RCNYILWX

Admittedly was much more focused on tomorrow's event today but I'd still be concerned about h9 verifying colder as long as you keep that northeast component from the surface high to the northeast. Going into 12/28/15, we were much more concerned about icing than the sleet because of the magnitude of the 850 mb warming but the low level cold wedge was still enough to produce mostly sleet north of I-80. I saw some of the 925 mb progs being warmer than previous runs, so if that occurs then the sleet concern will be unfounded, though we still have plenty of time for changes with this setup. Sent from my SM-G965U using Tapatalk

This sounding below demonstrates the NAM's dry air issues pretty well. It's over LaSalle or Kendall County valid at 21z. If you go to the Precipitation Type parameter on COD, this area is not being counted as snow by the model. Yet you look at that sounding and that's very clearly a snow sounding, a good WAA snow sounding at that. 850 mb is a pretty good proxy for saturation level at which snow reaching the ground is a decent bet. Since the NAM is 3 hour time steps, not counting these areas of the grid space of having qpf as snow cuts into the qpf and snow totals unrealistically. This is an area where meteorologists and weather enthusiasts can add value to the model forecasts when we know their typical biases. Sent from my SM-G965U using Tapatalk

I think analyzing the NAM is as simple as tossing its BMJ parameterization scheme which does terribly with antecedent dry air. Look at the composite reflectivity vs the precip output and that should be all you need to see. It will take some time to top down saturate, but with robust moisture transport and strong forcing you shouldn't lose 3 hours of precip. The NAMs have it snowing hard aloft at 21z. If anything, skew timing earlier and account for possibly earlier northward surge of warm nose and the forecast is in decent shape. Sent from my SM-G965U using Tapatalk

Sell on anything but brief (1-2 hours) sleet with this. Models always hang onto the low level cold wedge too long in strong warm advection setups. I think this a solid thump of snow, mix or brief change to sleet then freezing rain. Sent from my SM-G965U using Tapatalk

I'll take a stab at this. It would be cool to get a take from an actual modeler on the frustration we all share in the apparent degradation of short-medium range forecasts, especially in the winter. My perception since I've worked at NWS Chicago in 2010 is that the operational model forecasts haven't improved much if at all for our region. Meanwhile, I know that the modelling is so much more advanced now than it was then. I think there are a few feasible culprits to lack of good model consensus at fairly short lead times and changes in the forecast in short lead times. 1) Fast flow patterns have seemed to the rule in recent winters. Have seen it speculated on other forums/subforums that the fast flow may be linked to CC. Whatever the case may be, it makes logical sense that faster flow is tougher for even our advanced NWP of today to accurately handle key features that can often start as lower amplitude earlier in the forecast process. As we know, any errors at initialization get amplified the later into the forecast. 2) Lack of more recon sounding data over the all important Pacific basin. I don't know the full story behind data inputs that are no longer there, but it seems clear that satellite (despite the recent advances satellite data) and sparse aircraft soundings out over the open Pacific are insufficient to bridge the gap until RAOB sampling. 3) Doubled edged sword to ever higher resolution: Can very small scale features picked up by higher resolution modelling systems cause errors to increase quicker than more smoothed out data of the past? Further discussion on this can probably be sent to banter, but I'm curious if anyone else has any other ideas on this.

The GEMs especially but also the UKMET have been consistently farther south with wave 1, and we've seen a several run south bump from the other guidance since. Question is do we see the southward adjustments continue once the first wave is fully sampled? In making a hypothetical forecast blend at this point, I think I'd give a nod to the consistency of the GEMs and UKMET and weight them more than the Euro, GFS, NAM. Will be interesting to see trends with the 12z guidance when the wave is fully sampled. Sent from my SM-G965U using Tapatalk

All about how quickly the southern wave gets sheared out. Even on a stronger/NW solution like the UKMET, the wave opens up but is able to lift much farther north with incoming northern stream energy much slower. GEM is much stronger and faster and therefore farther southeast earlier with the northern stream energy, which shreds wave 2 in the confluence. So many pieces coming into play for wave 2 that won't be sampled fully until mid week and it's already a faster flow pattern that gives the models fits, so expect the operational runs to have large variance and large spread in the ensembles until more than satellite data from the key features can get inputted into the model initialization.

Also coming off the very dry surface high, the dew points start quite low and southeast surface wind through 06z or 09z is pulling from dew points still in the upper teens to lower 20s over central Indiana. Sent from my SM-G965U using Tapatalk

For those keeping score of the 00z cycle, the RGEM came in way south with wave 1. Sent from my SM-G965U using Tapatalk

I buy downtown Chicago having issues with a strong east or northeast wind during wave 2. Lake is still mild enough to cause problems immediately along the lake and a bit inland in a marginal air mass. Sent from my SM-G965U using Tapatalk

If you recall the Feb 2019 icing, that's something we got burned by. Much of the guidance kept insisting on warming the boundary layer to above freezing even though surface winds were progged to remain northeast/east-northeast at the surface. Putting down snow cover and not melting it going into the event is a wild card that could point toward sfc temps verifying colder if we get a scenario like shown on the 12z Euro.

Looking at the 12z EPS, wave 1 had a pretty big adjustment south similar to the operational. There's substantial west to east spread on the sfc low placement for wave 2, with the mean actually bumping east of the 06z. Another component we'll have to watch is the possibility of wave 2 getting sheared out in confluent flow to the northeast, which could tug the sfc low track more northeast with time depending on when that occurs. The EPS mean favors this sort of setup, so there's a decent amount of member support for the wave closing off, occluding and then shearing northeast. So many complexities to the setup. Edit: Included wave 2 in 2nd sentence of 1st paragraph

If we're to get a closed off low lifting close to due north like the Euro run will also have to watch for the magnitude of the high pressure to the north. The setup offered by the Euro is somewhat similar to 12/28/15 (Edit: the event referenced by@WI_SNOWSTORM), though that one had a stronger HP to the north/northeast. Should the surface ridging extending back west across the northern Lakes trend stronger, that could play a big role in sleet vs freezing rain distribution. 12z run verbatim looks icier farther north than 00z run did without looking at the ice accum output, still borderline IP vs ZR for a time around 12z north of I-80 using 925 mb temps and winds as a proxy. You could also keep the surface (away from Lake Michigan) at or below freezing for a longer duration with that odd due north track because that would prolong the period of northeasterly sfc winds and associated evaporative cooling. Would be an impactful mixed precip event for the LOT CWA as modeled on the Euro, especially north half and away from Chicago shore. Lots of time to go on this one and pretty much a wide variety of outcomes on the table.

Red flag for me was seeing 925 mb wind speeds being at 40-50 out of the east-northeast. When combined with the progged h8 and h9 temps, that's a strong signal for cold/dry air damming. Sleet will be something to watch for in this setup if we get a wound up low like on the 00z ECMWF. Sent from my SM-G965U using Tapatalk

Just with respect to the 00z ECMWF operational run, I think there would probably be a much larger zone of sleet. Valid 12z Friday when both Pivotal and WeatherBell algorithms are showing primarily freezing rain, I'd guess those 925 mb and 850 mb temps would be supportive of sleet, especially north of I-80. Don't have access to forecast soundings to confirm though. Also looked at the GEM for Tuesday night into early Wednesday and it appears the algorithm is accumulating ice when the soundings for at least a portion of the icing zone with northward extent would favor sleet. Does appear there would be a freezing rain zone south of the sleet zone. Regarding the overall 00z suite, focusing on wave 2 alone, all we can say is that the potential for a high impact event is there, but lots of moving parts. Going to take a while to sort this one out. Wave 1 looks to produce a swath of advisory to low to mid range warning snows, with a mixing zone south, and even wave 1 has a good deal of uncertainty associated with it. Sent from my SM-G965U using Tapatalk

Yeah, can't blame the sentiment. But now that the secondary low/southern wave is slowing to not coming up until Friday, that's an eternity in model land, and clearly not the final solution. Sent from my SM-G965U using Tapatalk

Fwiw, the 18z EPS doesn't have a lot of east outliers with the secondary low, has more tracks that would be too far west for much snow here, with a slight westward adjustment of the mean. It would be nice to get some of the front end snow, but failing that, I think overall best chance (not necessarily a good one yet) to get a 6"+ snow is with the secondary low/wave 2. While there's been a tendency for weaker/sheared/more east with secondary development, the magnitude of the eastern ridging makes a case for Alek and Hoosier's worry of a more amplified nw track cutter/rainer. I'd still lean toward miss east being a bigger worry, but miss NW is not too far off. I think we're in the game here, but as always need a lot to work out right to get a good event locally. Sent from my SM-G965U using Tapatalk

That pattern could produce snow events because of the assistance of the strong -NAO to modulate the +EPO some. It's certainly not a cold pattern. I'm not on the optimism train, just saying that's not a lock to be snowless like this month has been. The 850 mb positive anomalies are not obscene and the mean 850 mb 0C line is near or south of I-70. If we get a lower amplitude wave, could snow here with that. Sent from my SM-G965U using Tapatalk

To your point, here are the low locations on Wednesday and Thursday night into Friday for the 12z EPS, followed by individual member total snowfall. Good support for a weaker primary and a stronger secondary, so I agree that we're in the envelope. Unfortunately it's another pretty much thread the needle setup. Sent from my SM-G965U using Tapatalk

Well, the UKMET is something... One thing we can say for sure is it's not even close to the 00z GFS aloft. Other than that, much different evolution than we've seen on the other operational runs. Sent from my SM-G965U using Tapatalk

Yep just noticed the same thing and edited my post to reflect that. The GEFS still appears to follow the op more than the EPS does, 30 members vs 51 on the EPS may have something to do with that. That said, seeing a few members go toward a strong secondary might be a sign. Sent from my SM-G965U using Tapatalk

We'll see here shortly with the UKMET and then the Euro which way things are leaning. This past setup, I think the GFS having the support of the UKMET was key. It can be prone to some big swings but it's a solid model per verification stats, we just have less data available from it. If UKMET supports a GEM-like solution like the 12z run did and the ECMWF and EPS bumps again, we might be seeing a trend toward a GEM-like outcome. Edit: and vice versa could be true too. Did notice that the spread of the GEFS members increased, which despite the op-like solution being still more favored could be a sign of lowering confidence in which scenario is more probable.

Even though it was warmer than normal, January 2012 happened to be the best month of that winter, two decent storms and the only AN snow month of the cool season. The second event in January 2012 had nice high ratio fluff with the longest duration of 1/4 mile visibility in snow at ORD and MDW since GHD I and caused nightmare travel on a Friday afternoon commute. Sent from my SM-G965U using Tapatalk

It appears to have some better front end snow, but wouldn't think in reality the sfc low track on that run would be supportive of it. Sent from my SM-G965U using Tapatalk

Probably...so far UKMET is a no Sent from my SM-G965U using Tapatalk