virtualsphere

Excellent read as always Tony, many thanks! Think you may mean Nov-Dec 2010 for the fortnight of lying snow rather than 2012?

Agreed - I've thoroughly enjoyed the cold spell and think if it had occurred in 2008 or earlier it would have been rated very highly indeed. Having said that, although I have done well here, it has been fairly marginal in locations close by, especially those near to the coast. We had some decent snowfall last Friday, the first snow since December '10 here, and some brief flurries on Tuesday, but for me the unexpected persistent snow which fell nearly all day on Wednesday was the best part. We've now had a week of lying snow, and depths of nearly 5 inches in total, although as others have mentioned the temperatures haven't been that remarkable. For my location I'd rate this spell the third best since 1996, behind Dec '09 - Jan '10 (2nd) and Nov - Dec '10 (1st). On both of those occasions we ended up with 6 inches of snow or more, and in Dec '10 overnight temperatures fell below -13C which is very unusual for this area. For snow, the current spell may actually have matched or surpassed what we experienced here in 1995-6 and 1991, but I wasn't so interested in the weather then so can't be too certain. It's certainly been better than any cold snap between 1997 and 2008, almost all of which were one or two day events.

What a day - basically snowed all day here. Completely unexpected - was just expecting a few light showers if anything. Nearly had as much as last Friday.

Light snow here - surprised as I wasn't expecting anything this far west tonight. If it keeps up we might end up with a fresh covering come the morning!

NW Extra radar now enticingly showing the precip almost touching my postcode. I'm expecting it to stall about a mile to the east of my house, otherwise I'll have to eat my words about the snowy spell being over for my location.

Thanks for posting this - certainly illustrates that we're overdue a cold one (sub 2C). Surprising period in the early 1700s which only just broke 2C a few times though.

Quite an unusual FAX for T+120, not just the strength of the atlantic low, which is impressive in itself, but also the only strong low on the chart! Zonal... but not as we know it! So where do we go from here? GFS 12Z has the low spinning in situ for a while, without filling much, which means a few windy days especially for western areas. ECM agrees on a dartboard low: The UKMO keeps the deep low too, but brings the centre a little closer to the UK meaning a potentially windy spell for all on Sunday - Monday if this comes close to verification: Quite a change within a week. GFS ensembles suggesting a pressure rise after the low passes through although a lot of spread indicating uncertainty. The spread for 27th - 29th probably suggestive that GFS hasn't got to grips with the depth / track of the low yet either: Hold onto your hats this weekend! We've had a lot of rain over the last year but it feels like it's been a while since we had a notable windy spell of weather.

Quite a bit of snow here today but sadly nothing sticking - some of the snow we had on Friday has melted away as well. I've a feeling this might be the beginning of the end for this cold spell here.

Yes, the arc of rain currently over the tip of Cornwall is forecast to spread northwards and develop over Wales as the night and early morning goes on. For the current time (21:00) NAE 12Z has the precipitation accumulation in possibly a slightly more accurate location comparing to the radar than the 06Z with the precip showing closer to the coasts of the SW UK and Ireland, although the differences are slight.

. ...and I've just looked at the NMM and there are big differences here too between 06Z and 12Z. So not just "will it rain or snow?" but more "will it do anything at my location ... and if so, what?" Possibly those in SE Wales most likely to see some snow given both runs of both models show precipitation in that area, although still time for it to change. Interestingly, both NAE and NMM are going for snow rather than rain away from the coast.

Good post Jackone. The rain / snowfall NAE charts clearly show the differences that a shift in the position of the low can have on precipitation, and if it has changed that much between the 06Z and 12Z runs then there may well still be further shifts before the actual event. http://expert-images.weatheronline.co.uk/daten/proficharts/en/nae/2013/01/21/basis06/ukuk/prty/13012212_2106.gif http://expert-images.weatheronline.co.uk/daten/proficharts/en/nae/2013/01/21/basis12/ukuk/prty/13012212_2112.gif

Drove to Carmarthen this morning - quite a dramatic difference to home near Cross Hands. A good few inches here, but just approaching Carmarthen the fields quickly turned greener and there was very little on the ground. My parents reported heavy rain in Carmarthen at 2am - think we must have had snow all night as I heard no rain here. There's often a difference between Carmarthen and here, but never seen it as stark as today. On some occasions it has actually been worse there (the late 2005 northerly event for example) and I've driven along the A48 to find I can't get up the hill into town!

Yes, still a good few runs before Friday am and plenty of potential to shift. Wouldn't want to count anywhere out just yet.

Here's the shift west on NAE... Edit.. Benb beat me to it

I'd agree looking at the GFS chart, although would suggest the NAE does (in a very simplistic way) due to the different snow depths shown in the SE, mid and north Wales which, assuming a S / SSE flow would lead to areas sheltered by the valleys and mountains getting less. Not for a moment expecting these to be completely accurate though.

Snowfall accumulations up to midday Friday from NAE and GFS: Take these with a pinch of salt given the number of factors involved in predicting snowfall but it doesn't suggest there will be any significant break-point. The big issue at this stage is how far west will the snow extend...

Light snow here just outside Cross Hands, now starting to settle in places. Not looked much at the models or forecasts since last night and seems like an upgrade in that time. NAE now going for snow much further west on Friday and closer to the coasts, NMM also has snow further SW on the 12Z to the 06Z which looks better for Carms area.

This will be a slightly different type of post, so mods feel free to move if necessary. In the context of an evolving stratospheric pattern, some fascinating teleconnections and long range forecast work by members of the team, I thought it would be an interesting idea to see how well the models perform over an extended period by looking at a variety of charts for the same timeframe across different runs. With today (Monday 14th) being indicated some time ago as a possible date for the effects of the SSW to start to be felt in the troposphere*, I started saving charts a couple of weeks ago on a daily basis for 12:00pm today to see how well they would verify... Firstly, just a word of warning to note that I'll be looking at the weather over the UK when looking at the charts. The rest of the northern hemisphere may well be a reasonable match but if the UK synoptics are different then I would guess the majority would consider the chart to be of little use. It isn't a scientific analysis - just a simple look at the charts - but I'm also posting them here to save them as an indication of how well the models may perform as a SSW evolves and also leading up to a possible snow event. All charts are from the 12Z runs. The first chart is today's GFS T+0, which hopefully represents the actual synoptics at 12:00 today: Now the first model predictions, saved from Friday 4th - T+240. ECM: GFS: At this stage there is actually remarkable agreement between ECM and GFS considering the timeframe, although as you can see with high pressure centred over the UK they are both wide of the mark - in fact, as you can see in the T+0 chart above, low pressure was actually centred in a similar place to the high shown in these T+240 charts. On to Saturday 5th, and we're now at T+216: ECM: GFS: First thing to note is the stark differences between these charts, which illustrate that the models are still trying to find a clear solution. Neither are very accurate and there is actually more of a difference between them than in the T+240 above. Sunday 6th, T+192: ECM: GFS: While ECM is again wide of the mark, GFS compares remarkably well to our T+0, especially at that range. The big question at this stage is, if GFS has found the pattern, will it continue to show the low in that position off the NE coast on Monday 7th? Monday 7th, T+168: ECM: GFS: Notice here how the models seem to have started to come closer in terms of the northern hemisphere. Both show the Azores high, and similar positions to the surface lows, although there are still differences in the Atlantic and the UK. On balance I'd probably go for the GFS as most accurate here, but it's not as good a match as the GFS T+192. Tuesday 8th, T+144: ECM: GFS: UKMO: We now have UKMO charts. ECM has now started to catch on to the pattern and is probably more accurate in terms of the UK than GFS and UKMO at T+144. Wednesday 9th, T+120: ECM: GFS: Again a good match from ECM, with the low tracking down the UK from the NW. GFS has the Azores high positioning almost correct, but ends up with two cut off lows either end of the UK and a ridge in between, so not a correct pattern. Apologies - didn't save UKMO. Thursday 10th, T+96: ECM: GFS: UKMO: ECM has backed away slightly from the correct solution, with the low too far north. GFS has moved towards the ECM and the correct solution with the ridge removed from the UK but also has the low too far north. ECM probably still the better of the two. Friday 11th, T+72: ECM: GFS: UKMO: Now we're getting there. Better agreement here from ECM,GFS and UKMO on the elongated low tracking SE across the UK. ECM and UKMO probably slightly have the edge, with GFS having the low slightly too deep north of Scotland. Saturday 12th, T+48: ECM: GFS: UKMO: Again GFS persists with a similar shape of low, but it's now not quite as deep. Sunday 13th, T+24: ECM: GFS: UKMO: As you'd hope and expect, the pattern shown by all three models is a good match for the actual chart. Summary: if you've ever wondered why the models never verify, just look at how many different solutions we see day in, day out for the same time, of which only one can be accurate. It's strange to see both ECM and GFS agree on a 'false' solution at T+240, and I wonder if anyone would have been fooled by this agreement at long range. The first relatively accurate chart is from GFS at T+192, and although ECM doesn't seem to catch on too well until T+144, it then seems to be more accurate right up until T+48. Obviously this is just one example, and it would probably be a good idea to do the same in a standard zonal pattern to see how well they compare. You can of course just go to http://www.emc.ncep....gmb/STATS_vsdb/ to see the official verification stats, although these don't tell you specifically how the models are performing over the UK. * see http://forum.netweat...00#entry2448734 and http://forum.netweat...80#entry2457976 Incidentally, considering the snow in the east today, a good call from those who mentioned the 14th in the stratosphere thread a good few weeks ago...

A light covering of snow here earlier, first flakes since Dec 2010. Sadly turned to rain so didn't last long. Here's hoping for a bit more as the week goes on!

NMM has sleet for areas in mid and east Wales tomorrow, with a chance of some snow later on in the evening in SE (probably light and showery in nature). Sadly looks like rain here in the west at the moment. Sunday looks like a drier day. Jury is out on Monday at the moment in terms of specifics - all depends what happens with that front!

COAMPS = "The Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS®) has been developed by the Marine Meteorology Division (MMD) of the Naval Research Laboratory (NRL). The atmospheric components of COAMPS®, described below, are used operationally by the U.S. Navy for short-term numerical weather prediction for various regions around the world." http://www.nrlmry.navy.mil/coamps-web/web/view I'm not sure how well regarded it is, but considering I've never heard any of the respected members here or any other forecaster mention it, I would guess it isn't used much for the UK.

If you go to the ECMWF site, you can view ECM charts (and ensemble mean) highlighted with deviation from the mean. Next Monday for example shows considerable uncertainty in the Atlantic: http://www.ecmwf.int/products/forecasts/d/charts/medium/eps/ensm/essential!MSLP!Europe!168!pop!od!enfo!plot_ensm_essential!2013010712!!/

Saw this post earlier. I've been comparing the 12z runs for next Monday 14th. You're probably right about the old run data being out of date but by using the same model and run you get a chance to see how the latest data affects the predicted setup since the model's algorithm and the type of data collected will be the same, so any changes should be purely down to the data itself. For example yesterdays 12z and today's 12z GFS are very different, notably in our neck of the woods, as might be expected at this timeframe, but as we get nearer to the date it will be interesting to keep an eye on the same run to see how quickly it settles on a solution and compare to the other models. The differences between the same model and run in 24 hours demonstrate that a small change in input data can produce a very large change in synoptics for the UK currently at this range and so nothing is settled yet in terms of the final outcome. Why compare 12z to 12z and not just 12z to 18z for example? One reason is that apparently the 12z and 00z feature balloon data which isn't included in the 06z and 18z (although this article was written in 2009 so may no longer be accurate): â€For the GFS, look at a couple days of 00Z and 12Z runs for consistency. Avoid the 06Z and 18Z runs when 00 or 12 is available; in the U.S. these runs don't include the weather balloon network data (balloons are only sent up twice per day), and are therefore radically different and more likely to have bias.†http://m.accuweather...naccurate/18097 (Some other interesting info in there including verification stats for 1985 to present) There's a great feature on the NW data centre which allows you to compare charts at the click of a button between the current and previous run too - it would be even better if this could be extended to include the same run for the previous day at some point

Good point. I've started to follow the 12z and save one particular chart each day for 14th Jan to see how the models cope with this evolution. If the SSW modelling and forecast effects on the troposphere are correct I would be more inclined to trust yesterdays 12z than today's at this range given where the PV is modelled. Yesterday Today

Looking at recent model output, perhaps the biggest risk is for the block to be positioned wrongly and for the coldest air to be deflected south of the UK. The signals are there for blocking to set up to the north so what we need to look for as the colder synoptics near the reliable timeframe is for lower pressure to dig SE under the block which will allow the cold air to flood in from the continent. The other favourable resolution would be for the high pressure to back west allowing in low pressure from the east. There seems to be no evidence of a zonal pattern developing so getting a block set up in a favourable position is going to be what to look for in the next couple of weeks I think to avoid disappoment - which hopefully we will.