mdsnowlover Posted January 7, 2012 Share Posted January 7, 2012 Excellent anzwer and thank you. Im fsmiliar with the spaghetti plots. Link to comment Share on other sites More sharing options...
Kmlwx Posted January 7, 2012 Share Posted January 7, 2012 How can the ensembles look so different from the op run?? I realize there is smoothing done onthe ensembles. Thanks Because they are all running slightly different scenarios. It isn't the exact same model run over and over again. EDIT: Great answer above lol Link to comment Share on other sites More sharing options...
mdsnowlover Posted January 7, 2012 Share Posted January 7, 2012 Thanks kmlwx, i knew they were run eith slightly different initial beginning conditions Link to comment Share on other sites More sharing options...
Kmlwx Posted January 7, 2012 Share Posted January 7, 2012 Thanks kmlwx, i knew they were run eith slightly different initial beginning conditions They have kicked out some interesting solutions in the past that do not match the OP at all - certainly can be cause for some head scratching. Link to comment Share on other sites More sharing options...
Ian Posted January 7, 2012 Share Posted January 7, 2012 Hopefully it never snows Link to comment Share on other sites More sharing options...
Bob Chill Posted January 7, 2012 Share Posted January 7, 2012 Hopefully it never snows Your reverse psychology has been pretty much fail this winter. You may want to get back on the weenie train. Link to comment Share on other sites More sharing options...
Kmlwx Posted January 7, 2012 Share Posted January 7, 2012 Hopefully it never snows Can we vote to ban an admin? Link to comment Share on other sites More sharing options...
Wonderdog Posted January 7, 2012 Share Posted January 7, 2012 Hour 168 looks promising on the 12z GFS. Link to comment Share on other sites More sharing options...
RTutton Posted January 7, 2012 Share Posted January 7, 2012 The 12 GFS wants to give much of VA a bout of extended light precip next Friday night into Saturday. Can anyone tell if surface temps look good for Richmond, Fredericksburg, C'ville, Fairfax, etc? Link to comment Share on other sites More sharing options...
WinterWxLuvr Posted January 7, 2012 Share Posted January 7, 2012 It's called Chaos. On the right hand side of the plot below you'll see lots of lines. They represent various contour lines from the 500mb forecast from the ensemble members. The white lines show the control, essentially the operation but I think it is run with the same resolution as the ensemble members (DTK can correct me if I'm wrong). all the ensemble members are run with slightly different initial conditions than their counterparts or the operational. Those differences in the initial conditions grow during the run. BY 240 hours most of the members have a different idea of the evolution of the pattern than the control (operational). The ensemble mean (the average of all the members) therefore looks much different than the control because averaging smooths features but also because the operational is an outlier. It still could be correct but usually the ensemble mean will beat it at those time ranges. Hope I've semi answered your question. It sure looks like pretty good agreement with the spaghetti plots over the eastern parts of canada and the us. Seems to me that those things are usually all over the place. Link to comment Share on other sites More sharing options...
usedtobe Posted January 7, 2012 Share Posted January 7, 2012 The 12 GFS wants to give much of VA a bout of extended light precip next Friday night into Saturday. Can anyone tell if surface temps look good for Richmond, Fredericksburg, C'ville, Fairfax, etc? It's a 168 hr forecast so there is not much certainty in anything. It's not worth worrying about surface temps this early. Link to comment Share on other sites More sharing options...
Yorkpa Posted January 7, 2012 Share Posted January 7, 2012 Anything pertaining to snow in the northeast will not happen. Unless it's shown Less then 24 hours out. Fed up with this winter already. Hour 168 = who cares Link to comment Share on other sites More sharing options...
RTutton Posted January 7, 2012 Share Posted January 7, 2012 It's a 168 hr forecast so there is not much certainty in anything. It's not worth worrying about surface temps this early. Good point. As to not caring anything that happens at hour 168, at least it's better than a storm on the low res at 384 Link to comment Share on other sites More sharing options...
usedtobe Posted January 7, 2012 Share Posted January 7, 2012 It's a 168 hr forecast so there is not much certainty in anything. It's not worth worrying about surface temps this early. I think it too early to really worry much about the 168 hr light overrunning event but in this type of pattern that type of scenario is probably the easiest way to get snow. Still the next run of the GFS will undoubtedly be different so don't yet get your hopes up. Link to comment Share on other sites More sharing options...
WxUSAF Posted January 7, 2012 Share Posted January 7, 2012 12z GFS yesterday looked like a nice wintry run. Since then, it's been 4 runs of UGLY. Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 12z GFS yesterday looked like a nice wintry run. Since then, it's been 4 runs of UGLY. 12z UKMET, GGEM and ECMWF all have the precip Mon-Tues further north and a more impressive s/w. And that's just the first potential.... 1/12-15 is the next one. Link to comment Share on other sites More sharing options...
dtk Posted January 7, 2012 Share Posted January 7, 2012 How can the ensembles look so different from the op run?? I realize there is smoothing done onthe ensembles. Thanks Because they use a different version of the model, at degraded resolution, typically with a different set of physical parameterizations, and perturbed initial conditions. Link to comment Share on other sites More sharing options...
mitchnick Posted January 7, 2012 Share Posted January 7, 2012 12z UKMET, GGEM and ECMWF all have the precip Mon-Tues further north and a more impressive s/w. And that's just the first potential.... 1/12-15 is the next one. you ought to post more Link to comment Share on other sites More sharing options...
dtk Posted January 7, 2012 Share Posted January 7, 2012 It's called Chaos. On the right hand side of the plot below you'll see lots of lines. They represent various contour lines from the 500mb forecast from the ensemble members. The white lines show the control, essentially the operation but I think it is run with the same resolution as the ensemble members (DTK can correct me if I'm wrong). all the ensemble members are run with slightly different initial conditions than their counterparts or the operational. Those differences in the initial conditions grow during the run. BY 240 hours most of the members have a different idea of the evolution of the pattern than the control (operational). The ensemble mean (the average of all the members) therefore looks much different than the control because averaging smooths features but also because the operational is an outlier. It still could be correct but usually the ensemble mean will beat it at those time ranges. Hope I've semi answered your question. For all intents and purposes this is correct. The ensemble "control" run is initialized from a truncation version of the operational GFS analysis, but it is run at the ensemble resolution. Right now, the GDAS/GFS run at T574 with L64, whereas the ensemble is run at T190 L28. These numbers refer to the spectral truncation of the model (T; [the horizontal resolution roughly corresponds to 27km or so for the operational, and something like 90km for the ensembles] ) and the number of vertical levels (L). Also, the version of the forecast model that is used in the GFS/GDAS is actually different than the version used in the GEFS (this is for partly practical and partly scientific reasons). Of course, the ensemble members themselves are initialized from perturbed initial conditions (to attempt to account for uncertainty in the analysis state). Furthermore, to attempt to account for having an imperfect model, pseudo random forcing is applied to the model tendencies for each of the GEFS members (this helps increase the diversity/spread and sample more realistic uncertainty). It should be noted that our ensemble forecast system is in general quite under-dispersive. Link to comment Share on other sites More sharing options...
Ian Posted January 7, 2012 Share Posted January 7, 2012 12z UKMET, GGEM and ECMWF all have the precip Mon-Tues further north and a more impressive s/w. And that's just the first potential.... 1/12-15 is the next one. mon-tue is going to have some trouble not being rain if it happen Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 mon-tue is going to have some trouble not being rain if it happen Yep, it is at best a cold rain verbatim on the models. Maybe the CMC is snow in MD. The ECMWF sneaks the precip now into C NJ and SE MA. The thing that will help you guys out in MD is the timing on Monday AM. Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 The damn 12z ECMWF and GFS reversed their solutions from yesterday. Now the ECMWF breaks the southern low off into the E PAC while the GFS brings it out. Therefore, the GFS is more threatening Jan 13-14 while the ECMWF has more of an Arctic Front wave that misses the Mid-Atlantic. If the STJ low doesn't come out into the CONUS, then this threat will be limited to the Northeast. Link to comment Share on other sites More sharing options...
usedtobe Posted January 7, 2012 Share Posted January 7, 2012 mon-tue is going to have some trouble not being rain if it happen I don't think for us there is much hope for snow. The 168 hr event is more interesting but could easily also end up warmer if the wave were stronger. Without blocking that low near Nova scotia is likely to be transient. Without it we're a little warmer. Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 I don't think for us there is much hope for snow. The 168 hr event is more interesting but could easily also end up warmer if the wave were stronger. Without blocking that low near Nova scotia is likely to be transient. Without it we're a little warmer. I think the issue with the Jan 13-14 setup has more to do with the southern stream low that the GFS wants to bring out and phase while the ECMWF wants to leave in the E PAC (exact opposite of their 12z runs yesterday). The temporary heights lowered in E Canada will actually simulate a 50-50 low that is moving north as a s/w is amplifying. The cold air issue is more for Monday than the next one. I didn't check closely yet (outside of 2m and h85), but is the 12z ECMWF snow in Maryland on Mon AM? Link to comment Share on other sites More sharing options...
Ian Posted January 7, 2012 Share Posted January 7, 2012 The cold air issue is more for Monday than the next one. I didn't check closely yet (outside of 2m and h85), but is the 12z ECMWF snow in Maryland on Mon AM? no Link to comment Share on other sites More sharing options...
Ian Posted January 7, 2012 Share Posted January 7, 2012 maybe near the M/D line precip doesnt even make it to dc by 18z or it's just making it Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 maybe near the M/D line Thanks Matt. Based on what I saw from the 2m/h85 0C line, I would have guessed NW suburbs (immediate) of DCA were seeing some very light snow Monday AM. Looks like even with the ECMWF solution, there are snow showers galore with the cold shot at the end of the week. Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 precip doesnt even make it to dc by 18z or it's just making it That's not true. You've accumulated some light precip by 18z Link to comment Share on other sites More sharing options...
HM Posted January 7, 2012 Share Posted January 7, 2012 The 12z UKMET seems like a compromise between the GFS and ECMWF both with the Monday situation and then again with the second southern stream low (UKMET brings it out but drags it along more than the GFS). Here is the short-term UKIE: http://www.meteo.psu.edu/~gadomski/UKMET_12z/ukloop.html Link to comment Share on other sites More sharing options...
Ian Posted January 7, 2012 Share Posted January 7, 2012 That's not true. You've accumulated some light precip by 18z ok.. it must be like .01-.02 or something.. the .05" line is still 50 mi south or so and doesnt get to dc till 0z. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.