leemhoc Posted February 14, 2013 Share Posted February 14, 2013 Also new updated ggem showed a wetter solution than euro, even though running 4dvar as well. Link to comment Share on other sites More sharing options...
SR Airglow Posted February 14, 2013 Share Posted February 14, 2013 UKIE was the model that held this when everything else got rid of it a few days ago, so if it looses it at 0z, than it's goodnight to this storm. If it keeps it, it keeps me intrested. Going to need the EURO to be convinced though. -skisheep Link to comment Share on other sites More sharing options...
bluewave Posted February 14, 2013 Share Posted February 14, 2013 I haven't been around to look at anything for this Saturday potential, but regarding the storm last night, the GFS definitely beat the Euro overall on its forecasts the past several days. 3" totals were recorded as far north as Morris County NJ. Widespread 1-3" across eastern PA and portions of north and south Jersey. It was not up to some of the GFS runs that had 3-6" for PA/NJ, but I consider it much closer than the Euro's run after run of no precip north of the Mason Dixon line. The Euro came on board with a light event up here which was much different than some of the really juiced GFS and NAM runs. While the GFS was ahead of the Euro for showing measurable precip, it's over the top depictions outweighed any gains from seeing it first. I would rather start out dry than bump to a light event then haveto back away from a forecast for heavier snows. Link to comment Share on other sites More sharing options...
Isotherm Posted February 14, 2013 Share Posted February 14, 2013 The 18z runs are no worse than 12z. The fact that they're drastically different than 12z doesn't tell you that the event has "no chance" per se, but that the event is highly uncertain / questionable. Obviously there is some potential given the data was previously showing snow for the area. Will it happen? One run is too soon to make a call but if 00z data remains consistent w/ the 18z runs then the potential is likely shot. Link to comment Share on other sites More sharing options...
Edge Weather Posted February 14, 2013 Share Posted February 14, 2013 UKIE was the model that held this when everything else got rid of it a few days ago, so if it looses it at 0z, than it's goodnight to this storm. If it keeps it, it keeps me intrested. Going to need the EURO to be convinced though. -skisheep I agree, I am going with the UKMET due to consistency with this threat. If it loses it, then game over. But I am riding the UKMET until then. My gut instinct is that the Euro is messing this up again like it did the last one, as it does not do well with the northern stream dominated storms in my experience. The reminds me too much of December 26, 2010 when the Euro was the last one to pick up the storm in the short range. And the GFS simply does not have the balloon data this run. We shall see. Been an interesting discussion. Let's see what happens. Link to comment Share on other sites More sharing options...
donsutherland1 Posted February 14, 2013 Share Posted February 14, 2013 This is from an accuweather post about forecast models 2. Look at model trends. If the low pressure moved east with this run, what did it do the run before that? For the GFS, look at a couple days of 00Z and 12Z runs for consistency. Avoid the 06Z and 18Z runs when 00 or 12 is available; in the U.S. these runs don't include the weather balloon network data (balloons are only sent up twice per day), and are therefore radically different and more likely to have bias. I disagree with that assessment. Balloon data is only a small slice of overall data. Rarely is the balloon data by itself sufficiently important to lead to material differences in output. The absence of significant differences in verification scores speaks for itself. Link to comment Share on other sites More sharing options...
Isotherm Posted February 14, 2013 Share Posted February 14, 2013 Completely subjective. Your personal feelings dont determine which model sucked less. Agree. We can't really quantify the situation. Maybe someone should come up with a method, "If a model leads us on with a snowstorm, but then backs off slightly at game time, it gets 7 points. If a model doesn't see an event until late in the game, but then forecasts better even though it saw it 3 days late, 12.34 points awarded" Link to comment Share on other sites More sharing options...
bluewave Posted February 14, 2013 Share Posted February 14, 2013 Completely subjective. Your personal feelings dont determine which model sucked less. Don't take my word for it, look at the model scores for yourself. http://cliffmass.blogspot.com/2013/02/the-us-weather-prediction-computer-gap.html Link to comment Share on other sites More sharing options...
Isotherm Posted February 15, 2013 Share Posted February 15, 2013 Don't take my word for it, look at the model scores for yourself. http://cliffmass.blogspot.com/2013/02/the-us-weather-prediction-computer-gap.html The score chart is irrelevant to the recent storm. Since Philly averages around 22" of snow per winter, and Trenton about 25", does that mean every winter should see Trenton snowier than Philly? No. Although they're on average a slightly snowier location, Philly has had numerous storms w/ snowfall totals greater than that of TTN, and even seasonal totals. Similarly, while the ECMWF has the best scores on average, that does not preclude a bad stretch of calls from happening or a period in which it performs less consistently than the GFS. Link to comment Share on other sites More sharing options...
Isotherm Posted February 15, 2013 Share Posted February 15, 2013 The first 2 paragraphs negate your previous argument that you'd rather a model show no storm. And im not debating overall scores, put that strawman away. YOU declared the Euro the winner for last nights "event" based on completely subjective nonsense. Agreed. Link to comment Share on other sites More sharing options...
Storm At Sea Posted February 15, 2013 Share Posted February 15, 2013 "Off" runs of the NAM/GFS are good for one thing...suggesting trends between and bridging the 12z and 00z runs. It's a good thing that many of you weren't involved in amateur forecasting in the late 1990's. The run-to-run differences were much more extreme. Link to comment Share on other sites More sharing options...
SRRTA22 Posted February 15, 2013 Share Posted February 15, 2013 "Off" runs of the NAM/GFS are good for one thing...suggesting trends between and bridging the 12z and 00z runs. It's a good thing that many of you weren't involved in amateur forecasting in the late 1990's. The run-to-run differences were much more extreme. I agree with this. 18z/6z were always more of a bridge between 00z and 12z. Link to comment Share on other sites More sharing options...
killabud Posted February 15, 2013 Share Posted February 15, 2013 As an utter lurker / not a poster, it's not really my place to argue this, but I will make one observation... When I first started lurking here a couple years ago (actually longer than that, because I used to post on the "previous site" on occasion) the first thing I noticed that confused me a little was that every time the 18z models didn't agree with the wish consensus there was invariably a litany of posts along the lines of "it's the 18z, throw it out". So people suggesting as much here comes as no surprise to me - it's always happened. I just never understood the logic behind it. but when they "threw it out" they did it for a reason,not just because it wasn't what they wanted,i remember people on here would "toss" snowy 18z runs as often as bad ones i just feel like people have been "forecasting" with to much emotion these last two winters Link to comment Share on other sites More sharing options...
FPizz Posted February 15, 2013 Share Posted February 15, 2013 Opinions without backing of any sort are the issue. Bingo! Link to comment Share on other sites More sharing options...
Capt. Adam Posted February 15, 2013 Share Posted February 15, 2013 What happened with the 18z runs that made this entire argument go down? NAM and GFS that were heavy on precip yielded less. Not a total breakdown in the synoptics (in fact some would say upper level improvements), just less QPF "IMBY" so off the bandwagon many jump. Link to comment Share on other sites More sharing options...
Storm At Sea Posted February 15, 2013 Share Posted February 15, 2013 There were several juicy GFS runs, but in general, it had the overall idea pretty consistently of a 2-4 or 3-5 in. event for many areas in NJ. The Euro was dead last in terms of laying down any precip. YOU might prefer that a model not "tease" you with false hope, but a model that is off by 0.2-0.3 several days in advance while another show nada, wins the battle. May I also add that I consistently stated that BL temps were going to be problematic. There was plenty of wasted precip in the form of rain and PL. Once snow started falling, much of it was wasted as well. It was in the upper 40's for heaven's sake. It took time for snow to accumulate even on grass surfaces. The Euro came on board with a light event up here which was much different than some of the really juiced GFS and NAM runs. While the GFS was ahead of the Euro for showing measurable precip, it's over the top depictions outweighed any gains from seeing it first. I would rather start out dry than bump to a light event then haveto back away from a forecast for heavier snows. Link to comment Share on other sites More sharing options...
earthlight Posted February 15, 2013 Author Share Posted February 15, 2013 SREFs went the wrong way for sure...way less amped up compared to 15z. Link to comment Share on other sites More sharing options...
famartin Posted February 15, 2013 Share Posted February 15, 2013 I suspect the 0Z runs will be telling as to whether weather balloons really had an effect or if it was just some fluke data that "messed up" the 12Z runs. If the 0Z runs amp back up, then you can believe that there is some balloon data which is likely feeding the frenzy; otherwise, probably some random bad data which wasn't obvious got into the 12Z runs and has since been smoothed out/eliminated. Link to comment Share on other sites More sharing options...
famartin Posted February 15, 2013 Share Posted February 15, 2013 There were several juicy GFS runs, but in general, it had the overall idea pretty consistently of a 2-4 or 3-5 in. event for many areas in NJ. The Euro was dead last in terms of laying down any precip. YOU might prefer that a model not "tease" you with false hope, but a model that is off by 0.2-0.3 several days in advance while another show nada, wins the battle. May I also add that I consistently stated that BL temps were going to be problematic. There was plenty of wasted precip in the form of rain and PL. Once snow started falling, much of it was wasted as well. It was in the upper 40's for heaven's sake. It took time for snow to accumulate even on grass surfaces. Depends on where you were, of course. NYC area was really ripped off QPF wise, with only 0.03" at EWR and NYC, 0.02" at LGA and 0.01" at JFK. That was used up just cooling the BL off. Further south, much more precip fell (almost a half inch at ACY) but of course a lot got used up cooling the BL and then you still had issues with very low ratios due to surface temps just above freezing. There were lucky spots in far western NJ and interior SE PA where elevation helped decrease the BL issues and they had better ratios. Link to comment Share on other sites More sharing options...
ace0927 Posted February 15, 2013 Share Posted February 15, 2013 SREFs went the wrong way for sure...way less amped up compared to 15z. SREFs went the wrong way for sure...way less amped up compared to 15z. yup - in line with all the rest of the guidance since the euro no euro, no event...take that to the bank... Link to comment Share on other sites More sharing options...
eduggs Posted February 15, 2013 Share Posted February 15, 2013 SREFs went the wrong way for sure...way less amped up compared to 15z. I think they pretty consistently lag the previous NAM run. Link to comment Share on other sites More sharing options...
ace0927 Posted February 15, 2013 Share Posted February 15, 2013 I think they pretty consistently lag the previous NAM run. well the ARW's were the members that were producing....i guess that model and its members woke up Link to comment Share on other sites More sharing options...
SR Airglow Posted February 15, 2013 Share Posted February 15, 2013 SREFs went the wrong way for sure...way less amped up compared to 15z. They're based off the NAM, correct? If so, not suprised given the 18z NAM. 0z is when we see the trend back to something nice start if we want any hope of getting decent snow out of this. -skisheep Link to comment Share on other sites More sharing options...
96blizz Posted February 15, 2013 Share Posted February 15, 2013 Are we fast moving to a Euro only world? For tracking purposes, that would be sad. No single model can nail it, although the Euro is coming close, but that doesn't mean there isn't room for catch up and hard work for the rest! Link to comment Share on other sites More sharing options...
Isotherm Posted February 15, 2013 Share Posted February 15, 2013 The first real run of importance is the 00z NAM. The collective thought here should be expect nothing but hope for the best. Link to comment Share on other sites More sharing options...
snywx Posted February 15, 2013 Share Posted February 15, 2013 I think they pretty consistently lag the previous NAM run. Yeah I have noticed that.. usually the next SREF run follows the previous NAM solution. If the 0z NAM is amped up expect 3z SREFs to be amped as well.. Link to comment Share on other sites More sharing options...
SR Airglow Posted February 15, 2013 Share Posted February 15, 2013 NAM is running, good luck and think snow! -skisheep Link to comment Share on other sites More sharing options...
Dan76 Posted February 15, 2013 Share Posted February 15, 2013 http://cliffmass.blogspot.com/2013/02/the-us-weather-prediction-computer-gap.html?spref=tw Pretty intresting Link to comment Share on other sites More sharing options...
Storm At Sea Posted February 15, 2013 Share Posted February 15, 2013 At 21 HR, the H5 trof is less positively tilted which is good. Precip over Appalachia is breaking out but not as intense as 18z. Link to comment Share on other sites More sharing options...
nyblizz44 Posted February 15, 2013 Share Posted February 15, 2013 http://cliffmass.blogspot.com/2013/02/the-us-weather-prediction-computer-gap.html?spref=tw Pretty intresting I agree. very interesting read Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.