RCNYILWX

Some hindsight involved here, but I think the average of the ensemble 10:1 mean, probability and percentile stuff told a pretty consistent story. From the NWS perspective, I haven't been directly involved in the forecast process this time, but the inter-office collab and collab with WPC aspect is way more challenging than many realize. Internally, we felt that the QPF and ratios were too aggressive but there's only so much we can do when trying to form a consistent picture across large regions. To me this looked like a advisory to perhaps low end warning event, particularly on the non-GFS/GEFS data. It's easy to get caught up in the Kuchera ratio output when assessing the snowfall forecast, that method is too simplistic though. 10:1 gets bashed, and it should in marginal temp situations and more clear cut high ratio setups, but I think it's a good first guess plenty of the time for synoptic systems. Use the 10:1 and then adjust up where you're confident better banding is going to set up, through using model diagnostics like f-gen and cross sections, and utilizing the Cobb output. Sent from my Pixel 9 Pro using Tapatalk

Good voice of reason today Sent from my Pixel 9 Pro using Tapatalk

Noise level changes on the GEFS all in all if you look at the past 4 runs. Sent from my Pixel 9 Pro using Tapatalk

Solid agreement in the 12z ensemble data overall (10:1 total and 24-hour snow swath as a proxy), particularly the EPS, GEFS, and UKMET ensembles (MOGREPS-G), which is good to see at this juncture. The CMCE is farther southeast but has enough more amped members to not be too crazy far off from the other ensemble systems. Sent from my Pixel 9 Pro using Tapatalk

Might be worth starting a thread for the Wednesday system. Giving me vibes of the Feb 2019 ice storm (surface pattern is reminiscent to the observed pattern from our event recap page for 2/11-12/19). The GFS is slowly but surely starting to come around to the ECMWF/EPS idea. Also the EPS has lost a lot of the farther north surface low track members. This is important because the surface evolution is such that the warm front will remain well south until Wednesday evening, and then the track somewhere across the southern LOT CWA could very well prevent some places across interior northern IL from ever getting above freezing. If you compare the past several Euro runs, there's been consistency on a swath of 0.25-0.5"+ estimated FRAM output within the DVN, LOT, MKX CWAs and then extending into lower Michigan. Also noted in the AFD, the 00z and 06z EPS had 60-80% probs of >0.1"+ ice accums near and north of I-80. Sent from my Pixel 9 Pro using Tapatalk

For the 10 year anniversary of the Superbowl snowstorm, I updated the event recap page with another one of our forecasters, and we also the added new graphics for that page to the 2011 GHD Blizzard event recap. https://www.weather.gov/lot/2015_02_01_WinterStorm https://www.weather.gov/lot/2011blizzard Sent from my Pixel 9 Pro using Tapatalk

Today's overperforming temps and winds are a harbinger of things to come on Monday. 21z RAP looks realistic and it might still be a bit too cool. Thinking 45-50F is plausible in the LOT CWA. Given that we had low to locally mid 40s highs today with wind gusts up to 35-40 mph, Monday has an even better setup for strong west-southwest winds, so feeling pretty confident that we'll need a Wind Advisory at least for areas near and north of I-80. Sent from my Pixel 9 Pro using Tapatalk

Certainly too early to feel confident in it given uncertainty in the MJO realm, but nice look (500 mb height anomalies and 850 mb temp anomalies) on the 12z EPS in early Feb. GEFS was decent too, but would be cool to get something like the EPS evolution to play out in the extended. Anything to shake things up from the extended CAD doldrums. At least next week will be warmer here. Should still be a decent pattern (clippers and periods of LES) for the northern Lakes though.

Yep, should be a band that breaks through the dry air earlier in the day, most likely across far northern IL into southern WI, then the rest of the area with a period of snow late in the day through the evening. Sent from my Pixel 9 Pro using Tapatalk

Agree, really impressive airmass. The most intense cold with no snowpack in quite some time. For Chicago, have to go back to Feb 4-5, 2007 for a comparable event. Rockford had snow otg in that stretch, so it's been even longer there. Would've been interesting to see how close this got to the end of January 2019 with a solid snowpack in place. Edit: Rockford, at -11, at least tied 1/17/1954 for the coldest temp without snow cover since 1951.

It should be labeled more clearly what the sample point ranges indicate. At my office, we've chosen to stick with the standard color curve ranges. Sent from my Pixel 9 Pro using Tapatalk

That's the 25th to 75th percentile ranges for those locations from the NBM distribution and including the NWS deterministic forecast. I really don't like that we're doing this, because as you and others noted, our forecast is still there represented by the standard color scales. It's a misnomer to call those maps with the 25th-75th percentile sample point ranges plotted "official NWS forecast". Sent from my Pixel 9 Pro using Tapatalk

ORD lost its snow depth for the 00z ob. Either way, wasn't going to last through tomorrow. Sent from my Pixel 9 Pro using Tapatalk

Felt like it was good context to have vs. the 1996 event, plus the T snow depth in those older dates may have been close to an inch based on the daily data.

With officially 0 SD, it's only happened twice: 2/2/1996 (-5/-16) 2/3/1996 (-5/-19) I believe that -19 is the record low with no snow cover. With a T snow depth, it's happened 4 times: 2/23/1889 (-2/-11) 1/5/1912 (-5/-10) 1/6/1912 (-1/-11) 1/7/1942 (-4/-13) There's been 47 days total with sub-zero highs in the entire period of record. Sent from my Pixel 9 Pro using Tapatalk

Very good question - I will look that up. Sent from my Pixel 9 Pro using Tapatalk

You got it, impressive! I had forgotten that one. Sent from my Pixel 9 Pro using Tapatalk

Chicago posters, without looking it up, when was the most recent sub-zero low at ORD with 0 snow depth? Sent from my Pixel 9 Pro using Tapatalk

https://rapidrefresh.noaa.gov/ https://journals.ametsoc.org/view/journals/mwre/144/4/mwr-d-15-0242.1.xml The RAP is 13 km resolution, vs. 9km for the ECMWF, and 12km for the GFS and NAM, of the most commonly cited non-CAM guidance. The GDPS (GGEM) is run at 15km resolution and the RDPS (RGEM) is run at 10 km resolution. Finally, the UKMET is at 10 km resolution. What makes the extended RAP runs more prone to errors is that it's a hot start, hourly updating model with radar reflectivity as part of its data assimilation. This is just like the HRRR (which is initialized with RAP data) is more prone to errors in extended ranges. Think of the outer ranges of the RAP like days out on the ECMWF and GFS and beyond 24-36 hours out on the NAM. The RAP/HRRR can be right beyond 12 hours out, but in general they're more likely to be reliable within 12 hours, and even more so within 6 hours.

Should never have been issued so early, in hindsight.

I know you know that's overdone, but re. the LES, def a signal there for some accums, orientation and residence time of convergence TBD. Main limiting factor is lower inversion heights (around 850 mb), which would reduce flake size and snow ratios. Also the instability and lake to 850 delta Ts are on the modest side. The Canadians have higher inversion heights than the other guidance but not so much higher to explain the difference in QPF/snow amounts.

That plus the confluence from the upper low/PV lobe over eastern Canada and the northeast forced south by the NAO block.

Big drop in the 6"+ probs on the past few EPS and GEFS across our southern counties. 06z EPS only has 50-70% 6"+ probs roughly from STL to Lawrenceville. Barring a big change back in the other direction, we're not planning to hoist a watch this afternoon for the southern tier of counties in the LOT CWA.

The expected phase change back to a strong +PNA after the potential event next weekend caught my attention, thinking back to the Superbowl 2015 storm. It's not a perfect match but seeing some similar pieces in the Pacific pattern progression, when comparing the ECMWF ERA5 data on weather.us to the 500 mb anomaly forecast from the 00z EPS. A deep trough into the Aleutians forced a ridge spike off the west coast, dislodging what would become our main wave into the Rockies, while a deep trough also was in place over northeast NOAM and moving off the coast. This enabled height rises in between into the Midwest. Finally, a northern stream wave was able to dig in and at least partially phase in with the main wave for the deepening surface low we saw on 2/1-2. One of the biggest differences in the loading pattern was that the NAO was positive back in 2015 while it's forecast to be strongly negative in our timeframe of interest. That I think lends even moreso to the thread the needle aspect with the importance of a proper phase with the northern stream wave. If the phase is missed, the downstream blocking would suppress an overall weaker system. Back in 2015, the main wave being quite juiced and the lack of downstream blocking likely would've resulted in the good WAA thump that played out even without the phase for the prolonged high ratio deformation snow portion of that event. Overall, the analysis from [mention=525]Chicago Storm[/mention] and [mention=525]OHweather[/mention] is spot on in that the pieces could plausibly come together for a nice event, but with so many moving parts, things could certainly go wrong.

Decent looking snow squall setup on Wednesday on some of the guidance, 00z HRRR, past few NAM12 runs, and the 18z Euro. Sent from my SM-G998U using Tapatalk