SomeLikeItHot

Hmm the Met office have my home forecast for 12pm down as sleet, despite it being 1 deg and very low dewpoint. Would have thought whatever falls is unlikely to be anything but snow particularly at 90m ASL.

medium flakes floating down lightly outside my window in the city of London. Never going to settle but pretty to watch.

Seems to have shown the 0Z which ended in true gfs style with a SWesterly, rather than the pub which he referred to which didn't.

Or more likely there has been little model output for hours and the current runs are just starting to roll out to the interesting part ie where they may differ from the prior runs at say +120 hours.

Fergie mentioned the northerly solution was in the met office ensembles, yesterday I but a minority solution. I made a comment yesterday that this type of single forecast they issue is fairly useless in many cases as its either an average of two very different outcomes - and averages can be less informative than individual solutions in a bi-modal situation. Looking through models and ensembles today, I can at around 10-15 days either a northerly, the atlantic getting in, or continued high pressure on a few solutions. Giving the rough weightings of these three solutions would be more useful information in my opinion than just some single estimate with words about great uncertainty.

The thing to note that is that that chart is for London, the equivalent for say Aberdeen, sees many more cold members near the the Op run, (while still being on the cold side of them itself). So basically what we are seeing is a moderate confidence of scotland getting a cold northerly, with decreasing chances as we go south.

It will be its also interesting to note that it keeps the high in place just north of scotland for basically the entire Hi res part of the run then shifts it relatively rapidly as soon as it hits the low res.

GFS 12Z pushes the higher further north, compared with 6Z, but it never quite makes Iceland. Will it reach Greenland on later runs? 12Z 6Z

Depends when he meant. Early tomorrow the met are forecasting light rain showers over some parts before the really cold air moves in. Met office UK Forecast for Wed. Wednesday: Cloudier and colder than Tuesday, with the best of any brightness generally towards the west. Patchy rain and occasional hill snow dying out during the day.

Ok Sorry this is bizzare I was seeing a completely different chart. Sorry.

But why are you posting week old model output for this friday? - Edit: PLease ignore no idea what happened but was seeing a completely different chart first time I saw this.

Typically I find the short range forecasts are great, of course there are revisions but generally small ones. My approach is to use the Meto office for info over the next five days, but look at the models and members input for the 5-10 day period where the Met office is more vague and we are looking more for possibilities. I do however wonder at what the purpose and value is for the longer Met Office text forecasts? The first part covering 6-10 days is obviously fairly reasonable outline of what is likely to come but as it goes further out past 15 days it seems that it rarely conveys much information. Typically METO updates says "slightly above normal" or "slightly below normal", near normal etc, and it seems that this rarely going to verify as it is more likely to be either say mild SW influence or cold Easterly influence not some slightly below or above outcome, except briefly in transition. To me it would seem informative to be giving a probabilistic estimate for some of these scenarios? Is there a reason for not doing such - public perception etc? Giving two alternative outcomes may sound like hedging your bets, but its actually more verifiable than the inconclusive statements where two very different signals combine to give something that is a very washed out signal.

Probably too early to say, but some of the models if they verified would likely bring some snow later next week in some places. All could change with a shift north or south though so highly speculative.

More generally though ensembles are great for showing a solutions sensitivity to initial conditions, but ultimately they are all produced by they same model. If that model has a bias, all ensemble members will have the same bias.

Huh? If you are not ready why did you do it? Or didn't you , and you reason for not being ready was excellent agreement?

Belting down here in Billericay again now. 8cms at 6:30 am I am expecting it will get to 10cms soon if this continues for long.

8cms of snow out my way, and snowing currently at a reasonable rate

just over 7cms here and still pelting down.

About 5-6 cms here in billericay. Wet underneath and slippery as a result - slipped over taking a measurement. Still fairly heavy.

rain turned to sleet about 4pm turned to snow about 5:30pm coeing down fairly hard since then about 2 cms accumulation so far in Billericay. My gritted road seems to be turning white.

I think the low centre moving 50 miles is what he is talking about this can change what areas do and don't get snow/rain by much further. Obviously there is much uncertainty, the Met Office has been continually changing the forecast at my location over the last 12 hours from mostly sleet, to mostly heavy snow, to sleet and mostly light snow, and back again.

Not quite sure what you mean, perhaps I have missed some discussion that has occurred here previously. I agree its a good signal, but is the statement wrong? The air pool we are drawing from is warmer than usual, and although we can't know the outcome exactly it seems likely to be less cold than we would normally expect from an easterly flow.

And continues more or less to the end of the 6z run. Shame the air is so warm over East Europe otherwise we would be in the freezer, although it is cooling as it moves across. However I will believe this set up when the ECM shows show agreement.

GFS 18z very different and hopefully wrong. Really pushes through the atlantic, and although it build heights in scandi faster, its squeezing them out at +100hrs

Given that fairly major model disagreement at this time, eg the GFS 6Z shows this While the ECM shows And the ECM For the same time frame. I would think its too early to be worrying about exact snow location difference between two GFS runs. Both will almost certainly be wrong come the time, and indeed we should expect some change to the ECM which would at least change the likely location of any snow in this time frame.