bobbutts Posted February 11, 2013 Share Posted February 11, 2013 Lack of computing power is a crap excuse IMO when you consider Moore's law, so it's good to hear about upgrades taking place. Link to comment Share on other sites More sharing options...
Clinch Leatherwood Posted February 12, 2013 Share Posted February 12, 2013 GFS did have that nice QPF blob over LI extending north so it saw something. It also hinted at the more progressive pattern..otherwise the Cape snow they had on Saturday would ahve been ours in ern MA. So it wasn't it's greatest performance by any means..but it did hint at some thing that occurred. It's all how you use guidance. Even the lousier performing model like the GFS hinted at a more progressive situation which blended in with the aggressive models...gave you a great forecast. When you look at the GFS QPF output...taking into consideration the further NNW ML track it was pretty good. Oceanwx up at GYX made the point about the MLs...looking at it, shift that NW 100 miles and it's pretty much what happened. I'm still bummed it was a little more progressive and that the Euro missed that. Only because the snows we had here Saturday would have been 100 miles west, the band just off PTown all day would have been over me and EMA...and we all would have had the #1 storm. Link to comment Share on other sites More sharing options...
OceanStWx Posted February 12, 2013 Share Posted February 12, 2013 Lack of computing power is a crap excuse IMO when you consider Moore's law, so it's good to hear about upgrades taking place. Well the computing power is there, but the question is can we afford it. Obviously that answer is yes, but the problem is that there is very little support to invest in computing power when it is seen as adequate right now. There are also issues with effect use of resources. The ECMWF prefers to go with two op runs a day, in order to have more time to devote to a superior ensemble product. 51 members uses a lot of the time and resources. The end result is that we have some adequate models, but we are not leading the pack by any means. Link to comment Share on other sites More sharing options...
Mr. Windcredible! Posted February 12, 2013 Author Share Posted February 12, 2013 An interesting blog post on the lack of computing resources available for the operational models at the NWS: http://cliffmass.blogspot.com/2013/02/the-us-weather-prediction-computer-gap.html?m=1 Link to comment Share on other sites More sharing options...
CoastalWx Posted February 12, 2013 Share Posted February 12, 2013 That guy loves to bash NCEP. He definitely has a bias. Link to comment Share on other sites More sharing options...
CoastalWx Posted February 12, 2013 Share Posted February 12, 2013 Although he is 110% right about the resources spent to chase around climate change. Put it into something we know about like weather prediction. Link to comment Share on other sites More sharing options...
CoastalWx Posted February 13, 2013 Share Posted February 13, 2013 Link to comment Share on other sites More sharing options...
Ginx snewx Posted February 13, 2013 Share Posted February 13, 2013 LOL absolutely perfect, just pay the Euro's scrape ours and save money Link to comment Share on other sites More sharing options...
Ginx snewx Posted February 13, 2013 Share Posted February 13, 2013 Nam banding site was really good at T -12 hrs. Link to comment Share on other sites More sharing options...
bobbutts Posted February 13, 2013 Share Posted February 13, 2013 Well the computing power is there, but the question is can we afford it. Obviously that answer is yes, but the problem is that there is very little support to invest in computing power when it is seen as adequate right now. There are also issues with effect use of resources. The ECMWF prefers to go with two op runs a day, in order to have more time to devote to a superior ensemble product. 51 members uses a lot of the time and resources. The end result is that we have some adequate models, but we are not leading the pack by any means. Also it appears that the EC is offsetting at least some of the cost by selling access to the results (I lack of detailed knowledge on this). If the only return is an improved forecast, it's harder to quantify vs. direct fees. I'd imagine improved GFS forecasts would have significant monetary value, it's difficult/impossible to calculate though. Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 13, 2013 Share Posted February 13, 2013 The way I see it, this winter the GFS has performed similar to the last 5 years of Patriot's football - owned the regular season and then lose in the playoffs. ...so to speak. The operational GFS scored repeatedly better than the operataional ECMWF for well over a month's worth of middle and extended ranges, during a time which was ear-marked by anomalously fast, progressive flow. The flow characteristic situated perfectly into the GFS native bias, which happens to be longitudinal in nature. But ... slow the flow down even a little, which is what/when the recent blizzard took advantage, and even down to the last second the GFS screwed the pooch and tried to "stretch"/speed up the system too much. It's like within one hour of ingesting data it takes off like a hare and tries run the flow up England's arse - woh woh woooah, nice shootin' tex! It's so annoying too - it's so patently obvious that the GFS has this stretching speed up bias; why isn't fixed? Maybe it is an emergent property that cannot be resolved by tweaking physics; in other words, the physics are clad, they need better input data. Hm, interesting. Link to comment Share on other sites More sharing options...
Tropopause_Fold Posted February 13, 2013 Share Posted February 13, 2013 Haha Tip - nice three amigos reference Link to comment Share on other sites More sharing options...
SR Airglow Posted March 8, 2013 Share Posted March 8, 2013 Another big storm, another big bust by the GFS. Even at 0z last night when it was obvious that it was going to be big, it refused to cave... Discuss it's failures... -skisheep Link to comment Share on other sites More sharing options...
Baroclinic Zone Posted March 10, 2013 Share Posted March 10, 2013 Can we at least say the GFS had the original N idea in the day 5-7 range? The Euro latched on in the day 3-4 range. Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 The resources won't be allocated until people see science funding as a worthwhile endeavor. and btw, I believe most of ecmwf's money comes from private sources...so it's not really an equal playing field. Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 Although he is 110% right about the resources spent to chase around climate change. Put it into something we know about like weather prediction. how about we just do both? this shouldn't be an either/or scenario. Link to comment Share on other sites More sharing options...
mattb65 Posted March 10, 2013 Share Posted March 10, 2013 Can we at least say the GFS had the original N idea in the day 5-7 range? The Euro latched on in the day 3-4 range. I'm going to borrow a post from DT on facebook. I think the GFS was "right" in the 5-7 day range by accident, synoptically it was very wrong (so was the Euro in that time frame) From Facebook... Here's what DT had to say in response to someone who made a similar argument, and I agree with DT: First the operational GFS and the GFS ensemble mean took the coastal Low across southeastern Virginia and the Delmarva and that in turn bought the snow EARLY into Philly NYC CT and New England. But that did not happen. The GFS was saying that the snow would advance up from Washington, DC Maryland and Virginia which of course did not happen. Instead the Low stayed south.. and turned NE very late which is why the heavy precip did NOT move into NJ NYC and Philly yesterday You are suggesting that because the GFS had the snow first via a track which did not occur that the GFS should get credit. Sorry I dont follow that at all . In addition as people in New Jersey Philadelphia New York will tell you yesterday and no significant snow fell so the GFS was early and should not get points for that either. As I stated in my last forecast for New England it is the upper level shortwave... dropping through the Great Lakes over the middle Atlantic states which has caused the coastal Low to turn northeast close to Cape Cod and bring in a heavy snow back into New England and southeastern interior New York. The snow has come in FROM east driving westward towards the Hudson River Valley and New York City ... and not from the south. http://www.facebook.com/WxRisk?fref=ts Link to comment Share on other sites More sharing options...
CoastalWx Posted March 10, 2013 Share Posted March 10, 2013 how about we just do both? this shouldn't be an either/or scenario. Because we don't have the money for both. Let the money we have go into something we have a better understanding with..like atmospheric science to help save lives and property. The climate change stuff can stay in the private sector or university funding...AKA UAH. Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 Because we don't have the money for both. Let the money we have go into something we have a better understanding with..like atmospheric science to help save lives and property. The climate change stuff can stay in the private sector or university funding...AKA UAH. we do have money to do both...NOAA gets an absolute pittance of tax payer money. Link to comment Share on other sites More sharing options...
ROOSTA Posted March 10, 2013 Share Posted March 10, 2013 Moving forward I suggest that removing the title and time (in effect) not knowing which model is which. Sure would remove the debate of which is the best. Ah, that would be the EURO in many aspects but not always. Be easier watching trends and middle blend "all them together" Yeah model bias plays a major roll in formulating a forecast. Model hugging is never a good idea, proven again and again. Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 Hi, Cliff. Two points:- First, when you plot the capacity for Titan, is that the whole machine or just the allocation granted to NOAA? The machine is not a dedicated climate machine, but is also to be used for nuclear reactor design, superconductivity, and other DOE projects. Also, although we at GFDL (NOAA's climate modeling center) have been granted an allocation on Titan, we can't make much use of it! Much of the new upgrades to Titan are Graphical Processing Units (GPUs) which our current models cannot make use of. Already DOE has expended a significant amount of resources to make one of NCAR's new climate models to run on a GPU system...and the version GPU version is already lagging behind the scientific development of the traditional CPU version. ahh...looks like the article made climate change research the boogie man. Link to comment Share on other sites More sharing options...
CoastalWx Posted March 10, 2013 Share Posted March 10, 2013 we do have money to do both...NOAA gets an absolute pittance of tax payer money. Well not the way NOAA puts it and from what I hear. That gets into another issue though..so not going there. I'm not an NCEP basher because those guys and ESRL have developed some good mesoscale models. Some of the money has gone into that...and we've made great strides with modeling convection. I don't think the UK has anything close to the mesoscale models we have, but I could be wrong. Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 Well not the way NOAA puts it and from what I hear. That gets into another issue though..so not going there. I'm not an NCEP basher because those guys and ESRL have developed some good mesoscale models. Some of the money has gone into that...and we've made great strides with modeling convection. I don't think the UK has anything close to the mesoscale models we have, but I could be wrong. what do u mean? NOAA says they get plenty of money from the government? Link to comment Share on other sites More sharing options...
SR Airglow Posted March 10, 2013 Share Posted March 10, 2013 Can we at least say the GFS had the original N idea in the day 5-7 range? The Euro latched on in the day 3-4 range. Agreed, the GFS actually beat the EURO with the overall setup in the beginning, but, when it came to the endgame, they couldn't get it done. Tip's Patriots analogy works very well for the GFS and this storm... -skisheep Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 NESDIS gets the most money out of the NOAA pie, followed by the NWS. Fisheries, Ocean Service, and Office of Oceanic and Atmospheric Research get less. Link to comment Share on other sites More sharing options...
CoastalWx Posted March 10, 2013 Share Posted March 10, 2013 what do u mean? NOAA says they get plenty of money from the government? Well funding to NWS and what not is getting reduced and that has been part of the argument, no? Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 Well funding to NWS and what not is getting reduced and that has been part of the argument, no? well yeah I don't agree with that but that has to do with the politicians and the general public not NOAA's allocation of resources. Link to comment Share on other sites More sharing options...
CoastalWx Posted March 10, 2013 Share Posted March 10, 2013 well yeah I don't agree with that but that has to do with the politicians and the general public not NOAA's allocation of resources. It just sounds like the way the money is being allocated...we don't have the money for developing something similar to what the ECMWF have in terms of computing power. That's what I mean by not getting into the details because politics are involved unfortunately. I do think the GFS stuff is flying in the face of the good developments in modeling that NCEP and ESRL has done regarding short term hi res models. Link to comment Share on other sites More sharing options...
OSUmetstud Posted March 10, 2013 Share Posted March 10, 2013 It just sounds like the way the money is being allocated...we don't have the money for developing something similar to what the ECMWF have in terms of computing power. That's what I mean by not getting into the details because politics are involved unfortunately. I do think the GFS stuff is flying in the face of the good developments in modeling that NCEP and ESRL has done regarding short term hi res models. ECMWF has a tremendous amount of money because it's privately funded. I'm not sure it's even reasonable to hope that we can get to ecmwf level with government funding. I just don't see how climate change is a real issue with this...since the NWS gets more money than that branch...and that article was dismantled to some degree since those computers are not really computers devoted to climate change...they have many other jobs as well....as I posted earlier. Link to comment Share on other sites More sharing options...
TheSnowman Posted March 10, 2013 Share Posted March 10, 2013 Not only did the GFS Nail this storm, this storm WAS the GFS's. Yes when it got to the middle range, it fell apart, but that's not its strongpoint. Which is why I'm so confused at why I was so bullish on this storm about 12" - 20" solidly from Monday on and hardly anyone else was. The GFS had this on Saturday night (Video to prove and I had NEVER EVER Filmed myself looking at a model, and I caught the biggest surprise ever and Called it! Cause I said "could be the biggest surprise") and when the Best Long Term model has something no one else does, go with it, then in the middle range the EURO latched on to a big hit. It was primed for a big hit. This was a huge coup to the GFS. Basically 96-120 hours out. We just have to realize it is the best for long term, and not much else. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.