Model output discussion - into 2018

[blizzard81]

If ecm is anywhere near the mark this morning, it has been one of the worst weeks performances for it's op runs ever. 

[Weather-history]

ECM op 0z run has gone from that 3 days ago to this now at the same time frame. A 1035mb high cell around Iceland to a 975mb low southwest of Iceland, It maybe just the op run but its almost embarrasing.

Edited January 5, 2018 by Weather-history

[Weather-history]

And 2 days ago. Look at that!

[radiohead]

2 minutes ago, Weather-history said:

And 2 days ago. Look at that!

Absolutely dreadful. It's a reminder that even though it's the best performing model, it is still capable of getting things very wrong.

[bluearmy]

7 minutes ago, Weather-history said:

And 2 days ago. Look at that!

TEN DAYS OUT KEV !!!

at the risk of repeating myself ................

the ecm ens mean remains the best tool in this type of circumstance as its neither over progressive or under progressive.  A check on the spreads against this mean would have consistently told you that the Atlantic was going over the blocking. It would also have told you that the uppers were not going to be low enough to bring convectively snowfall from any easterly flow.  Before the Icelandic clusters were available, I would have posted the above days ago. Yet they didn't point in this direction. I'm going back to tried and tested methods now !!!!

[pureasthedriven]

4 minutes ago, radiohead said:

Absolutely dreadful. It's a reminder that even though it's the best performing model, it is still capable of getting things very wrong.

All verification stats aside, I really don’t see it as the best performing model. Time and again it over-amplifies, especially in the 7-10 day range and it also has good form for last minute flips, too. The oft-maligned GFS, volatile as it often is,  to me, calls the broad scale pattern changes the best, if one doesn’t get too obsessed by the micro-detail. That said, we’re not ‘there’ yet in terms of the actual weather. 

[comet]

1 minute ago, radiohead said:

Absolutely dreadful. It's a reminder that even though it's the best performing model, it is still capable of getting things very wrong.

Just to reiterate its the best performing model hemispherically  wise not necessarily for just our neck of the woods.

Must admit to being very disappointed at the output this morning. My expectation was very high following the 18z last night and expecting to see much more amplification in the pattern ahead. Note how to weekend cold spell has almost disappeared with the sinking high which I shared my concerns about several times in the last few days. oh well lets see what the 12z bring.

[shotski]

6 minutes ago, pureasthedriven said:

All verification stats aside, I really don’t see it as the best performing model. Time and again it over-amplifies, especially in the 7-10 day range and it also has good form for last minute flips, too. The oft-maligned GFS, volatile as it often is,  to me, calls the broad scale pattern changes the best, if one doesn’t get too obsessed by the micro-detail. That said, we’re not ‘there’ yet in terms of the actual weather. 

If that’s the case why bother with the operational runs at all ? GEFS and ECM ens should be the pin up charts. Saying that, the extended EPS flip daily from one extreme to another.....

edit: should have quoted Blues post. 

Edited January 5, 2018 by Weather toad

[Duane S.]

I swear the ECM OP used to be better three years or so ago, before the more recent ‘upgrades’. 

[Frosty.]

The Ecm 00z shows an increasingly chilly weekend coming up with overnight frosts and a quiet start to next week but then the atlantic bulldozes through with spells of wet and windy weather and it also becomes mild at times. 

[pureasthedriven]

11 minutes ago, Weather toad said:

If that’s the case why bother with the operational runs at all ? GEFS and ECM ens should be the pin up charts. Saying that, the extended EPS flip daily from one extreme to another.....

edit: should have quoted Blues post. 

Well, quite! 

[mulzy]

12 minutes ago, Weather toad said:

If that’s the case why bother with the operational runs at all ? GEFS and ECM ens should be the pin up charts. Saying that, the extended EPS flip daily from one extreme to another.....

edit: should have quoted Blues post. 

Use operationals up to day 5, ensemble guidance up to day 10.

[comet]

I think you would have to say judging by the operational output we have in front of us this morning and the runs we have seen in the last few days,the ecm has been woefull. 

Maybe it's a massive mild outlier.

[ArHu3]

3 minutes ago, comet said:

I think you would have to say judging by the operational output we have in front of us this morning and the runs we have seen in the last few days,the ecm has been woefull. 

Maybe it's a massive mild outlier.

It was yesterday, judging by the plume but massive mild operational outliers have a habit of verifying whilst for cold you need almost the entire ensemble suite there 

For the last cold spell (2013!) the plume looked like this 

 

 

Edited January 5, 2018 by ArHu3

[Weather-history]

47 minutes ago, bluearmy said:

TEN DAYS OUT KEV !!!

at the risk of repeating myself ................

the ecm ens mean remains the best tool in this type of circumstance as its neither over progressive or under progressive.  A check on the spreads against this mean would have consistently told you that the Atlantic was going over the blocking. It would also have told you that the uppers were not going to be low enough to bring convectively snowfall from any easterly flow.  Before the Icelandic clusters were available, I would have posted the above days ago. Yet they didn't point in this direction. I'm going back to tried and tested methods now !!!!

I know. 

[blizzard81]

48 minutes ago, bluearmy said:

TEN DAYS OUT KEV !!!

at the risk of repeating myself ................

the ecm ens mean remains the best tool in this type of circumstance as its neither over progressive or under progressive.  A check on the spreads against this mean would have consistently told you that the Atlantic was going over the blocking. It would also have told you that the uppers were not going to be low enough to bring convectively snowfall from any easterly flow.  Before the Icelandic clusters were available, I would have posted the above days ago. Yet they didn't point in this direction. I'm going back to tried and tested methods now !!!!

I think people's issue with the ecm is not the lack of cold uppers it was showing. That was very apparent to most. The issue is the performance of consecutive ops and the synoptics they were showing. The ops are supposed to be higher resolution when compared to the ensembles. The ecm ops have been woeful this week and there have been similar bad episodes from the ecm over the last few years. 

[Troubleatmill]

11 minutes ago, ArHu3 said:

It was yesterday, judging by the plume but massive mild operational outliers have a habit of verifying whilst for cold you need almost the entire ensemble suite there 

Yes but easy to forget that for our neck of the woods if we were a dice 1-5 would be mild and 6 would be cold so always more chance of calling mild correct, synoptics would quite likely look different at t0 to whenever the model's' called it 

Edited January 5, 2018 by Nicholas B

[mountain shadow]

The only straw I can grasp is that the UKMET T+144 still has the Scandi ridge in place with some energy underneath. 

Need to wait for ECM ens for the clusters but that ECM op is very progressive.

[blizzard81]

2 minutes ago, mountain shadow said:

The only straw I can grasp is that the UKMET T+144 still has the Scandi ridge in place with some energy underneath. 

Need to wait for ECM ens for the clusters but that ECM op is very progressive.

I think ukmo will look more like ecm (and what gfs has been showing consistently - credit where credit due) on the evening run. 

Edited January 5, 2018 by blizzard81

[nick sussex]

Hello I'm the ECM.

On Wednesday I promised you a two week holiday travelling on first class to New York.

On Thursday I promised you just a week in Paris.

On Friday I cancelled your holiday.

Which sums up the descent of the ECM to some dire looking output for coldies bar the last frame tease! Of course now that the ECM has by far the worst looking output it will probably be right!

By T96hrs you knew it was a horror show with it being much more progressive  than the rest, having said that theres a large spread in the ensembles in some key areas shown by the ECM spreads.

I fear the worst because the ECM has form here, dropping the coldest solution and then kicking coldies when they're down.

Tonight we'll see whether the ECM is correct, given the timeframes thankfully the drama should end by this evening!

 

 

[TEITS]

The wildly different Op runs is making me sceptical of any output at the moment as none of them agree. The UKMO remains the pick of the bunch though.

Think I shall just sit back and watch this unfold rather than commenting on each run until we get some order amongst the chaos!

[snowspotter]

Block setting up nicely dry weather here we come . UKMO hinting at very good possibilities 144 . Quietly confident on this spell coming up.

Its gonna get colder peeps 

[ArHu3]

3 minutes ago, Nicholas B said:

Yes but easy to forget that for our neck of the woods if we were a dice 1-5 would be mild and 6 would be cold so always more chance of calling mild correct, synoptics would quite likely look different at t0 to whenever the model's' called it. Suppose rather than a six sided dice we are looking for the bonus number out of 49 for cold though ?

I don't live in your neck of the woods though but it it's still true here ?. But in the last 3 years I haven't seen a hlb actually surviving but I have seen dozens being modeled but in the end they all just toppled even when toppling only had a very slim chance of happening in the ensemble suite 

[Frosty.]

I would like to see the cold block at least put up a decent fight, the Ecm 00z just blows it apart with ease. I still think next week could be interesting if the block holds its ground / fights back and the atlantic struggles to make too many inroads as the ukmo suggests..it's not game over..yet!

Edited January 5, 2018 by Frosty.

[West is Best]

I agree Frosty. The ECM is pretty grim.

Though it pains me to say so, we ignore or (worse) diss the GFS at our peril. Unless it comes on board nothing is either likely or certain. It remains the gold standard model: head and shoulders above the others imho. Dammit.