ILMRoss
-
Posts
658 -
Joined
-
Last visited
Content Type
Profiles
Blogs
Forums
American Weather
Media Demo
Store
Gallery
Everything posted by ILMRoss
-
I spoke a little too soon, next frame the r/s line makes a bit of a comeback, almost back to the previous run
-
Solid 40 mile shift N with r/s line, weaker high here.
-
18z FV3 already looking like a pure weenie run
-
Wouldn't be surprised if there's a diminished gradient because more members are throwing some of that front end thump further east. that initial finger of precip extending ahead of the storm can be a doozy. I would still expect the main rain/sleet line to developed somewhere along RDU/CLT during warmest portion of storm
-
Really I wouldn't even focus on p-types much right now at all, not until the NAM-3km and RGEM has taken a couple of cracks at this storm. Until then, a rough guideline would be to take the modeled 850 0C line and imagine another line 50-75 miles north of that... there's your sleet zone between those lines. We're still talking NC snow, lol.
-
FV3 is fine, the low is going to gravitate towards the Huntsville/Knoxville area a little before the transfer because of the CAD forming lower pressures on the OTHER side of the Apps... One more thing to think about is the overall pressure field remains similar, but that little "L" icon is going to shift. It's there because Levi's or whoever's code puts that L at every local minimum... and in a transfer like this, that could be a lot of areas! Run to run, the "L" could jump by hundreds of miles, but in the grand scheme of things, it's probably not a big shift. It's a psychological thing! Be careful!
-
FV3 looked fine to me, a touch slower but overall maintained the same general r/s line.
-
GFS is wild. Check this out. I got two trend gifs. First one is the 500 vort over Texas: So, there's some variation, and these differences are important! But for the most part, it's just noise, and the front end outcome hasn't changed much with this system. Now fast forward 24 hours later: There are worlds of differences here- vort maxes aren't just in different states, they're in different regions- Especially on the last 3 runs. This has a lot implications for how our trough is oriented, where the best lift is, where the SLP ends up, and consequently, where the high gets pushed. Don't really know what's going on upstream for the GFS to cause these differences. I also don't know if other models have this, it's just the first thing that caught my mind on the GFS. Any of these solutions are still probably plausible but it's definitely weighing on my mind that the GFS has no idea what's going on once this shortwave gets into the SE.
-
GFS looks like it lost that positive tilt it had in 18z so this run should at least be juicer in qpf, like previous non-18z runs
-
Blah, wish I could give you a straight up answer. Perhaps? Someone posted verification scores for all the models a few days ago which showed the FV3 was verifying slightly more accurately than the GFS. Based on that, maybe, but this is a question above my pay grade.
-
Two things: 1. Just because a low is jumping around doesn't make a model run unrealistic. Lows are going to jump to where the combination of vertical lift and low level spin are maximized. 2. 18z GFS FV3 was a pretty cool run. Caught a wide swath of people with significant snow. Seemed like a pretty reasonable run, as far as dynamics go. Let's hope that continues.
-
If I were chasing I would choose fancy gap, hillville, or Wytheville, somewhere that’s still going to get CAD bleeding in but also at elevation. Proximity to I-77 and route 52 or 58 is also critical, interstates and us routes will be plowed first and foremost making sure you don’t get stranded longer than intended.
-
There's also counties in that dma that are pretty far south of the triad. I think she is being a little conservative but it goes without saying that a full fledged "omg it's going to snow" message would freak people out, and if that forecast is blown less people may watch that station. There is palpable pressure, from the sales department to the news director, to have this right. Before I got into broadcast, I thought, "oh hue hue hue I'm going to be brazen and I'll speak my mind!" but it's *a lot* harder said than done.
-
18z got derailed very early on, when configuration of the parent shortwave got pushed *way* more positive over the southwest. This kind of put a positive slant on the s/w for the rest of the run and to me, really tampered with the dynamics, and created less qpf. Is it possible? I mean sure, but I'm not putting much stock in this particular iteration
-
Raleigh will be in the running for 1-3 inches of front end thump before switching to... something else. Learn from the NE from that mid Nov storm... that front end thump can stymie even the most experienced models and forecasters, especially with a solid CAD in place. Might not happen, but I’d keep an eye on that possibility. Lastly, I would not make any absolute statements about surface temps until we’ve seen a few cycles of the nam, raleigh still firmly capable of a significant sleet/ice event in my opinion
-
I don’t know how much of that is the Canadian’s depiction of the CAD, or its general cold bias (it always seems to have the coldest temps in general of all the storms I’ve tracked around here). When I still lived in Wilmington, that model was my saving grace a lot lol
-
Yeah, as others have alluded to, the NAM, and it’s 3km nest, usually have the most accurate depiction with CAD. The reason? I think it’s all resolution- since CAD is driven by topography, you need to accurately represent that topography to accurately model a CAD! The lower resolution of globals see mountains as 25x25 km raised lumps... while NAM and other higher resolution has the resolution to pick out individual mountains, valleys, ridges, and peaks. I think that is really important in terms of accurately modeling what kind of cold air bleeds down from the parent high, and how long it sticks around.
-
Remember folks, if the gradient isn't running through the heart of Wake County, then you can toss that snowmap. It's not verifying.
-
A thought on this... There's a lot of convective feedback off the coast here... Look at the pattern of the pressure contours! I think the models are going to have a LOT of trouble with this, especially the GFS, which seemed especially prone to botching this kind of stuff during our early January storm this past winter. This screwy representation has a lot of implications, especially with moisture transport. Check out how dry NC is here! I think we trend wetter in this particular instance, especially once convection allowing models get in the game.
-
18z GFS is interesting. GFS has the shortwave landfalling in Cali as a much sharper wave... the orientation is a lot different. I included the trend. I reckon this is what eventually tugged the entire system to the N. 18z output sucked, even for me, but how on earth is the GFS going to make such a large change to the shortwave over one of the most under-sampled regions? I'm not tossing the 18z GFS but I'm not putting much stock in it either.
-
Ehhh... I don't really know if this issue is ever going to go away. The surface highs are nice, but the air mass we're inheriting is a rotting, modifying arctic air mass from earlier in the week. This air mass is going to be a pool of -3 to +3 C temps at 850, and I don't really know a mechanism that's going to change that. As of now, models have for the most part graced us with adequate temperatures, but this is a situation that might be somewhat more sensitive to mishandling of mid level temps than others. The entrenched cold air is never the thing that "wowed" us with this system, so far it's the qpf.
-
Agree with this, the places that have won with previous storms will likely win again. This is just such a beefy, large storm that the jackpot area could be 14-20 inches instead of 10-14 inches like our last couple of miller As. There was a storm in late January 2009 that this storm kind of reminds me of, don't know if the telecommunications/setup was the same but I remember that storm as another rumbling, west to east long duration event. I stayed in a suburb of Winston Salem with family friends and got a great storm. I would though, not hang my hat on the fact that this CAD high will be there. The 00z GFS, while an outlier, was also completely plausible, with an arctic trough/front creating warm air advection ahead of it in the NE and completely sabotaging the cold air supply.
-
I'm not really sweating this model suite too much, although I also have the luxury of being in the S WV mountains and I'll probably be good for at least a safe 3-6 under most scenarios. To the folks that say "The GFS looks kind of weird", you're right. The main change on the GFS wasn't our actual shortwave, but instead the placement of a strong arctic trough at 500mb in the NE. Look at the trend I posted- the 00z GFS backs this feature further SW by a solid 500 miles! This messes with everything- Not only does is completely shift areas of ascent and descent, and warp our high pressure, but it also acts to sweep out our low pressure very quickly- that's why the GFS is faster. Other runs and other models feature a slow, lumbering southern wave traveling unfettered from Cali to the east coast. Things get quicker when it gets caught up with whatever this trough is making an appearance.
.thumb.gif.4fdf989c90e64edf2ea392396fe1847e.gif)
-
I've been sniffing around these forums since winter of 11/12 and this storm is already in the top tier of consistency in the long range compared with previous storms. In my experience, it joins the echelon of 2/13/14 and 1/23/16 in terms of seeing a signal that somebody was going to get smashed 8 days beforehand. I've heard a theory that sometimes these potential "big dogs" can get sniffed out very quickly and show more consistency than smaller-scale, more nuanced storms. Which makes sense- bigger systems would probably have larger parent shortwaves and would be easier to model. In any case, looks like a fun system. One thing I will say that I've noticed- The high is nice, but I'm really not very impressed with thicknesses or the depth of the cold air. We're not negotiating with record cold after, to put it lightly. I would watch trends with mid level temperatures. That being said though, still a very strong signal this far out.
-
I honestly couldn't tell you exactly what's going on with the TT snowmaps, but something is funky with how the algorithm is coded and it's producing a lot of spurious totals. Trash those maps. We're not getting any snow whatsoever with this system. We're familiar with the tropes surrounding CAD wintry mix events. They often (but not always) trend colder and stronger, especially at event onset when models have a very hard time modeling the in-situ nature of some CADs. I could see that happening a little here. Generally, this is still a pretty textbook look for a CAD wintry event and I'd be on my toes for a lot of folks in the Triad and along i77 to maybe Statesville to see advisory level sleet/ice accums. Now that being said, let's hold our horses just a little bit. If this were January, I think this would have potential to be a major ice storm for a lot of NC. But, fortunately, it's still only November, and these cold air lobes are still in their nascency. I don't like curtailing a forecast based on climatology, because that's bitten me in the butt before, but I think it's important to take a step back and realize that this is already a rare event for mid-November, and a colder, more expansive icy area is just making an already anomalous event *more* anomalous. One last thing to address- Some folks have rightly acknowledged that ice is a self limiting process, and I think that is a concern with this. I don't think that cold air advection will do the trick keeping things below freezing. In a CAD event, your cold air source is coming further north along the mountains. In this event, that air won't be fresh cold, dry air! In Greensboro, that cold air advection will be shuffling in air that has *already* had a lot of latent heat added to it from icing upstream!
.gif.1112908655ad8bf54cf632e1a3ae9cdf.gif)