block...

Don't wanna spoil the party Truly. At the same time, we see a area of low pressure developing at the right of the Greeny block. This system will develop and move a little bit to the southeast and expand and very difficult to move on. At the same time the cold eroses the Greeny Block. Perhaps (and I hope) I'm wrong.

 

[Battleground Snow]

GFS OP, has finally adjusted to the other models in terms of snow for Thursday

[Panayiotis]

Just a note before the 12z. That was one the quickest ‘turn-arounds’ across the models I’ve seen for a long while. From what looked like a decent cold spell (predicted by most models) has now turned into something uninspiring. Let’s see what translates over the next few days... 

Edited January 13, 2021 by Panayiotis

[If Wishes made Weather]

3 minutes ago, northwestsnow said:

Icon swings back to a stronger Greeny block...

I have noticed this sort of cyclical nature in the outputs.  Maybe something to do with simple diurnal variation in the input data. Probably suggestive of taking a sort of average view of outputs over a period of a couple of days rather than suggesting that any one chart is predictive of itself.

[sheikhy]

3 minutes ago, Battleground Snow said:

GFS OP, has finally adjusted to the other models in terms of snow for Thursday

Any chartss mate?can i only assume gfs p remains the same?!!

[Kasim Awan]

5 minutes ago, Paul said:

Keep an eye out for the 15z as we've made a change to our prec type algorithm on there which'll come into effect then (the prec type isn't a direct data feed from the UKV, it takes various data params from it and then our algorithm takes over to produce the output). The reason for the change is that in this type of situation we felt that the snow was a bit underdone in places, so we've put a correction in. 

For those expecting it to go nuts and put snow everywhere, that's not going to happen I'm afraid but there will likely be some more pink on there than there is right now (unless the raw data on the 15z goes the other way of course!). 

The lying snow maps (which are directly from the model) from the 12z are a good indicator of what's expected right now.

 

 

This is good, however I don't think that level of "patchiness" is particularly realistic given the intensity of the precipitation we're looking at. Some smoothing I think wouldn't go amiss

[Battleground Snow]

Just now, sheikhy said:

Any chartss mate?can i only assume gfs p remains the same?!!

Sorry yeah here, was just trying to get off a work zoom call.

Para  top v OP bottom. Like Paul says don't take them too literally

[That ECM]

From here on in is FI.

[sheikhy]

3 minutes ago, Battleground Snow said:

Sorry yeah here, was just trying to get off a work zoom call.

Para  top v OP bottom. Like Paul says don't take them too literally

Defo upgrade in snow accumulations!!!!the para has been really consistant with this!!mids gets around 5cms from this!!

[Ninman]

10 minutes ago, Paul said:

Keep an eye out for the 15z as we've made a change to our prec type algorithm on there which'll come into effect then (the prec type isn't a direct data feed from the UKV, it takes various data params from it and then our algorithm takes over to produce the output). The reason for the change is that in this type of situation we felt that the snow was a bit underdone in places, so we've put a correction in. 

For those expecting it to go nuts and put snow everywhere, that's not going to happen I'm afraid but there will likely be some more pink on there than there is right now (unless the raw data on the 15z has a massive change of direction of course!). 

The lying snow maps (which are directly from the model) from the 12z are a good indicator of what's expected right now.

 

 

Looks more like what I am expecting. More for northern and eastern areas of the UK with some lying snow in more central areas with elevation. Certainly don't expect any lying snow where I live. It will be interesting to see how this pans out against the updated algorithm!

[chris55]

25 minutes ago, mushymanrob said:

Make of this what you will

 

Similar. However we have the low pressure to the south of the high this time and slightly less cold air incoming from the north/north east (at least initially) but this is once again suggesting a high risk of boundary frontal snowfall somewhere in the UK. Some places could get buried while others see rain.

[Kasim Awan]

Just now, Ninman said:

Looks more like what I am expecting. More for northern and eastern areas of the UK with some lying snow in more central areas with elevation. Certainly don't expect any lying snow where I live. It will be interesting to see how this pans out against the updated algorithm!

You are very well placed in Nottingham, again this post is not guaranteeing you'll see snow, but based on due care analysis of the model output it's fairly likely.

[nick sussex]

There’s a lot going on even within the day 4 range .

The cold pool to the east  is reluctant to give way so the next shortwave is struggling to clear into the North Sea .

 

[Tim Bland]

Aperge going for snow on sat for similar areas as tomorrow. Accumulation chart is all ‘falling snow’ between. Ie and then 

ps * UK view not available yet before I get lynched...*