Model Output Discussion

[Thundery wintry showers]

Maybe the GFS takes individual frost hollows and extends them to cover an unrealistically large area? I think the temperature progs for other areas of the country are realistic but there's no way south-east England will see temperatures that low on a wide scale.

[Cheese Rice]

Maybe the GFS takes individual frost hollows and extends them to cover an unrealistically large area? I think the temperature progs for other areas of the country are realistic but there's no way south-east England will see temperatures that low on a wide scale.

Why would the GFS be programmed to even know where frost hollows are? Perhaps in the higher resolution on NW extra but otherwise I would be surprised as it is hardly of use, like-wise why wouldn't GFS prog frost free zones? Doesn't really make sense.

...

OT- Got down to -0.6c this morning, with extensive frost on cars and grass etc, pretty notable for May (I think).

Edited May 11, 2010 by 10123

[mushymanrob]

well the 12z still holds the promise of a gradual change to warmer weather as the azores high is expected to nudge in over the weekend and into early next week. it still looks pretty dry though (saw loads of showers here today, but the rain never reached the ground)

[Stargazer]

Maybe the GFS takes individual frost hollows and extends them to cover an unrealistically large area? I think the temperature progs for other areas of the country are realistic but there's no way south-east England will see temperatures that low on a wide scale.

The GFS certainly seems to be sticking to its guns (if I recall correctly it was predicting similar temperatures several days ago). UKMO gives a minimum of +3 for Reading and +2 for Benson (a well-known frost hollow), values which I'd say are more realistic given that the 30 year low in May for most parts of the south is around -1 to -2. [Edit: but then Benson is shown by the Met Office as the coldest spot in the region on Sunday night at -1.6, so who knows...]

Edited May 11, 2010 by Stargazer

[Optimus Prime]

UKMO always overestimate nightime minima while GFS mostly underestimates them.

Benson is the coldest spot for the southern portion of England. Some of these low minima for that location;

May 2005 -2.3c (15th)

May 1997 -4.1c (7th)

May 1996 -4.2c (4th)

May 1995 -2.6c (14th)

[Stormyking]

The warmer weather thats been showing lately has become rather consistent so far and I think 12z looks better than 06z and now as some have said now that its now getting into the reliable time frame :lol:

Fingers crossed for nice warm or hot weather.

[SNOW GO]

Some warmer weather still in the running but currently as always the typical UK weather has to through a spanner in the works in the potential of heavy rain as the chart shows as we approach the weekend of the 22cnd/23rd of May

[Milhouse]

The warmer weather thats been showing lately has become rather consistent so far and I think 12z looks better than 06z and now as some have said now that its now getting into the reliable time frame

Fingers crossed for nice warm or hot weather.

This evenings ECM is good but no where near as good as it could be. However its that time of year now where we dont just need a southerly to reach 20c. By the looks of it most parts should see at least 20c sometime next week.

[SP1986]

Encouraging outlook but I must concede although the area of the -5C seems overdone given that Benson is currently just 1C, it could well get some -4 or -5C there tonight, given it is a notable frost hollow

[damianslaw]

The GFS chart doesn't look right. It doesn't show any frost in the central northern Pennines areas i.e. Teesdale/Upper weardale, also it seems to omit Cumbria from frost, these areas will be the coldest in the country tonight. I always take these charts with a pinch of salt, usually very inaccurate.

Back to the charts, signs of a slow warm up later this week, but it will be a slow one especially in the north with a north westerly flow persisting.

Next week increasing signs of azores ridging, but I suspect it will be very weak ridging and any real warmth will be resigned to the south east.

[SP1986]

Damian I think the highest temperatures will be in upland areas at this time of year at night in these setups, despite the low upper level temperatures. I looks like any real frost tonight will occur in low lying sheltered areas (although upper level frost hollows like Sunnybridge will get hard frosts). At this time of year a 1mph wind can prevent an air frost.

[damianslaw]

Damian I think the highest temperatures will be in upland areas at this time of year at night in these setups, despite the low upper level temperatures. I looks like any real frost tonight will occur in low lying sheltered areas (although upper level frost hollows like Sunnybridge will get hard frosts). At this time of year a 1mph wind can prevent an air frost.

Yes low lying areas under such set ups should see the lowest temps, however, the GFS chart is not accurate much of inland Cumbria is due to hit 0 degrees at least and I suspect many sheltered low lying parts of North Yorkshire, the charts do not show this. In my post I mentioned valley locations where cold air quickly drains especially in steep sided valleys such as in the Lake District, perhaps less so the Pennine valleys which are not so deeply incised.

Also, some higher ground such as Shap Fell often shares similiar low temps to lower ground its a notorious frost hollow.

Edited May 11, 2010 by damianslaw

[Thundery wintry showers]

Re. Stephen Prudence, it's perhaps most accurate to say that the lowest temperatures are usually found within valley basins, i.e. low ground relative to the surrounding area, but often contained within areas of high ground. For example, Braemar and Aviemore in the Scottish Highlands are notorious frost hotspots.

Damianslaw has raised some good points regarding the GFS's overnight temperature progs and effectively refuted my previous theory- it seems that the temperature estimates are just inaccurate sometimes with no obvious consistent pattern of inaccuracy. There are certain traits of the GFS which are consistent though, e.g. consistently overestimating minima in many coastal regions and much of East Anglia.

[Optimus Prime]

The areas you mention are tiny, and England as a whole is a very small country so I wouldn't take much notice of your all so anal regional areas. So it's not that the GFS is 'very inaccurate' as you so bluntly put, it's because England is such a small area to map withb a lot of cities sqeezed in and upland districs making all the difference.

I've done so many tests on the GFS at 7 days ahead and it usually comes within 0.4c of the outcome. So accurate I don't bother trying to criticize it anymore.

[Thundery wintry showers]

I should also mention that these problems with GFS seem to relate mainly to cold nights under good radiation conditions. In other circumstances the GFS's temperature projections are often reasonably accurate.

[mushymanrob]

the gfs and ecm both go for the azores high nudging in early next week, not sure the fax at t120 supports that though..

showers? what showers? we had a sprinkle out of a laden sky... and theres no real sign of any real rain.

[summer blizzard]

http://www.wetterzentrale.de/pics/Rtavn1801.png

GFS stil going for a warmer and sunnier spell from 144 hours out, until then a rubbish showery, coole spell.

[LadyPakal]

For those arguing about the projected GFS overnight temperatures: http://news.bbc.co.uk/1/hi/uk/8677263.stm

[Milhouse]

http://www.wetterzen...s/Rtavn1801.png

GFS stil going for a warmer and sunnier spell from 144 hours out, until then a rubbish showery, coole spell.

Wow. What a chart!!! Could be the first time of the year when we use the word 'hot'.

[SP1986]

For those arguing about the projected GFS overnight temperatures: http://news.bbc.co.uk/1/hi/uk/8677263.stm

It was cold granted, but nowhere near the widespread -4C to -7C that the GFS predicted at interludes earlier this week. Still I think we can put this one to bed as perhaps a case of extended frost pockets, whatever it is, after tonight, frost is unlikely to be seen for many locations until next Autumn.

[phil nw.]

It was cold granted, but nowhere near the widespread -4C to -7C that the GFS predicted at interludes earlier this week. Still I think we can put this one to bed as perhaps a case of extended frost pockets, whatever it is, after tonight, frost is unlikely to be seen for many locations until next Autumn.

Come on Stephen i think you are being somewhat pedantic on this.You do GFS a disservice imo.

Ok one can always find isolated examples of inaccuracy,s but It`s not a mile off based on last night`s mins.

GFS is a global model and it`s grid pattern isn`t going to be able to hone in on every tiny location with the accuracy that you seem to demand incessantly.

You have had a bee in your bonnet about this for months and i agree with you it`s time to let this go.

[BlueHedgehog074]

An improvement sure looking evident from the lastest ECMWF and GFS updates showing the high pressure gradually extending North-Eastwards towards the UK. The UK Met Office also follow a similar trend showing the high reaching the edge of southern England around 120 hours. The most settled weather will likely to be the South and East, with the placement of the high helping to drag in warmer air from the South-West or West. Still a chance of spells of rain/showers spilling in from the Atlantic at times, particularly the further North and West you go (where the pressure won't be quite as high), but with warmer temperatures and the high being more of a feature during next, I think it should feel quite pleasant.

Perhaps still a while away to be sure whether this will verify, but considering what the models have been showing in some of their lastest updates, temperatures of 18*C and above is looking more feasible for next week.

Thereafter, according to the GFS, it looks as though the cooler wet weather returns as the high retreats back to the South-West and low pressure to the East, with another duration of cool winds from the North West to North East. Probably likely to chance, though.

Edited May 12, 2010 by Rainbow Snow

[SP1986]

Come on Stephen i think you are being somewhat pedantic on this.You do GFS a disservice imo.

Ok one can always find isolated examples of inaccuracy,s but It`s not a mile off based on last night`s mins.

GFS is a global model and it`s grid pattern isn`t going to be able to hone in on every tiny location with the accuracy that you seem to demand incessantly.

You have had a bee in your bonnet about this for months and i agree with you it`s time to let this go.

If you want me to let it go, don't give me reason not to! Let this be the last sentence ever on this matter.

[jonathon]

If you want me to let it go, don't give me reason not to! Let this be the last sentence ever on this matter.

i dont agree with you that it will be the last frosts till auturm. the met office is still forecasting snow for northern hills in there 15 day outlook and frosts also from midlands northwards till at least 2 or 3 weeks time.

[SP1986]

Well true in the interior of the country frosts can last into June, and also in frost hollows, but for the coast and cities it will be, if it isn't then this year will go down in memory for being one of frosty Mays if not the frostiest in some living memory.