Cory Posted November 1, 2012 Share Posted November 1, 2012 U.S. forecast's late arrival stirs weather tempest "A European forecast that closely predicted Hurricane Sandy's onslaught days ahead of U.S. and other models is raising complaints in the meteorological community." http://www.usatoday....models/1668867/ There have been a couple of other stories on this from other outlets as well. Link to comment Share on other sites More sharing options...
yoda Posted November 1, 2012 Share Posted November 1, 2012 Washington Post also had a short mention of it on Tuesday I believe Link to comment Share on other sites More sharing options...
hm8 Posted November 1, 2012 Share Posted November 1, 2012 The Euro did kinda make the GFS it's b**ch. That being said, this event was so anomalous, it's not really a great measuring stick for how good a model is. Link to comment Share on other sites More sharing options...
isohume Posted November 1, 2012 Share Posted November 1, 2012 Haha...gotta give Cliff Mass credit for keeping the fire burning! Hey, he could run for Senator someday. I somehow doubt 99.99% of the public cares it took the GFS a few runs to latch onto the ECMWF soln 5 days out. The next time the ECMWF fails to predict counter to the GFS, I'm holding a press conference. Link to comment Share on other sites More sharing options...
Wx4cast Posted November 1, 2012 Share Posted November 1, 2012 The Euro did kinda make the GFS it's b**ch. That being said, this event was so anomalous, it's not really a great measuring stick for how good a model is. Umm yes it is! The more anomalous an event the more likely for some models to underperform while others measure up. GFS has a history of poor verification scores vs EC and even the UK on a daily basis. Link to comment Share on other sites More sharing options...
Srain Posted November 1, 2012 Share Posted November 1, 2012 I find it interesting just how quickly some have forgotten TS Debby and that fiasco... Link to comment Share on other sites More sharing options...
Ed Lizard Posted November 1, 2012 Share Posted November 1, 2012 Pure wishcasting from 240 hour Euro, Hurricane Sandy gets captured by the polar trough, and produced multi-day Halloween Storm Part Deux for the SNE subforum that people will make movies about. Much better than yesterday's 12Z moderate tropical storm for Florida. So much better. I hope I'm not 5 posted for a month for this... ETA Just checked SNE subforum banter thread, and I'm not the only person in Weenie Wonderland http://en.wikipedia....rricane_of_1804 240 hour Euro was suggesting the phasing... Link to comment Share on other sites More sharing options...
THEREALTOR1 Posted November 1, 2012 Share Posted November 1, 2012 U.S. forecast's late arrival stirs weather tempest "A European forecast that closely predicted Hurricane Sandy's onslaught days ahead of U.S. and other models is raising complaints in the meteorological community." http://www.usatoday....models/1668867/ There have been a couple of other stories on this from other outlets as well. The European model is widely seen as the best at predicting hurricanes, Mass and others say, as demonstrated with Hurricane Isaac in late August. ...as well as just about every other major meterological event that's occurred in our region in the past few years. Anybody that's spent any time on here knows that the Euro is the rock on which all models eventually crash themselves on to (usually). I guess it could be interesting for this to go mainstream though, as i've read for years as mets (and everyday weather enthusiasts such as myself) have complained that OUR models just don't compare with the Euro when it comes to sniffing out a solution, nor the accuracy at which those solutions typically unfold. Maybe this will lead to major upgrades to the current models? Although I think I rather prefer to see maybe more funding or grants and whatnot allotted for individuals and the private sector to maybe make an effort to build a better mousetrap (metaphorically)? Link to comment Share on other sites More sharing options...
Ian Posted November 1, 2012 Share Posted November 1, 2012 A lot of mets don't even understand statistical scores from what I can gather. There was a lot of GFS bashing over the last week on Twitter from tv mets etc. I see rando pubilc being angry that the Europeans have a better model in the last few days. Just what we need is more bad information swirling around. Euro superior--sure.. but the GFS is pretty damn good. Link to comment Share on other sites More sharing options...
isohume Posted November 1, 2012 Share Posted November 1, 2012 Welp...looks like the latest ECMWF runs are now going with upper midwest sfc cyclogenesis at 120 hr. Welcome to the party. The GFS had this last time I worked the medium range...3 days ago. Someone get me Shepard Smith on the line. Link to comment Share on other sites More sharing options...
wi_fl_wx Posted November 1, 2012 Share Posted November 1, 2012 Kerry Emanuel writing about it makes it a big deal. His words give significant weight to ideas that everyone has been thinking about or quietly discussing for a long time. With everyone talking about the great forecast for Sandy and how it saved countless lives and millions if not billions of dollars, it is an opportune time to bring up the topic. For day to day weather, the difference between the ECMWF and GFS doesn't really matter that much. Nobody's going to lose their life or have property damaged if the high tomorrow is 72 or 75. Both models can accurately predict if it is going to rain in 3 days--and of course there are counter-examples, the ECMWF is not always right. But for an event like Sandy with so much at stake, the fact that the ECMWF had the right track several days ahead of GFS is not good for the GFS. I doubt anything will change in the near future, but it is good to have this out in the open now. Link to comment Share on other sites More sharing options...
LocoAko Posted November 1, 2012 Share Posted November 1, 2012 A lot of mets don't even understand statistical scores from what I can gather. There was a lot of GFS bashing over the last week on Twitter from tv mets etc. I see rando pubilc being angry that the Europeans have a better model in the last few days. Just what we need is more bad information swirling around. Euro superior--sure.. but the GFS is pretty damn good. That, and the "off-run" myth just refuses to die amongst hobbyists and meteorologists alike. Statistical scores be damned. I give up. Link to comment Share on other sites More sharing options...
usedtobe Posted November 1, 2012 Share Posted November 1, 2012 Haha...gotta give Cliff Mass credit for keeping the fire burning! Hey, he could run for Senator someday. I somehow doubt 99.99% of the public cares it took the GFS a few runs to latch onto the ECMWF soln 5 days out. The next time the ECMWF fails to predict counter to the GFS, I'm holding a press conference. The euro was first but the gfs also was pretty doggone early forecasting the storm giving plenty of lead time. lol at Mass. He's a smart guy who is like hyperbole and being critical of the NWS. He ought to save it for when the models really bust. Link to comment Share on other sites More sharing options...
donsutherland1 Posted November 1, 2012 Share Posted November 1, 2012 Each model has its strengths and its biases. Simplistic Euro vs. GFS comparisons on a single storm do not accurately convey the value-added. If Maas were committed to improvements in forecasting high-impact events, not what seems to be an unending crusade against NCEP and the U.S. models, he would do well to recall that it was the GFS that first indicated that the Boxing Day blizzard was going to bring a substantial snowfall to portions of the East Coast. His analysis would cover a representative sample of high-impact events. Clearly, on this event, the Euro did better. However, its edge is not so decisive that one can ignore all the U.S. guidance. Doing so would lead, overall, to inferior forecasts. That decline in forecast quality would, in turn, increase risks to public safety, among other issues. Link to comment Share on other sites More sharing options...
mnweather Posted November 1, 2012 Share Posted November 1, 2012 The ggem caught onto the scenario pretty early, Granted it wasnt as consistent as the euro as to where it would make landfall but it still had it early. Link to comment Share on other sites More sharing options...
CoastalWx Posted November 1, 2012 Share Posted November 1, 2012 Funny how people are praising the Euro, but have we forgotten Debbie? Does nobody care about this because Debbie didn't have the impact? The euro is a good model, but it has flaws like every model. It's also wise to use ensembles guidance beyond 5 or 6 days out, not necessarily deterministic guidance. I hope Sandy doesn't cause people to just focus on the op runs. Link to comment Share on other sites More sharing options...
blizzardof09 Posted November 2, 2012 Share Posted November 2, 2012 some good points in this thread, the gfs was for lack of a better word owned by the euro in the storm as stated earlier, but every model isnt perfect and during the winter were gonna see the gfs sometimes verify better than the euro. there really isnt a huge disparity between the two in verification scores or the gfs just wouldnt be used or very heavily overhauled. in the end though if i were to put stock in a forecast i would use the euro but the gfs wouldnt be too far behind. Link to comment Share on other sites More sharing options...
Ground Scouring Posted November 2, 2012 Share Posted November 2, 2012 Funny how people are praising the Euro, but have we forgotten Debbie? Does nobody care about this because Debbie didn't have the impact? The euro is a good model, but it has flaws like every model. It's also wise to use ensembles guidance beyond 5 or 6 days out, not necessarily deterministic guidance. I hope Sandy doesn't cause people to just focus on the op runs. I would definitely agree with the bolded and also argue that, in the case of Sandy, the ECMWF ensembles (which actually performed better in regard to the tropics than the operational did all year) as well as the operational latched on to the track much earlier because the pattern was much more amplified and the wavelengths more frequent than they were most of the hurricane season. Excluding its coup on Sandy, the ECMWF clearly lost the medium-range battle to the GFS in nearly every other case this hurricane season. Given the increasingly sophisticated upgrades to the MAVS and other areas of the background modeling, the GFS should continue to improve its medium-range forecasting in situations that the ECMWF might have conceivably reigned supreme just a few years ago. Link to comment Share on other sites More sharing options...
Msalgado Posted November 2, 2012 Share Posted November 2, 2012 No one is going to trust a 7 day forecast for any meaningful decision anyway. It just doesn't make sense to compare the strength of models to predict in the medium range unless we're actually going to put the medium range to work making meaningful decisions and in that case neither model is going to make the grade. The NHC forecast on Sandy 5 days out was fantastic and that is the bottom line. Link to comment Share on other sites More sharing options...
Hoosier Posted November 2, 2012 Share Posted November 2, 2012 The ggem caught onto the scenario pretty early, Granted it wasnt as consistent as the euro as to where it would make landfall but it still had it early. GGEM might've even had it into the US a run or two before the Euro. Link to comment Share on other sites More sharing options...
mnweather Posted November 2, 2012 Share Posted November 2, 2012 GGEM might've even had it into the US a run or two before the Euro. Im thinking the same thing. Hopefully, someone might be able to confirm that for us. Link to comment Share on other sites More sharing options...
Shsg Falls Posted November 2, 2012 Share Posted November 2, 2012 The ECMWF has been a statistically superior model for over 15 years, hasn't it? I remember when the ECMWF gave more accurate output in the 1990s and the NMC (today's HPC) didn't even mention it's existence. Better late than never, I suppose. I enjoy reading about the ECMWF superiority in the mainstream media. However, as is typical the journalism has been lacking. ECMWF is a intergovernmental organization who's only mission is the production of medium range/global (and now seasonal) numerical model output. They are not responsible for the diversity of products and services provided by the NWS and NCEP. If 100% of the NCEP budget was theoretically dedicated to only medium range numerical output, I would think they could produce a superior model to the ECMWF. However, with a finite budget and a diverse array of products and responsibilities I don't think it is possible to seriously compete with ECMWF. Since NCEP and ECMWF have different goals I'm not sure a comparison is fair. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.