North Balti Zen Posted February 13, 2012 Share Posted February 13, 2012 The GEFS actually isn't terrible. The ensembles (er, ensemble mean, I guess) have been awful this winter. Glad they are showing something, I guess, but unless or until the OP is showing it, not buying ensembles as a "saving grace".* *Except for the control run of the Euro. That's clearly the one to hang your hat on... Link to comment Share on other sites More sharing options...
TowsonWeather Posted February 13, 2012 Share Posted February 13, 2012 verification statistics are not applicable to assessing the potential accuracy of a single event on a regional basis. next. Well, of course. But we also can't have it both ways. We can't mock the models that perform worse and then turn around and dismiss verification stats when someone is using them to defend a model's legitimacy. Is the UKIE solution correct? Almost certainly no, as we all know. But it is a legit model and its solution is one more data point to add onto the pile. Link to comment Share on other sites More sharing options...
snowfan Posted February 13, 2012 Share Posted February 13, 2012 Behind the JMA, right? CRAS. Get it right. Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 Well, of course. But we also can't have it both ways. We can't mock the models that perform worse and then turn around and dismiss verification stats when someone is using them to defend a model's legitimacy. Is the UKIE solution correct? Almost certainly no, as we all know. But it is a legit model and its solution is one more data point to add onto the pile. we should probably consider the fact that a lot of weenies completely misuse the verification stats. Link to comment Share on other sites More sharing options...
stormtracker Posted February 13, 2012 Share Posted February 13, 2012 The 12z Euro run is the most important run in ji's life. Link to comment Share on other sites More sharing options...
H2O Posted February 13, 2012 Share Posted February 13, 2012 I think this puts it all in perspective. http://mag.ncep.noaa...el=&imageSize=M Look at how fast the flow is and all moving parts. It's possible that the ns and ss can link up for the weekend but thinking any model is capable of locking in at this range isn't a good idea. The good news is split flow and vorts zipping along in both streams. Can't get a storm without it. Bad news is no blocking or decent amplification of the flow. It's the proverbial "perfect timing" or "thread the needle" scenario. We'll see I guess but the odds are clearly stacked against us. Haven't had much luck this year so maybe can hit a 16 and draw a 5 against the dealer's 20. To follow up, look at how many vorts the GFS shows just at 123hr. No way it will be able to lock down a storm this far out accurately with so much going on. I'm not saying that this is the reason it doesnt have a storm but one reason it's solutions vary so much run to run. http://mag.ncep.noaa.gov/NCOMAGWEB/appcontroller?prevPage=Model&MainPage=indexℑ=&page=Param&cycle=02%2F13%2F2012+12UTC&rname=UPPER+AIR+PARMS&pname=500_vort_ht&pdesc=&model=GFS&area=NAMER&cat=MODEL+GUIDANCE&fcast=123&areaDesc=North+America+-+US+Canada+and+northern+Mexico&prevArea=NAMER&currKey=model&returnToModel=&imageSize=M Link to comment Share on other sites More sharing options...
stormtracker Posted February 13, 2012 Share Posted February 13, 2012 To follow up, look at how many vorts the GFS shows just at 123hr. No way it will be able to lock down a storm this far out accurately with so much going on. I'm not saying that this is the reason it doesnt have a storm but one reason it's solutions vary so much run to run. http://mag.ncep.noaa...el=&imageSize=M This is much more logical than DT's idea. The model has so many damn vorts out there and they all change position with each run, so it's likely that 18z will look different. And then 0z will look different....and then... Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 To follow up, look at how many vorts the GFS shows just at 123hr. No way it will be able to lock down a storm this far out accurately with so much going on. I'm not saying that this is the reason it doesnt have a storm but one reason it's solutions vary so much run to run. http://mag.ncep.noaa...el=&imageSize=M we've seen this story a lot the past two winters. it does F with solutions but it also makes it harder to get good amplification. we need like 5-10 things to all come together right at once for any big storm. Link to comment Share on other sites More sharing options...
Bob Chill Posted February 13, 2012 Share Posted February 13, 2012 To follow up, look at how many vorts the GFS shows just at 123hr. No way it will be able to lock down a storm this far out accurately with so much going on. I'm not saying that this is the reason it doesnt have a storm but one reason it's solutions vary so much run to run. http://mag.ncep.noaa...el=&imageSize=M Exactly. Without having block and/or a big amplified pattern, it's unreasonable to expect models to pinpoint a solution 5-6-7 days in advance. Get one thing wrong early on and then everything else in the later panels exaggerates the error. IMO- the reason the gfs doesn't have the storm is because it's much more likely to miss the connection than have the 2 streams dig and phase perfectly. It wouldn't boost my confidence at all even if the gfs does show a storm at this range. All we can really hope for is some sort of a stormy solution come wed-thurs and then let the details work themselves out as we approach. Link to comment Share on other sites More sharing options...
chris87 Posted February 13, 2012 Share Posted February 13, 2012 Well, of course. But we also can't have it both ways. We can't mock the models that perform worse and then turn around and dismiss verification stats when someone is using them to defend a model's legitimacy. Is the UKIE solution correct? Almost certainly no, as we all know. But it is a legit model and its solution is one more data point to add onto the pile. you're missing the broader statistical point. it's fine to make an argument for why the UKMET solution could be correct, but using verification statistics as evidence is not the way to go. Link to comment Share on other sites More sharing options...
yoda Posted February 13, 2012 Share Posted February 13, 2012 EURO out to 96 Link to comment Share on other sites More sharing options...
mitchnick Posted February 13, 2012 Share Posted February 13, 2012 you're missing the broader statistical point. it's fine to make an argument for why the UKMET solution could be correct, but using verification statistics as evidence is not the way to go. I think the diff between the two of you is he is making the argument in general that the UK has a better chance of verifying than other models and your suggesting he make an argument why it is more likely to be right in its depiction of this storm and I don't think anyone can make that argument this far out with this system Link to comment Share on other sites More sharing options...
WinterWxLuvr Posted February 13, 2012 Share Posted February 13, 2012 You can always look for things that might be giving you a clue. At 5+ days, you'd think the ensembles would be useful. If not, why have them? Anyway, at 0z, 0 members had 0.1+ precip in Va. at 1AM Sun morn., 6z had 2 members that did it at the same time frame, and 12z has 5 that do it. This wouldn't necessarily argue for amplification into a big storm, but at least it might suggest less suppression. Be interesting to see what eveyones daddy has to say. Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 euro not markedly different than 0z thru 114, perhaps even less impressive with moisture at least. flow is very flat in the east.. not sure how gom moisture is going to climb. Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 may be a bit slower tho now thru 120.. hard to say if that changes anything. Link to comment Share on other sites More sharing options...
yoda Posted February 13, 2012 Share Posted February 13, 2012 euro not markedly different than 0z thru 114, perhaps even less impressive with moisture at least. flow is very flat in the east.. not sure how gom moisture is going to climb. At 120 there are two s/w... one in N KS/S NE and one SW TX ejecting out... 1008 L W GOM http://vortex.plymouth.edu/gifs/120213182316227628000.gif Link to comment Share on other sites More sharing options...
mitchnick Posted February 13, 2012 Share Posted February 13, 2012 http://vortex.plymouth.edu/cgi-bin/gen_grbcalc2.cgi?re=namer&id=&zoom=.6&ti=0≥=1280x1024&mo=ecmwf≤=500&va=hght&in=60&pl=cf&ft=h120&cu=latest&overlay=no&mo=≤=&va=&in=&pl=ln&ft=h24&cu=latest Link to comment Share on other sites More sharing options...
TowsonWeather Posted February 13, 2012 Share Posted February 13, 2012 you're missing the broader statistical point. it's fine to make an argument for why the UKMET solution could be correct, but using verification statistics as evidence is not the way to go. I understand, and I think we agree here. I think we veered off track because when Ji was responding to people who were mocking the UKIE by pointing out that it was a legit model, you took that as a defense of the model's specific solution rather than a statement about the model's legitimacy in general. Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 At 120 there are two s/w... one in N KS/S NE and one SW TX ejecting out... 1008 L W GOM http://vortex.plymou...16227628000.gif is a little better in the northern stream at 126 and by 132 500mb is definitely better than 0z. gom low a bit west too. it might try to come further north. Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 well this run is much better.. trough comes in later but digs a lot by 138. it might do it or be close. low near fl/al/ga border at 138. Link to comment Share on other sites More sharing options...
yoda Posted February 13, 2012 Share Posted February 13, 2012 144 -- 1003 Low is in C GA moving NE ( I think based on Ian's 138 hr location) Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 light snow up into s va by 144.. low on ga/sc border at coast. trough prob not quite sharp enough. Link to comment Share on other sites More sharing options...
mappy Posted February 13, 2012 Share Posted February 13, 2012 http://vortex.plymou...t=h24&cu=latest Do you think the next time you post a panel image, you could add some explanation behind it. I am sure many who are reading/lurking would love to know what this panel shows. Link to comment Share on other sites More sharing options...
Midlo Snow Maker Posted February 13, 2012 Share Posted February 13, 2012 12z euro hr 150 Link to comment Share on other sites More sharing options...
Bob Chill Posted February 13, 2012 Share Posted February 13, 2012 You can always look for things that might be giving you a clue. At 5+ days, you'd think the ensembles would be useful. If not, why have them? This is a very good point. I look at the ensembles alot. Helps me understand various possibilities instead of just taking the op and running with it. I never post much about the ens though. I'll never be able to wrap my head around each solution and say "this is more likely because of this" or vice versa for each member. There are very few non-mets on the board who have the skill to be able to disassemble each ens and understand what is right and wrong with each member. I find too many people post indiv members just because they have snow and have no skill to explain why the solution is more correct than the other 95% that show nothing. Link to comment Share on other sites More sharing options...
Ian Posted February 13, 2012 Share Posted February 13, 2012 wow at 150.. snow up into dc area.. looks pretty nice.. low off nc coast Link to comment Share on other sites More sharing options...
Midlo Snow Maker Posted February 13, 2012 Share Posted February 13, 2012 hr 150 1000 mb low near ilm 0.25" to ric Link to comment Share on other sites More sharing options...
yoda Posted February 13, 2012 Share Posted February 13, 2012 light snow up into s va by 144.. low on ga/sc border at coast. trough prob not quite sharp enough. I have it about 50-100 miles back west on Plymouth of that... Link to comment Share on other sites More sharing options...
Srain Posted February 13, 2012 Share Posted February 13, 2012 WEATHER RECONNAISSANCE FLIGHTS CARCAH, NATIONAL HURRICANE CENTER, MIAMI, FL. 0100 PM EST MON 13 FEBRUARY 2012 SUBJECT: WINTER STORM PLAN OF THE DAY (WSPOD) VALID 14/1100Z TO 15/1100Z FEBRUARY 2012 WSPOD NUMBER.....11-075 I. ATLANTIC REQUIREMENTS 1. NEGATIVE RECONNAISSANCE REQUIREMENTS. 2. OUTLOOK FOR SUCCEEDING DAY.....NEGATIVE. II. PACIFIC REQUIREMENTS 1. FLIGHT ONE -- TEAL 72 A. P-56/ DROP 9 (44.3N 151.0W)/ 15/0000Z B. AFXXX 18WSC TRACK 56 C. 14/1845Z D. 16 DROPS ON TRACK AS PUBLISHED E. AS HIGH AS POSSIBLE/ 15/0600Z FLIGHT TWO -- NOAA49 A. P-56/ DROP 9 (44.3N 151.0W)/ 15/1200Z B. NOAA9 19WSC TRACK56 C. 15/0730Z D. 16 DROPS ON TRACK AS PUBLISHED E. 41,000 TO 45,000 FT/ 15/1800Z 2. SUCCEEDING DAY OUTLOOK: A. P-56/ DROP 9/ 16/0000Z B. P-56/ DROP 9/ 16/1200Z Link to comment Share on other sites More sharing options...
TowsonWeather Posted February 13, 2012 Share Posted February 13, 2012 wow at 150.. snow up into dc area.. looks pretty nice.. low off nc coast Dr. Maybe! Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.