Typhoon Tip Posted January 6, 2011 Share Posted January 6, 2011 Wow, I'm really impressed with the 00z model agreement re the event later next week - sans the ECM. One aspect about that, which may explain the apparent earlier agreement and better behavior with the overall idea thus far, is that the flow is relaxed(ing) in the deep S prior to the trough arrival through the lower OV and TV regions. The heights over Florida have dropped to 576dm and the mean wind speeds less than 50kts is a sign of lowering resistance/shear, and this would definitely offer less overall wave interference from the L/W scales, but allow more of the centric trough dynamics to be conserved as well... Very encouraging sign that regardless of whatever uncertaintly plays out along the NE coast over the next 84 hours, the denizens may not have to wait long for something appearing more bona fide. Link to comment Share on other sites More sharing options...
powderfreak Posted January 6, 2011 Share Posted January 6, 2011 I think there is some underplaying of ratios going on. These norluns can sometimes produce some of the most prolific ratios for NE (outside of the LE /orographic stuff). When I see .5-1" of qpf for ME in an inverted trof setup, I think 12"+ Still curious as to the wsw for NYC metro and NJ. Maybe fluff factor could be enough to push totals up there... Agreed... for whatever reason these events always seem to have good ratios. Must have to do with the banded nature of norluns. I have no idea but I agree that they often produce good 15:1+ ratios. ALB mentioned this in the AFD this morning and is forecasting 18:1 ratios for the event. Link to comment Share on other sites More sharing options...
baroclinic_instability Posted January 6, 2011 Share Posted January 6, 2011 That's not an overlay. Here's the 12h 0z NAM forecast vs Init. Slight differences in strength and positioning in the two features which were entirely on land at the time (west of lakes which specifically left to its solution last night) And below that the 24h 0z forecast, and now the 12h 12z. Just a SLIGHt change with the handling of the energy coming into PA entirely due to the crap 0h product from 0z last night. Paint it however you like, but it clearly had bad data that either didn't get caught or shows up elsewhere as the GFS did not have these issues. Also a good illustration of why those verification scores aren't true to what's happening. The contours may not be moving all that much but the vorticity is, and it's being pegged poorly often in strength and position in the very, very short term. If we cannot get that right we can give up getting really accurate foreacasts long term. You make some valid points. Errors do grow rapidly in time. This is why numerical modelers/modeling community puts so much effort into data assimilation. As for the NAM suddenly sucking--this is nothing new. I posted about the NAM "Phase Shift" problem over a month ago regarding a very different setup. The NAM has not suddenly become worse, but in reality, has never been the most reliable to begin with. It can be used to good effect though when analyzed properly. http://www.americanw...tward-too-slow/ Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted January 6, 2011 Share Posted January 6, 2011 Wow, I'm really impressed with the 00z model agreement re the event later next week - sans the ECM. One aspect about that, which may explain the apparent earlier agreement and better behavior with the overall idea thus far, is that the flow is relaxed(ing) in the deep S prior to the trough arrival through the lower OV and TV regions. The heights over Florida have dropped to 576dm and the mean wind speeds less than 50kts is a sign of lowering resistance/shear, and this would definitely offer less overall wave interference from the L/W scales, but allow more of the centric trough dynamics to be conserved as well... Very encouraging sign that regardless of whatever uncertaintly plays out along the NE coast over the next 84 hours, the denizens may not have to wait long for something appearing more bona fide. Isn't your rule of thumb that you don't want heights over 580dm in FL....that is rough point at which the geopotential medium becomes compressed to the point of shearing out and surpressing cyclones... Link to comment Share on other sites More sharing options...
Clinch Leatherwood Posted January 6, 2011 Share Posted January 6, 2011 Wow, I'm really impressed with the 00z model agreement re the event later next week - sans the ECM. One aspect about that, which may explain the apparent earlier agreement and better behavior with the overall idea thus far, is that the flow is relaxed(ing) in the deep S prior to the trough arrival through the lower OV and TV regions. The heights over Florida have dropped to 576dm and the mean wind speeds less than 50kts is a sign of lowering resistance/shear, and this would definitely offer less overall wave interference from the L/W scales, but allow more of the centric trough dynamics to be conserved as well... Very encouraging sign that regardless of whatever uncertaintly plays out along the NE coast over the next 84 hours, the denizens may not have to wait long for something appearing more bona fide. You've been hammering high wind speeds down south and I'd largely agree. That's part of what is giving the models a hard time. -- OSU, again though it isn't microanalysis to look at a 12 hour prog from a short term model and see that it was off by hundreds of miles with the key vorticity. That isn't analysis, that's a bad, bad forecast for 12-24 hours. But again your 500mb verifications will show it was fine because it just misplaced the critical vorticity by 300 miles, not really the overall heights which is why your scores will continue to be decent when ground truth whacks it. The GFS has seemed less prone to these swings over the last year. For what reason I dont know, but the NAM.... Link to comment Share on other sites More sharing options...
moneypitmike Posted January 6, 2011 Share Posted January 6, 2011 Kevin is busy, so I took care of it for him... Mt. Tolland is the next county to the left, Dave. A for effor thoguth! Link to comment Share on other sites More sharing options...
OSUmetstud Posted January 6, 2011 Share Posted January 6, 2011 That's not an overlay. Here's the 12h 0z NAM forecast vs Init. Slight differences in strength and positioning in the two features which were entirely on land at the time (west of lakes which specifically left to its solution last night) And below that the 24h 0z forecast, and now the 12h 12z. Just a SLIGHt change with the handling of the energy coming into PA entirely due to the crap 0h product from 0z last night. Paint it however you like, but it clearly had bad data that either didn't get caught or shows up elsewhere as the GFS did not have these issues. Also a good illustration of why those verification scores aren't true to what's happening. The contours may not be moving all that much but the vorticity is, and it's being pegged poorly often in strength and position in the very, very short term. If we cannot get that right we can give up getting really accurate foreacasts long term. I know that...I can see the differences side by side too. I didn't feel like making an animation...sorry... The vorticity is completely dependent on the the orientation and tightness of the isoheights...so if there were an issue you would also see it in the height correlation statistic. I wish I could find verification statistics on the NAM...but I don't know where to find them. I'm just not sure how these differences at short and medium ranges are any different than what we were seeing 1, 3, 5, 10 years ago. I'd imagine they are actually, overall, smaller run-to-run. I still wouldn't be shocked if someone presented me with stats that the models have performed poorly this month, however, just due to how anomalous this pattern is. I'm not sure why you think that is a huge step back in modeling. Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted January 6, 2011 Share Posted January 6, 2011 I know that...I can see the differences side by side too. I didn't feel like making an animation...sorry... The vorticity is completely dependent on the the orientation and tightness of the isoheights...so if there were an issue you would also see it in the height correlation statistic. I wish I could find verification statistics on the NAM...but I don't know where to find them. I'm just not sure how these differences at short and medium ranges are any different than what we were seeing 1, 3, 5, 10 years ago. I'd imagine they are actually, overall, smaller run-to-run. I still wouldn't be shocked if someone presented me with stats that the models have performed poorly this month, however, just due to how anomalous this pattern is. I'm not sure why you think that is a huge step back in modeling. Exactly; I agree with your position on this 100%.....the data surplus doesn't help, either. Link to comment Share on other sites More sharing options...
ski MRG Posted January 6, 2011 Share Posted January 6, 2011 Nice and cold here at 2k. Love the model mayhem and the bickering back and forth. Snow is incoming don't need the models to see and feel it. Link to comment Share on other sites More sharing options...
Clinch Leatherwood Posted January 6, 2011 Share Posted January 6, 2011 You make some valid points. Errors do grow rapidly in time. This is why numerical modelers/modeling community puts so much effort into data assimilation. As for the NAM suddenly sucking--this is nothing new. I posted about the NAM "Phase Shift" problem over a month ago regarding a very different setup. The NAM has not suddenly become worse, but in reality, has never been the most reliable. It can be used to good effect though when analyzed properly. http://www.americanw...tward-too-slow/ Will give it a read, thanks. I think part of why we all get on one another is we're talking about different things. OSU thinks I'm referring to the trough and lack of snow. I really haven't cared much about this system because when I see such wild swings at 12-18 hours I know we're going to see wild swings in the data 60 hours down the road. I'd be happy with passing snow showers at this point. The ETA wasn't this bad. The NAM wasn't that bad prior to the last 18 months. Why I have no idea, but I've over analyzed it for years and it's never been like this specifically with vorticity placement and strength. When you can look at the water vapor and RUC 3 hours in and see that it's off the rails, not good for our "best" model. What I question is how it was so wrong at 0z 0h last night unless part of the problem is in assimiliation. It was mostly out on its own right at 0h last night. How's that happening? Link to comment Share on other sites More sharing options...
CoastalWx Posted January 6, 2011 Share Posted January 6, 2011 Exactly; I agree with your position on this 100%.....the data surplus doesn't help, either. Highly anomalous patterns will produce highly anomalous results and models will struggle. Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted January 6, 2011 Share Posted January 6, 2011 Nice and cold here at 2k. Love the model mayhem and the bobbing back and forth. Scott is incuming don't need the models to see and feel it. Oh boy Link to comment Share on other sites More sharing options...
Clinch Leatherwood Posted January 6, 2011 Share Posted January 6, 2011 I know that...I can see the differences side by side too. I didn't feel like making an animation...sorry... The vorticity is completely dependent on the the orientation and tightness of the isoheights...so if there were an issue you would also see it in the height correlation statistic. I wish I could find verification statistics on the NAM...but I don't know where to find them. I'm just not sure how these differences at short and medium ranges are any different than what we were seeing 1, 3, 5, 10 years ago. I'd imagine they are actually, overall, smaller run-to-run. I still wouldn't be shocked if someone presented me with stats that the models have performed poorly this month, however, just due to how anomalous this pattern is. I'm not sure why you think that is a huge step back in modeling. It didn't get the 0h init right from 0z last night. That doesn't have much to do with anything other than something going wrong in the process, IMO. Do you think the average person understands this is a different pattern when their forecasters can't get things right and are telling them they cannot rely on the models? Whether or not it's relevant to the pattern of the last two winters or not we will find out in time. The bigger question is why can't it forecast 6 hours out and why is it struggling at times even at init? Link to comment Share on other sites More sharing options...
OSUmetstud Posted January 6, 2011 Share Posted January 6, 2011 It didn't get the 0h init right from 0z last night. That doesn't have much to do with anything other than something going wrong in the process, IMO. Do you think the average person understands this is a different pattern when their forecasters can't get things right and are telling them they cannot rely on the models? Whether or not it's relevant to the pattern of the last two winters or not we will find out in time. The bigger question is why can't it forecast 6 hours out and why is it struggling at times even at init? no...but I assumed you weren't the average person and would understand.. I've personally never liked the NAM. I think the GFS is decent. Link to comment Share on other sites More sharing options...
HoarfrostHubb Posted January 6, 2011 Share Posted January 6, 2011 Mt. Tolland is the next county to the left, Dave. A for effor thoguth! terribke terribke geography skills Link to comment Share on other sites More sharing options...
moneypitmike Posted January 6, 2011 Share Posted January 6, 2011 Nice and cold here at 2k. Love the model mayhem and the bickering back and forth. Snow is incoming don't need the models to see and feel it. Yup, a bit nipply out there. Good news is that We're all gonna get snow. It's a good think no one would pay me for a forecast though! Looks like the start on this gets pushed way back in this run of the NAM/ More in line with EC timing? 23.2/12 Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted January 6, 2011 Share Posted January 6, 2011 Highly anomalous patterns will produce highly anomalous results and models will struggle. Couple that with sickos like you and I up until 4AM staring at every detail, and messenger comparing the pube length of the vort from each run....there you go. There are so many more models and they are run more often.....also forums like this, where people get in a circle jerk and scrutinize them. Link to comment Share on other sites More sharing options...
Damage In Tolland Posted January 6, 2011 Share Posted January 6, 2011 Can we stop the my penis is bigger than yours model talk? And get back to figuring out how much snow we're gonna get Link to comment Share on other sites More sharing options...
mahk_webstah Posted January 6, 2011 Share Posted January 6, 2011 Nice and cold here at 2k. Love the model mayhem and the bickering back and forth. Snow is incoming don't need the models to see and feel it. Pete's bones FTW! Link to comment Share on other sites More sharing options...
CoastalWx Posted January 6, 2011 Share Posted January 6, 2011 Regarding this storm, I do like how the mid level lows shifted south at 12z. The band over se NY and the Catskills could be a great one. If you notice, that band lies exactly along the pivot point of H7 and H5, and also collocated with he surface trough. All those features will support that band for several hours anyways. The NAM brings this band ne rather quickly until it reaches ME. After that, it almost has a backside deformation band that pivots in from the west. I think we still need to look out for some sort of an OES/CF thing going on in ern mass. This is why these are a b*tch to forecast. Widespread snows can be hard to find sometimes with these setups, but the NAM wasn't far from this, IF we can get the mid level lows to the south and allow for more Atlantic inflow. This will probably change again at 18z, 00z, 06z and maybe 12z again, but we can at least try and narrow down the "hot spots". Link to comment Share on other sites More sharing options...
HoarfrostHubb Posted January 6, 2011 Share Posted January 6, 2011 Oh boy Kevin will have a field day... Link to comment Share on other sites More sharing options...
moneypitmike Posted January 6, 2011 Share Posted January 6, 2011 terribke terribke geography typing skills Link to comment Share on other sites More sharing options...
Gibbsfreeenergy Posted January 6, 2011 Share Posted January 6, 2011 AccuCast the orientation/configuration will be something similar to this with the first surge of instability in the inverted trof. The NAM really had a second surge of instabilty setup from about BOS on north late Saturday into Sunday as second solid shot of strong PVA is whipped around anchored 504dam vortex in PA...which is certainly possible given all the maxes getting whipped around the closed vortex. The GFS shows a similar redevelopment as well...would think AccuMap would need additional coverage of snow in NE Mass/SE NH/Maine at the expense of some of the 3-6 coverage in eastern CT/RI/SE Mass. I really think that the initial surge of instability with the inverted trof will be impressive, and stall out east of NYC in western CT and Hudson Valley. Then redevelop up north with renewed vort max. Again these are details that will prolly change several times from here on out, so maybe I shouldnt waste me time, lol. Have to pass the day with a my first cold of the season some how I guess, lol. Link to comment Share on other sites More sharing options...
free_man Posted January 6, 2011 Share Posted January 6, 2011 Agreed... for whatever reason these events always seem to have good ratios. Must have to do with the banded nature of norluns. I have no idea but I agree that they often produce good 15:1+ ratios. ALB mentioned this in the AFD this morning and is forecasting 18:1 ratios for the event. I missed that, but exactly what I was thinking! Something to look for anyway with regards to a "surprise" snowfall Link to comment Share on other sites More sharing options...
codfishsnowman Posted January 6, 2011 Share Posted January 6, 2011 Last storm was impossible to measure. I am claiming 4.5" in my sig, but if I had tried measuring the next day I'd have been hard pressed to find much above 2". I think it's quite possible the next 60 hours puts more snow on my lawn than the bomb did, and the fact that it's an anomalous meso-banded inverted trough setup I consider a good thing. I remember when the Sox were down 0-3 to the yanks in '04 Maddog Russo was saying on the radio that if the sox were ever going to win, the prelude would have to be some weird set of circumstances, a "cataclysmic event" he termed it. In this case, I am thinking of upside as being advisory criteria, but in general I say bring the atypical setups. These run-of-the-mill bombs consistently suck around here it seems. I lived in Florence for four years and there were a bunch of good east coast bombs and ONLY ONE produced over a foot of snow and that was bc it was heavily moisture laden coming up from the gulf and it tracked well inside the bm...it poured snow in mid Mar 07 with three inch hr rates right before it mixed with sleet and then tappered off giving us 13-14 inches...interesting was that Springfield only had six or seven inches before the change over to ip.. The Jan 05 event produced about a foot but much of that fell from system snow with the clipper feature, once the low actually bombed near the bm we only had one band of heavy snow move through right before the snow ended Now those were the two best scenarios, the rest of the so called bombs that buried other parts of sne were all horrible up there with less than six inches always and many times snowfall rates that hardly approached an inch/hr at the storms height while other parts of sne saw two to four inch hr rates etc.... In fact Springfield often did somewhat better although not that much. When I lived up there I asked the locals what was the heaviest snow they had ever seen on the level in a single storm and most of the old timers said a foot and a half tops and some even said a bit over a foot! Link to comment Share on other sites More sharing options...
HoarfrostHubb Posted January 6, 2011 Share Posted January 6, 2011 This is how Blizz typed "terrible" once, and MaineJayhawk and I use it now and then. At this time, does it seem like anyone can see more than 2-4 or 3-5 out of this? Link to comment Share on other sites More sharing options...
CoastalWx Posted January 6, 2011 Share Posted January 6, 2011 Couple that with sickos like you and I up until 4AM staring at every detail, and messenger comparing the pube length of the vort from each run....there you go. There are so many more models and they are run more often.....also forums like this, where people get in a circle jerk and scrutinize them. But I'm not scrutinizing them. They have their issues, and we all know the NAM does too, but this season has put them to the test. They're going to struggle with the pattern we have. I go back to Analog96's post about how models suck, and that is simply untrue. They are guidance and it is up to us to interpret them. This year has been a great case of why ensemble guidance is such a good tool to use. They try to help the forecaster out, but taking in many possible scenarios, and spitting out the solution. In many cases, they have been pretty good 5-7 days out. Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted January 6, 2011 Share Posted January 6, 2011 But I'm not scrutinizing them. They have their issues, and we all know the NAM does too, but this season has put them to the test. They're going to struggle with the pattern we have. I go back to Analog96's post about how models suck, and that is simply untrue. They are guidance and it is up to us to interpret them. This year has been a great case of why ensemble guidance is such a good tool to use. They try to help the forecaster out, but taking in many possible scenarios, and spitting out the solution. In many cases, they have been pretty good 5-7 days out. I didn't mean you, specifically.....I just meant that people on this forum tend to. Link to comment Share on other sites More sharing options...
Clinch Leatherwood Posted January 6, 2011 Share Posted January 6, 2011 no...but I assumed you weren't the average person and would understand.. I've personally never liked the NAM. I think the GFS is decent. Cute. You know what I meant. The GFS and NAM don't even agree quite often at init. Sure they always had differences in strength with the NAM being stronger. But lately they cannot even get the position in reasonable agreement. That's a change, been doing this long enough to know that even 5 years ago that didn't happen often and when it did we had a bad run. We had the odd comments two weeks ago about data error. Last night there were significant differences at init, sure grid scale etc, but when the NAM blows the differences up 6hrs later did they really exist or was it bad data? EDIT: Check the wind speed/direction dead middle of Iowa on the init, real or imagined? Are we having a problem with assimiliating data just maybe? Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted January 6, 2011 Share Posted January 6, 2011 This thread is already a disaster and I haven't even bit****. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.