Jump to content
Snow?
Local
Radar
Cold?
IGNORED

12z Model Comparisons Results


Recommended Posts

Ian, have you split your results with the type of weather conditions experienced, so that we might be able to better judge whether GFS is better at zonal conditions as an example.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

I haven't (yet), but it would be easily possible to do so from the results I've obtained so far.

Link to comment
Share on other sites

Posted
  • Location: just south of Doncaster, Sth Yorks
  • Location: just south of Doncaster, Sth Yorks

hi Ian

As Julian suggests that would really be big insight into each model with differing weather patterns. More work for you I'm afraid.

John

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

Done. :D

I tried a number of different categories. I categorised the general west-east flow into Blocked, Zonal, and then into different types of block (Block to East, Block around Britain, Greenland Block, Mid-Atlantic Block).

I then categorised synoptic types into Cyclonic, Anticyclonic, Westerly, Southerly, Easterly and Northerly.

There is too small a data sample to make any major conclusions, but here are some pointers:

1. The models seem to score more highly under blocked than zonal conditions. This may be because with high pressure in place there is more margin for error, whereas in a low-pressure situation a difference of low pressure positioning by just 200 miles can mean the difference between a SW'ly and a northerly.

2. The models seem better at handling northerly and easterly types than westerly and southerly types. The reasons may be related to the above.

Model synopsis (each model is judged relative to the performances of the other models, rather than in terms of absolute figures)

GFS: Good at handling mid-Atlantic blocking and northerly wind situations. Surprisingly the model seems to perform about the same relative to the others in both blocked and zonal conditions. Top performing model for Mid-Atlantic Block.

UKMO: Good in blocked conditions, when it is very close to the ECMWF, but relatively poor in zonal conditions and at handling blocking over Greenland.

ECMWF: A consistent performer across the board, though less good relative to the other models in northerly wind situations. ECMWF is the top performing model for all situations bar Northerly and Mid-Atlantic Block.

NOGAPS: A low-quality model, though decent at handling mid-Atlantic blocks.

JMA: A good model for Northerly types (best of all models in this case) but very poor at handling zonal conditions, particularly of the zonal westerly type.

GEM: Good at handling easterly types, very poor for zonal, westerly and northerly types.

model_comparisons.xls

Link to comment
Share on other sites

Posted
  • Location: just south of Doncaster, Sth Yorks
  • Location: just south of Doncaster, Sth Yorks

hi

That is really fascinating reading Ian and thanks a lot for going to so much trouble. Another 12 months(!!!) of this and we should be able to use the models more confidently. Maybe even switching models based on the predicted weather type by the best model for that type. Still be huge pitfalls but its objective analysing like this that will eventually help get more accurate forecasts for longer ahead.

Can I ask my usual favour please?

Thanks to Microsoft and their refusal to help I am still unable to download Excel format in Internet Explorer. I've also tried using other types and for some reason it will not work in any.

many thanks

John

ps

sorry I meant an e mail to me please with the excel file enclosed.

johnh@johnholmes.charitydays.co.uk

Edited by johnholmes
Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

An email has been sent. I sent it in a Word document as I recall it not sending properly the last time I sent it as an Excel document.

Link to comment
Share on other sites

Posted
  • Location: Liphook
  • Location: Liphook

Very intresting observations there, its intresting to see that some people's ideas and my own are actually somewhat proven by those results.

One thing I'd love to know is whether the runs from GFS also differ from each other. I seem to have the feeling that the 06z tends to be the blocked run and that the 12z is the more zonal run...one day I may just look into that unless someone already has because its intresting to know each runs strengths and weaknesses.

(ps, prehaps the 12z possible bias to zonality is why it had the bad spell on John's obs of the 12z earlier in the year when things became more blocked.)

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

16th June 2006

Very slack 1020mb high over Britain, as a ridge from the Azores. Slack 1015mb low off NW Scotland. Very light WSW winds.

GFS: A very good T+144, high displaced slightly out into the Azores but with a strong 1016-1020mb ridge shown over Britain, and the lows in the right places. T+120 less good, still had us in high pressure but with the high centred at 1025mb to the east. T+96 was perfect.

I award this run 8/10.

UKMO: Poor T+144, with a high out over Scandinavia, SSE winds for the NE, and a low at 1010mb over southern England. T+120 was pretty good, the high slightly overdone but everything correctly placed. T+96 was similar to the GFS at T+120, with high at 1025mb, slightly to the east, and overdone.

Only a 6/10 for this one. The T+144 curse returns.

ECMWF: Good T+144, high marginally underdone, but at 1015-1020mb and a ridge over Britain, it was a good output, and lows in the right places. T+120 was similar to the GFS T+120, with the high a bit to the east and at 1025mb. T+96 was excellent.

Overall, this run scores 8/10- the same as GFS.

NOGAPS: High over Scandinavia at T+144 with SSE winds and a possible thundery breakdown. T+120 similar to GFS and ECMWF, high at 1025mb and slightly to the east. T+96 was pretty good with 1020mb high over Britain, though it had cut off the connection with the Azores High due to the low to the NW being overdone.

Overall 6/10.

JMA: Pretty much perfect at T+144. T+120 slightly overdid the high, but it was another good output. T+96 also good, but overdid the high again.

Overall another 8/10.

GEM: A good T+144 high slightly underdone, but in the right place, lows also in the right places. T+120 was the same, while at T+96 the high was of the correct intensity, but the low to the NW underdone.

Another good run which I award an 8/10.

So very good from the GFS, ECMWF, JMA and GEM. NOGAPS and UKMO had problems at T+144.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

17 June 2006

High at 1020mb over the south, slack westerly winds over Britain. 1010mb low off NW Scotland and 1000mb low out in the Atlantic.

GFS: High overdone and to the east at T+144, giving anticyclonic southerlies. A good T+120, with the high slightly underdone but all features in the right places. Too progressive at T+96, with a 1010mb low over N Scotland and moderate westerlies everywhere.

Only a 6/10, because of the poor T+96.

UKMO: Similar T+144 to GFS, with the high overdone and displaced to the east, giving us anticyclonic southerlies. High to the NE at T+120, with easterly winds shown over the British Isles. 1015mb trough shown over Britain at T+96, with high pressure to the north, south and east.

I can't believe that this is a UKMO run that I am seeing this in, but it was a very poor run indeed. Only 3/10.

ECMWF: High a bit to the NE at T+144, with easterlies over the south. Same mistake at T+120. Fairly accurate at T+96, with the high in the right place but a little underdone.

Overall only 6/10.

NOGAPS: High a bit to the NE at T+144, with easterlies over the south. Same mistake at T+120. Same at T+96, but with a low over southern England pumping up thundery showers.

Only 4/10.

JMA: 1020mb high displaced a bit to the east at T+144 with anticyclonic southerlies. High to the NE at T+120, with easterlies over the south. A mess at T+96, with pressure at 1015mb over the British Isles and neither low nor high pressure.

Only a 6/10.

GEM: High displaced to the east at T+144, with SSW winds over Britain. T+120 was quite good, high overdone though and covering the whole country, not just the south. High to the NE at T+96, with easterlies over the south.

I award this 6/10.

Overall, a fairly poor set of outputs. A recurring theme was to have the high positioned to the NE, with easterly winds over southern Britain.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

18 June 2006

1005mb low off N Scotland and a deeper 1000mb low off NW Scotland, giving SW winds over Britain and rain spreading east.

GFS: At T+144 the main low was positioned a bit too far south, giving southerly winds over Britain and rain only just approaching the west (possible thundery breakdown even?) T+120 was quite good, low about 100 miles too far west though and so we were shown in a westerly flow, rain only just arriving. T+96 overdid the low, at 996mb, with a pronounced SW flow shown over the British Isles.

I award this run a 7/10, it was generally quite accurate but niggly differences from the actual outcome existed at all three timeframes.

UKMO: High over Scandinavia at T+144, with a 1010mb low over southern England, and probably thunderstorms. At T+120 the output resembled the correct outcome but with the Atlantic low some 300 miles too far west, and slack westerly winds for Britain. T+96 was very accurate.

I give this 6/10, but really apart from T+96 it wasn't as good as the GFS, T+144 was particularly poor.

ECMWF: High over Scandinavia at T+144, with a pronounced ESE flow and low pressure over France; even worse than the UKMO T+144. T+120 was accurate, the low marginally to the NE of where it actually ended up, giving us a more pronounced SW flow. T+96 was disappointing though, the low too far north, and a ridge from the Azores affecting the southeast.

I only award this a 5/10.

NOGAPS: Same as ECMWF at T+144. At T+120 had the main low well to the south of where it ended up, giving us a cyclonic southerly flow. Accurate at T+96.

I award this a 5/10 because of the good T+96, but otherwise it was a poor run.

JMA: 1010mb low over S England and easterlies elsewhere at T+144; poor again. T+120 quite accurate, main low about 200 miles to the west of where it ended up and overdeepened, but Britain correctly shown in WSW winds. Quite a good T+96, but low slightly too far east.

I award this a 6/10.

GEM: 1020mb high shown over southern Britain at T+144, Atlantic systems well away to the north. Low overdeepened at T+120, with strong SSW winds over Britain. Low too far SE and underdone at T+96, with weak 1010mb feature over Scotland and slack SW winds over Britain.

Overall just 5/10.

All of the models had problems, I think the GFS was the closest this time. The Euro models, together with NOGAPS and GEM, seemed to like the idea of having a Scandinavian High and easterly winds at T+144.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

19 June 2006

1005mb low over Scotland and westerly winds over England, Wales & Ireland. A mostly cloudy day with drizzle (I think, at least that's the way it was over N England). 990mb low well out into the Atlantic.

GFS: T+144 was pretty good. The main 1005mb low was a bit to the south of where it actually ended up, with westerlies only for the southern half of the country, but overall it was pretty accurate. T+120 was less good, the low overdeepened at 996mb and way too far north, giving westerlies for all. Very good at T+96 with everything almost exactly spot on.

I award this a 7/10 because of the disappointing T+120, the other timeframes were good.

UKMO: At T+144 the low was slightly overdeepened at 1000mb and positioned to the NW of Ireland, giving moderate SSW winds over Britain. T+120 had similar problems to GFS with the low too far north, though less so than with the GFS T+120, just north of Scotland. However, the low was even further north at T+96, even with a ridge from the Azores influencing the south.

Only 5/10. The UKMO doesn't generally seem so good in zonal conditions.

ECMWF: Low way too north at T+144, with westerlies over the N, high pressure in the S. The low was even further north at T+120, with high pressure over England & Wales. Even at T+96 the low was way too far north, with high pressure covering England and Wales.

Only 3/10. It was never particularly close, the low was always too far north and it always had the south under high pressure (and I mean its own anticyclone, not just an Azores ridge!)

NOGAPS: Low at 995mb over Iceland at T+144, giving SSW winds over the rest of Britain. Low marginally too far north at T+120 but a pretty good output overall; cyclonic westerlies shown over Britain and low over the far north of Scotland. Spoilt by an ECMWF-style T+96, with high pressure over the south, and the low well to the north of Scotland.

Only 5/10 due to the poor T+96.

JMA: T+144 was similar to UKMO- low out to the West with SSW winds shown over Britain. The low was impossible to detect at T+120, with a 1010mb trough over the country, and very slack winds. At T+96 the low was way too far north, with WSW winds and an Azores ridge extending into the south.

Only 4/10.

GEM: Low too far north and overdeepened at T+144; WSW winds over Britain, an Azores ridge in the south. Low too far NW at T+120, over Iceland, but with 1010mb trough over western Britain and a slack regime, similar to JMA. Low off NW Scotland at T+96, with westerly winds shown over Britain.

I rate this at 5/10 as it was a bit closer to correct outcomes than the JMA, but not by much.

The GFS today was not brilliant, but the other models were rather poor, consistently positioning the low too far to the north and west. With accurate outputs at two out of the three timeframes, the GFS was the most accurate of the models.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

20 June 2006

A low at 980mb off NW Scotland, propelling very strong south-westerly winds over to the British Isles. Quite a simple-looking chart, but here's how the models did...

GFS: At T+144 the low was correctly positioned but underdone, at 1000mb, thus the south-westerlies over Britain were slacker and an Azores ridge affected the SE. Not bad overall though. T+120 again had the general pattern right, but the low was situated over Iceland- positioned too far north- with westerly winds for Britain. At T+96 the output was very accurate indeed, the low marginally too far north resulting in a slightly slacker SW flow, but no more than by about 50-100 miles.

Overall a 7/10 again, it was quite a good run but the GFS had problems with the low at both T+144 and T+120.

UKMO: Low underdone and too far west at T+144, with high pressure over southern England and the wet/windy weather only just approaching the west. At T+120 the low was underdone and too far north; quite accurate for northern Britain but had high pressure over the south. Low too far west at T+96, with westerlies over Britain and high pressure over the south.

Only 4/10.

ECMWF: Lows too far north at T+144 with a strong Azores ridge over England & Wales. Low displaced to Iceland at T+120, but unlike the GFS, had slack high pressure over Britain. T+96 was quite good, but the low was underdone and a bit far to the NW, allowing high pressure to cover the south.

Overall this run scores 5/10, being better than the UKMO, but there were more problems with the low than on the GFS.

NOGAPS: Similar to GFS at T+144, low correctly placed but underdone, meaning slacker SW winds over Britain and high pressure in the south, but not a bad output. T+120 was also good, the low marginally too far west, but the strong SW flow and rain correctly shown over Britain. At T+96 the low was underdone with more of a westerly flow shown for Britain, but still, the general pattern was correct, rain in roughly the right place.

I give this run an 8/10.

JMA: Low way too far west at T+144. A trough over northern Scotland with slack westerlies, high pressure over the south. Low too far west at T+120, with SSW winds and high pressure over the SE. Low too far west and underdone at T+96 but a better output, with westerlies over Britain and SWs coming into the west.

Another 5/10.

GEM: Low too far north and underdone at T+144, with cyclonic SW winds and high pressure over the SE, though not too bad a T+144 output. Low too far NE at T+120, situated to the north of Scotland with slack westerlies over most of Britain, strong over northern Scotland. T+96 was better, similar to JMA with the low too far west and westerly winds over Britain, SW winds coming into the west.

This run scores 6/10.

It seems that the Euro models and also JMA struggled with this situation. GFS was solid if unspectacular, but for once, the award for "most accurate model" goes to NOGAPS- possibly for the first time ever!

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

21 June 2006

Big 985mb low over the north-western tip of Scotland with very strong westerlies over Britain. One of those days where "autumn feel" could be argued for.

GFS: Low too far north at T+144, by a long way, with moderate WSW winds over Britain- the westerly dominated pattern being correct is the only good thing I can say about that output. T+120 was better, but the low was still too far north, with the strong W winds confined to Scotland, lighter winds and higher pressure in the south. T+96 was pretty good, orientation of low slightly different though, giving more of a W/NW flow.

I only rate this at 5/10.

UKMO: At T+144 the low was off to the NW, with SW winds over Britain and high pressure over the SE, low also very much underdone. T+120 was getting there, but low still too far W, with a SSW flow over Britain. T+96 was an excellent output though.

I think this was slightly better than the GFS, but only slightly, so 6/10.

ECMWF: T+144 had the low way too far north and only slack westerlies for Britain, worse even than the GFS T+144. T+120 similar to GFS T+120- low too far north, with strong winds for Scotland but only light winds in the south. T+96 was excellent though.

I rate this also at 5/10.

NOGAPS Low a bit too far north at T+144, but had strong cyclonic W winds over the whole of Britain, easily the best of the T+144s so far. T+120 had the low correctly positioned but underdone, with moderate to strong cyclonic westerlies over Britain. No T+96.

A 8/10 for the NOGAPS.

JMA: Low way underdeepened and too far north at T+144 with just slack westerlies. Good T+120, similar to NOGAPS, with low a bit underdone but in the right place. T+96 very accurate.

I award this 7/10.

GEM: Low way too far north at T+144, with slack westerlies. Low way underdone at T+120 but in the right place, with moderate cyclonic westerlies. Low too far north at T+96 but quite close, with strong WNW winds for Britain.

This run scores 5/10.

I don't rate the NOGAPS highly in general, but it seems to have been the best model at handling the approach of this deep depression, which really surprised me. Maybe this might be one of NOGAPS's strengths- as always with these comparisons, time will tell.

22 June 2006

A deep 990mb low off NE Scotland with strong NW winds over Britain. An occlusion ensured that most places did not see "sunshine and showers", instead being mostly cloudy.

GFS: Low too far north at T+144, 1010mb trough over N England and light westerly winds. Low way too far NE at T+120 with a strong ridge from the Azores, and north-westerly winds. Accurate at T+96.

I award this 5/10 because of the good T+96, but the general tendency was to overdo the Azores High and underdo the low pressure.

UKMO: Low too far south at T+144 but quite a good stab, the low over Scotland with strong WNW winds elsewhere. Low too far north at T+120 with WNW winds and an Azores ridge into the south, but still closer than the GFS was. A good T+96.

This scores 6/10.

ECMWF: Similar to GFS at T+144, low too far north, and 1010mb trough over Britain with slack westerlies. Decent T+120, low a bit to the north, but with a strong NW flow shown and the high to the south not overdone. A good t+96.

I award this 7/10.

NOGAPS: An excellent T+144, very little to fault. However at T+120 the low was too far north, with an Azores ridge into the south. No T+96.

I award this 6/10.

JMA: Low underdone and too far west at T+144 with a westerly flow over Britain, but not too bad. Low way too far north at T+120 with an Azores ridge over the south. Quite good T+96, low slightly underdone.

This scores a 6/10.

GEM: Low vastly underdone at T+144- just a 1015mb feature (!) with a slack northerly flow. Low too far W at T+120 with a westerly flow for Britain. Accurate T+96.

This scores 5/10 due to the good T+96, but otherwise poor.

Situation back to normal with the ECMWF producing the best output, although all models were over-eager to bring in West is Best's fabled Azores High.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

Some Apologies

Haven't been updating these for a while, I have to come clean and admit that my birthday celebrations were the main thing that put a temporary hold to the updates.

Updates will commence from 30 June to 7 July, whereupon I will be away for 2 weeks, then updates will continue from around 25 July onwards.

Hope this isn't too much of a problem, if anyone wants to cover during 7-25 July they are welcome.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

30th June 2006

Only T+96 is available, but it's a start. A high at 1030mb over Scandinavia, 995mb low SW of Iceland and a slack SSW flow over Britain.

GFS: Pretty accurate T+96, had the slack SSW flow correct although it had the low a bit further to the east, and the Scandinavian High likewise, so pressure was shown as being a bit lower.

I say 8/10.

UKMO: Low near Iceland underdone, and Scandinavian High a bit overdone, with stronger southerlies shown over the British Isles, with easterlies for the far SE, though the general pattern was correct.

This scores 7/10.

ECMWF: An accurate output, although again the low SW of Iceland was slightly underdone. No significant errors otherwise, the low perhaps a little to the east of where it ended up with slightly lower pressure over Britain.

8/10.

NOGAPS: Not a great T+96, the low too far east and a secondary feature to the SW. We were still shown in southerlies but with more of a cyclonic bias.

6/10.

JMA: An accurate output, similar to the Euro models with the low to the NW slightly underdone in intensity, but everything else correct.

Overall 8/10.

GEM: Low way underdeepened at 1005mb and too far east, resulting in a cyclonic S/SW flow over Britain and low pressure in charge.

Only 5/10.

Something a bit more meaningful will be released tomorrow, as I will have both T+96 and T+120 to go on.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

1 July 2006

1030mb high over southern Scandinavia. 1000mb Icelandic Low, 1020mb trough over southern France/Spain and southerlies over Britain.

GFS: T+120 was pretty good, with the high in the right place and southerlies over Britain. The lows in the Atlantic were rather underdone though. T+96 was again good, but again low pressure was a bit underdone in the Atlantic, it had a southerly over Britain and the 1020mb trough right though.

I say 8/10 for this run.

UKMO: 1010mb low over W Ireland at T+120, with the high too far east, though it did have us in southerlies. T+96 had the opposite problem- the high was too far west, giving easterlies for southern and eastern Britain.

Only 6/10.

ECMWF: T+120 similar to UKMO- lows too far east, giving more of a cyclonic S/SW flow- if anything I think this output was worse than the UKMO, in that low pressure was dominant rather than Britain sandwiched between low and high pressure. T+96 was excellent though.

I think the T+96 was sufficiently good to make this run a 7/10, but not overall as good as the GFS.

NOGAPS: Strong southerlies and low pressure over Ireland at T+120- quite poor. No T+96 so I'm unable to do a fair comparison with other models, so no mark, but I'm guessing it would have been a low one.

JMA: Had a High erroneously positioned to the SW, so we were in a slack area of 1020-1025mb pressure with highs to the SW and NE at T+120. T+96 still had a high wrongly placed to the SW, giving us slack high pressure.

Overall just 6/10.

GEM: No T+120, but T+96 resembled the GEM, with too much high pressure to the SW.

The GFS just shaves it today, tomorrow all three timeframes will generally be available (except where they were missing on WZ, which can't really be avoided)

Link to comment
Share on other sites

Some Apologies

Haven't been updating these for a while, I have to come clean and admit that my birthday celebrations were the main thing that put a temporary hold to the updates.

Updates will commence from 30 June to 7 July, whereupon I will be away for 2 weeks, then updates will continue from around 25 July onwards.

Hope this isn't too much of a problem, if anyone wants to cover during 7-25 July they are welcome.

I'll do at least some of the summaries, as per http://www.netweather.tv/forum/index.php?showforum=68 and then you could mark the models on your return.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

Thanks a lot for the offer for assistance- it will be much appreciated.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

2 July 2006

High pressure at 1030mb over the Baltic, with light easterlies over Britain and 1015mb trough over France.

GFS: T+144 was quite good, though the 1015mb trough was situated over Britain rather than France, making a potentially big difference to the weather. T+120 not so good, had strong SE winds over Britain and a stronger, 1010mb, low to the SW. T+96 was pretty good, but low again a bit overdone to the S at 1012mb.

I think this run should get 7/10 for being quite close to correct with the general pattern right at all three timeframes, but dodgy aspects particularly at T+120.

UKMO: No T+144. T+120 was absolutely spot on and I can't find anything to fault at all, while T+96 slightly overdid the high to the north but was overall very accurate in positioning of all of the features and had us in slack easterlies.

Fortunately for the UKMO there is no T+144, otherwise it might have dented the mark. Based on the T+120 and T+96 I give it a 9/10.

ECMWF: Very accurate at T+144. However T+120 had a way overdeepened low (1005mb) to the SW, worse than the GFS, with strong cyclonic SE winds. T+96 was accurate, high to the north marginally overdone, like on the UKMO.

I mark this down to 7/10 due to the poor T+120.

NOGAPS: Cyclonic southerlies at T+144- poor run. High way overdone at T+120 and too far SW, with the 1030mb high almost on top of NE Scotland. High to SW overdeepened at T+96 with strong SE winds for Britain.

Only 4/10.

JMA: Massively overdeepened 1000mb low to the NW at T+144, though to be fair, not really influencing the British Isles. Other aspects of the T+144 were good. T+120 was very good with little to fault, as was T+96, identical to the ECMWF T+96.

I award this an 8/10, but the T+144's overdeepened low at 1000mb is a typical JMA error.

GEM: 1010mb low over Britain at T+144- and not even a trough like the GFS. A col at 1015-1020mb at T+120. And again at T+96, though it at least had the Scandinavian High right this time.

Only 4/10.

Overall, well done to the UKMO model, though it "cheated" a bit by having a missing T+144 (Entire Grid Undefined). Still, the T+120 and T+96 combined were the best such outputs of any model.

Link to comment
Share on other sites

Posted
  • Location: Lincoln, Lincolnshire
  • Weather Preferences: Sunshine, convective precipitation, snow, thunderstorms, "episodic" months.
  • Location: Lincoln, Lincolnshire

3 July 2006

1015mb weak low over France with pronounced easterly over Britain. 1025mb high over Baltic and 1025mb high ridging from the Azores intoi NW Scotland.

GFS: 1010mb low to the west at T+144, with SSE winds, and low pressure in charge, so not the greatest of outputs. T+120 was decent but the low to the S was displaced slightly to the north and the Azores ridge was underdone. Low to S overdone at T+96 at 1010mb with strong easterlies.

I give this a 6/10, it kept wanting to overdo low pressure.

UKMO: T+144 was good, though like the GFS T+120 it underdid the Azores ridge into the NW, overall it had the 1015mb low and easterly flow in the right places, slightly overdoing the Scandinavian High. T+120 had 1030mb high to the north, somewhat overdone. The high was still a bit overdone at T+96, but both timeframes had us in an easterly with 1015mb low to the south.

I give this 7/10.

ECMWF: Overdone low at 1005mb to the SW at T+144, with cyclonic SE winds for the south, easterlies for the north. T+120 was very accurate, acores ridge into NW slightly underdone. T+96 very good, high to N slightly overdone at 1025-1030mb.

I award this 7/10 as well, let down by the T+144.

NOGAPS: High way overdone at T+144, extending into Scotland at 1030mb. 995mb Atlantic low off W Scotland at T+120, with 1010mb trough over S England and southerlies. Only T+96 was vaguely accurate, but to be fair it was quite good, with the low slightly overdeepened and placed to the SW, but easterlies correctly over Britain and 1025mb high to the north.

A 5/10.

JMA: Good T+144, low slightly too far north to our S but very good overall with E winds shown and 1025mb high to the north. T+120 also top notch, Azores ridge slightly underdone being the only error. T+96 was also very good.

I award this 8/10.

GEM: Lows to S and W overdone, slack pressure at 1010mb over Britain and high way too far NE at T+144. High underdone at T+120 but with lows in the right places, very slack easterlies shown. Accurate T+96.

I award this 6/10.

Overall, the JMA, surprisingly, seemed to be the best model at handing the easterly flow. Other models had a tendency to overdo the low pressure; of the main models the two Euro models slightly outperformed the GFS.

Link to comment
Share on other sites

  • 2 weeks later...

11th July 2006

Actual Chart

Marks ECM 6/10 Low Pressure not deep enough over Iceland, HP well forecast.

NOGAPS 6/10 Low Pressure not deep enough over Iceland, HP too strong over UK.

UKMO 8/10 Low Pressure west of Iceland, only slightly under developed, Ridge over UK well handled, but central pressure slightly too high.

GEM 9/10 Good Forecast on Low West of Iceland, HP slightly wrongly positioned.

JMA 4/10 Low Pressure west of Iceland not forecast, and High wrongly positioned.

GFS 6/10 Low Pressure not deep enough over Iceland, only slightly under developed, HP not ridged for enough over UK.

Link to comment
Share on other sites

12th July

Based on the T+96 AND T+120.

GFS Very Good on Low Pressure East of Iceland, and effect in Northern UK in terms of position and pressure. High Pressure to South also Good 9/10.

T+120 High Pressure Still good, but High slightly over High at centre, good dealing with secondary depression, effect on Northern UK Good, but underdone low pressure 8/10. Giving 8/10 overall.

GEM T+96 very good, HP very close, secondary low off Greenland and low pressure off Iceland, only slightly out.

T+120 low over Iceland fairly accurate, Secondary low not on charts, but high pressure also fairly close Overall 7/10.

ECM Low over Iceland well handled but effect on UK understated, ridge over southern UK slightly overdone

T+120 High Pressure excellently dealt with including ridge over southern UK. LOW Pressure over Iceland understated but effect on UK very good.7/10 overall

UKMO T+120 Low Pressure fairly close but not quite deep enough and HP also quite good. T+96 Low Pressure correct pressure but slightly too far left, Ridge over UK dealt with well but Over all High Pressure too high, giving a 6/10 overall.

NOGAPS T+120 Has Ridge over UK way overdone, and has 2 cells around Iceland.

T+96 Ridge over UK overdone, Low Pressure close to Iceland about right pressure but too far north understates effect ion UK. 5/10 overall.

JMA T+120 and T+96 High Pressure Ridge too close to UK, Pressure over done, and Low Pressure over Iceland not nearly developed enough. But the T+120 at least a low pressure East of Iceland, Not That Good 4/10.

Link to comment
Share on other sites

15th July

ECM(96) : High Pressure slightly too high in centre, but High Pressure block very close. Low Pressure over Iceland very well predicted, low Pressure SW of Iceland, slightly over done but well aligned. 9/10

ECM (120) : High Pressure correctly predicted in terms of pressure but badly positioned over UK, Low Pressure round Iceland reasonable. Low Pressure SW of Iceland slightly too far but correct pressure. Over Developed feature West of Greenland , which does not feature In actual charts. 8/10

ECM(144) : High Pressure very close to bring right, Low Pressure around Iceland decently handled and low pressure SW of Iceland too far West. 9/10 (For T+144)

8/10 Overall

UKMO(96): High Pressure correct pressure but wrongly aligned in south, Low Pressure over Iceland reasonably predicted, low Pressure SW of Iceland, well over developed by 15mb. 7/10

UKMO(120): High Pressure correct pressure overall but wrongly aligned and too low pressure for Southern UK. Low Pressure over Iceland reasonably predicted but slightly over developed. Low Pressure SW of Iceland, well over developed by 20mb. 7/10

UKMO (144): High Pressure around UK but not over UK, Low Pressure SW of Iceland missed and situation over Iceland mishandled, a poor effort 3/10.

6/10 Overall

NOGAPS(96): High Pressure far too dominant over UK, Low Pressure over Iceland, general picture correct, but pressure in wrong place. Low Pressure SW of Iceland, too far South and overdeveloped. 4/10

NOGAPS (120) High Pressure Over Done. 6/10

NOGAPS (144): High Pressure over UK in reasonable alignment, Low Pressure around Iceland reasonably dealt with, no Low SW of Iceland. 6/10 (For T+144)

5/10 Overall

GFS (96): High Pressure good attempt but slight wrong alignment. Low Pressure over Iceland, general picture correct, but pressure in wrong place. Low Pressure SEW of Iceland, too far South and overdeveloped. 6/10

GFS(120): High Pressure good attempt but too weak in Azores area.

Low Pressure over Iceland, general picture correct, and pretty good attempt. Low Pressure SEW of Iceland, too far South and overdeveloped by 20mb. 6/10

GFS(144) : Reasonable attempt at High over UK, Low Pressure SW of Iceland well handled, and Low Pressure around Iceland reasonably handled. 8/10.

7/10 Overall

GEM(96): High Pressure slightly too high in centre, but High Pressure block very close, but underdeveloped in Azores. Low Pressure over Iceland well predicted, low Pressure SW of Iceland, too far west and slightly over developed. 7/10

GEM(120): High Pressure wrongly aligned and too weak, general picture around Iceland OK but wrong positioning and Low Pressure SW of Iceland omitted completely. 4/10

GEM(144): High pressure over UK but completely wrong alignment, Low Pressure SW of Iceland well dealt with, but Low pressure around Iceland underdeveloped. 5/10.

5/10 Overall

JMA(96): High Pressure good attempt but slight wrong alignment. Low Pressure over Iceland, general picture correct, but pressure in wrong place, low Pressure SW of Iceland, too far west and slightly over developed. 8/10

JMA(120): High Pressure good attempt but slight wrong alignment. Low Pressure over Iceland, bur pressure too high in Greenland, low Pressure SW of Iceland, slightly too far West and wrongly aligned. 8/10

JMA(144): High Pressure around UK but not over UK, Low Pressure SW of Iceland missed and situation around Iceland not good, but some indication of a depression, 4/10.

6/10 Overall

Link to comment
Share on other sites

16th July

ECM(96) : High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west. Low Pressure over Iceland/GIN area well dealt with. 8/10

ECM (120): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west. Low Pressure over Iceland/GIN area slightly overdone. Overdone low pressure SW of Iceland 7/10

ECM(144): High Pressure too weak and cut-off from main Azores High, low pressure over Gin reasonably dealt with. 6 /10.

7/10 Overall.

UKMO(96): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too big. Low Pressure over Iceland/GIN area well dealt with. 7/10

UKMO(120): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure slightly too far west. Low Pressure over Iceland/GIN area dealt with well. Well Overdone low pressure SW of Iceland 7/10

UKMO (144): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure slightly too far west and cut-off from Azores High. Low Pressure over Iceland/GIN area dealt with well. Well-Overdone low pressure SW of Iceland 6/10

7/10 Overall

NOGAPS(96): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west, Low pressure over Gin area generally right, but has overdeveloped low SW of Iceland massively, more so than GFS 5/10.

NOGAPS (120): High Pressure overdone at 1035mb, with alignment of High pressure not good and mainly cut-off from Azores High. Overdone Low pressure in GIN and SW of Iceland, Weak overall. 4/10.

NOGAPS (144): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure slightly too far west. Low Pressure over Iceland/GIN area dealt with well, but overdone low pressure NE of Iceland. More accurate than T+96 and T+120. 7/10

5/10 Overall

GFS (96): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west, low pressure over Gin well dealt with but Low Pressure SW of Iceland far too developed. 6/10

GFS(120): High Pressure overdone at 1035mb, with alignment of High pressure not good and mainly cut-off from Azores High. Overdone Low pressure in GIN and SW of Iceland, Weak overall, looks very similar to NOGAPS T+120 4/10.

GFS(144): High Pressure overdone at 1035mb, with alignment of High pressure not good and mainly cut-off from Azores High. Overdone Low pressure in GIN and SW of Iceland 5/10.

5/10 Overall

GEM(96): High Pressure at 1030mb correctly. But centred too close to UK and pressure not deep enough over Azores. Way overdone pressure SW of Azores. 5/10

GEM(120): High Pressure at 1030mb correctly. But centred too close to UK. Way overdone pressure SW of Azores. Low pressure not well dealt with over western GIN. 6/10

GEM(144):Low Pressure over GIN vaguely accurate, with High Pressure way out to the west, Poor, a generous 4/10.

Overall 5/10

JMA(96): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west. Low Pressure over Iceland/GIN area well dealt with. 8/10

JMA(120): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west. Low Pressure over Iceland/GIN area well dealt with. Overdone low pressure SW of Iceland 7/10

JMA(144): High Pressure at 1030mb correctly and generally well aligned, but centre of High Pressure too far west and too big. Low Pressure over Iceland/GIN area well dealt with, but overdeveloped low pressure to SW of Iceland, 7/10

Average 7/10

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • UK Storm and Severe Convective Forecast

    UK Severe Convective & Storm Forecast - Issued 2024-03-29 07:13:16 Valid: 29/03/2024 0600 - 30/03/2024 0600 THUNDERSTORM WATCH - FRI 29 MARCH 2024 Click here for the full forecast

    Nick F
    Nick F
    Latest weather updates from Netweather

    Difficult travel conditions as the Easter break begins

    Low Nelson is throwing wind and rain at the UK before it impacts mainland Spain at Easter. Wild condtions in the English Channel, and more rain and lightning here on Thursday. Read the full update here

    Netweather forecasts
    Netweather forecasts
    Latest weather updates from Netweather

    UK Storm and Severe Convective Forecast

    UK Severe Convective & Storm Forecast - Issued 2024-03-28 09:16:06 Valid: 28/03/2024 0800 - 29/03/2024 0600 SEVERE THUNDERSTORM WATCH - THURS 28 MARCH 2024 Click here for the full forecast

    Nick F
    Nick F
    Latest weather updates from Netweather
×
×
  • Create New...