WINTER STORM FCST 12/14/07 - 12/17/07: SOUTHERN US - NORTHEAST US

[Dan Robinson]

Interesting to note that the 0z NAM (WRF) scraps the storm. Outlier, yes - but interesting nonetheless.

 

[Rob Lightbown]

One problem that I have in forecast snowstorms here in northern Maine (Where there is 2 feet of snow on the ground) is snow to water ratios. Is there any tried and true method on accurately determining this aspect of snowfall forecasting. Take this weekend's storm. The 00 UTC GFS model is painting around 0.90 inch of water equivelent and the BUFKIT snow tool is indicating 20-30:1 ratios, while the Kuchera algorithm is showing around a 20:1 ratio. One thing I have read in the past is that coastal storms usually have a 12-15:1 ratio, so I guess I'm not sure what to believe and let me tell you the discrepencies in snow ratios gives some large snow total possibilities. Any thoughts would be welcome.

 

[Mikey Gribble]

Tricky forecast for the snow storm that will affect Kansas and Oklahoma tomorrow night. The GFS had been steady as a rock over the last several days (along with several other models) with the track and intensity of the storm. The NAM was completely out to lunch (I mean waaayyyy off) until the 12Z run this morning and now it has fallen in line with the GFS. The 12Z of the GFS moved the track northward though so now I'm not sure what to believe. I am leaning towards the track of the 12Z NAM with a blend of precipitation intensities of both the NAM and GFS. The current track of the NAM is more in line and consistent with what the other models have been showing in the past several days and the NAM has done a pretty good job on past storm tracks once you get within 48 hours. Right now my thinking is that an area from Ashland to Ark City to Wichita cone (facing NE) will be the western edge of the heavier swath of snow, with that cone of heavier snow extending East to east central Kansas. I am thinking 5-8 inches should be common in this area. If I had to stick a number on snow fall amount for Wichita right now I would go with 6 inches. That is a good snow storm for us. For some reason we always dodge the bullseye on good snow storms, but we are actually in pretty good position for this one. I still need to do my forecasting this morning and check the 12Z run of all the models, so right now I'm only fairly confident on the track of the heaviest snow.

 

[jladue]

One problem that I have in forecast snowstorms here in northern Maine (Where there is 2 feet of snow on the ground) is snow to water ratios. Is there any tried and true method on accurately determining this aspect of snowfall forecasting. Take this weekend's storm. The 00 UTC GFS model is painting around 0.90 inch of water equivelent and the BUFKIT snow tool is indicating 20-30:1 ratios, while the Kuchera algorithm is showing around a 20:1 ratio. One thing I have read in the past is that coastal storms usually have a 12-15:1 ratio, so I guess I'm not sure what to believe and let me tell you the discrepencies in snow ratios gives some large snow total possibilities. Any thoughts would be welcome.

Rob, you sure hit on a sore topic in the realm of what we should know vs what we do know. You know there's no decent network that measures ratios over hourly to 24 hr time spans. You come close with SNOTEL sites but they're in the mountains. The METAR sites are terrible because there's no adequate shielding around the bucket and so the catch efficiency goes below 50% with a wind above 10 mph. Snow spotters typically don't measure ratios. COCORAHS members do measure ratios but only once per day. In the course of a snowstorm, the crystal density habits change drastically from one hour to the next, especially when you get convective bands passing by.

So with all the lack of quality obs, there's been a terrible lack of good research. Nevertheless a few folks made some decent attempts. You can take two of our lessons (lesson 5) on snow ratio basics from our winter weather course at
http://www.wdtb.noaa.gov/courses/winterawoc/IC6/index.html

There's two techniques besides just going with 10:1 or some other guess in the lesson.

1. The Caribou snow technique which forecasts high ratios when the level of maximum vertical motion phases with the dendrite production zone (-12 to -18 C). The thinking is if most of the vertical motion is where dendrites form, then fluffy snow forms.

2. The Roebber et al. 2003 technique which forecasts snow ratios based on a neural network. The best inputs come from month, compaction due to wind speed, and warmest temp in the layer. Here's a link to their online forecasting site: http://sanders.math.uwm.edu/cgi-bin-snowratio/sr_intro.pl

Note that the Caribou technique doesn't take into account settling and compaction and so it may be better related to short duration snow ratio measurements. I find that the Caribou technique often overestimates snow ratios. If most of the vertical motion is confined to the dendrite production zone and there's no warm layers, it might work better. But broad vertical motion where riming and graupel can form around dendrites, the ratios would go down. But I'm guessing because there's a huge dearth of ground truth (i.e. no snow ratio, snow flake type reports).

regards,

Jim

 

[Chris Hayes]

I'm hoping the forecast discussion is right for our area. NWS mentioned 8-12 inches of snow possible from Saturday to Sunday over NE Ohio. Just need the storm to go south and east just enough to keep it all snow, and hopefully no dryslot move in.

 

[Andrew Francis]

Difficult forecast. Everything was off on the heaviest ice accum for the ice storm by about 60 mi SE of where the original heavy precip core was forecasted. So maybe the snowstorm here in KS and OK will also be shifted SE. I just want to see a good amount of snow lol :'[

 

[mark plate]

Difficult forecast. Everything was off on the heaviest ice accum for the ice storm by about 60 mi SE of where the original heavy precip core was forecasted. So maybe the snowstorm here in KS and OK will also be shifted SE. I just want to see a good amount of snow lol :'[

New 00Z WRF/NAM has trended back that way...closer to its outlier 12Z run from 12/11. Will see if other models follow suit.

 

[Andrew Francis]

This morning I was dissapointed when i checked and the Winter Storm Watch had been dropped for central OK. And the forecast changed to 1-2 inches.

Now nearly all of central and NW OK just got upgraded to a Heavy Snow Warning with 5-7 inches likely?!!

How does that change that drastically in a matter of hours??? Maybe the dropping of the winter storm watch was an error?


Anyways, snow breaking out now in SW KS and TX panhandle

Time for a NOW thread

 

[mark plate]

Looks like the deformation zone sets up Saturday morning along roughly the I-44 corridor. 5-7 inches seems excessive and will be very localized if it occurs. 2 to maybe 3 inches will be more widespread within the deformation zone during the day Saturday.

Heaviest snow tonight will target the Wichita vicinity.

 

[Mikey Gribble]

I'm not sure what to think anymore. I've been pouring over model data for the last two days trying to work out the small scale details and I'm no better off than I was 48 hours ago. My current thinking is in line with the NWS, that the heaviest swath will run SW-NE from DDC to Emporia, just North of the Wichita area. Four days ago I was afraid the heaviest snow would go South of us, but with each run of the GFS it kept creeping the heaviest snow North. Now it is going to go to far North for us to get the heaviest stuff. It will be close though. You never really know the small scale amounts on snow. The bottom line is that the DDC-Emporia swath (either side of that line) is going to get a good 4-8 snow. Localized higher amounts are definitely possible, but that just depends on where the bands set up. I always bite my nails thinking we are going to miss out on the snow until we get a big band coming at us.

 

[Bill Schintler]

Winter Storm FCST: Eastern IA

Yet another winter storm will affect portions of the Midwest on Saturday, however the heaviest snowfalls will remain well south of the eastern Iowa area.

Cedar Rapids:
Light snowfall will start at 10 AM CST Saturday and continue until 8 PM. Expect 0.5 inch of snowfall at the Eastern Iowa Airport with lesser amounts further north. Marion and County Home road will see a dusting of accumulation.

Iowa City:
Light snowfall will start at 8:30 AM and continue until 9 PM. Expect 1.0 inch of accumulation.

Marengo:
Light snowfall will start at 8 AM and continue until 8 PM. Expect 0.8 inche of accumulation.

Synopsis:
At the SFC, a CF pushed through the area last night, bringing a reinforcing shot of cold air with NW winds gusting to 20 mph. WV satellite imagery indicated digging cutoff low over AZ/NM while the developing moisture conveyer belt aloft was evident in the form of wispy CI and a baroclinic leaf structure from TX through IA. In WRN KS, radar indicated developing returns with moisture reaching the ground in KDDC after 18Z in a zone of isentropic lift on the 285-295K SFCS.

Discussion:
MDLS in decent agreement with regard to the evolution of a coupled H3 jet structure and regions of ULVL divergence and attendant UVM, with the strongest lift remaining well S of the area. A 70kt H5 streak over NV will dive S and phase with the SRN stream wave now over NM/TX. The GFS is slightly faster more open with the EWD progression of the trough while the NAM maintains a more closed solution. This trend continues though Sunday as additional energy dives S from CAN to phase with the SRN stream energy while a major nor’easter develops in the NERN CONUS.

Tricky part of FCST is predicting the NRN extent of the precipitation given SWRD intrusion of dry air at the SFC through the mid-levels. Model soundings indicate dry air from H8 to H6 at the onset, so will adjust MDL QPF downward accordingly. The strongest forcing will occur between 14Z and 19Z as the right-entrance region of a 130kt H3 streak along with a healthy shot of isentropic lift overspreads the area. A suit of snowfall FCST methods were considered. Investigating QPF, the GFS is most bullish and brings over 0.2 inches of QPF along I-80. Given the aforementioned dry air, this will be adjusted down to 0.1 inch along I-80 and 0.05 along US-30, which also brings it into better agreement with the other MDLS. Cold temperatures will be in place aloft, with H8 in the -10C range and H7 about -13C. A 14/1 snow/water ratio will be used, which gives 1.5 inches along I-80 with lesser amounts to the N. Looking at the Garcia method, considering 1-2 g/kG mean mixing ratio lifted along the 290K SFC for a duration of 6 hrs suggests about 1 inches of snowfall at best.

Looking ahead, the recent active pattern will slow down for the next six days as the pattern transitions to a split-flow regime and a parade of weak disturbances transitions E within the NRN branch. Finally, late in the week a potentially strong system will then affect the upper-Midwest.

- bill
1:30 PM CDT, 12/14/07

 

[Robert Dewey]

Westward trend continues... Looks like quite a snowstorm will setup from MI eastward. I think HPC's latest snowfall graphic is a bit too far east, but that's just a first guess and not really an accurate portrayal of amounts.

I'll probably spend more time looking over things tomorrow morning, or perhaps the 00Z run if things get even more interesting.

Needless to say, I suspect some areas in eastern MI will see totals in the 12-16 inch range (perhaps lake enhancement areas, or if the westward trend continues as I suspect it will; and yes Rob, I am "predicting the models" ).

The BUFKIT forecast shows a mean snow ratio of 18:1, for a storm total of 16-18 inches over Detroit, with a solid 30-40knt wind gust potential. That certainly has my attention.

Areas further east (parts of NY, for example) will likely see 18-24 inches, or even more. I noticed BUF has "up to 30 inches" in some of their WSW products.

 

[Joel Wright]

Looks like a nice 5-8" band of snow will set up from southern MO, through the St. Louis area, up through the Ft. Wayne IN area, and up towards Detroit. A few locations up near 10" aren't out of the question. Particularly in northern IN and southeast MI. Snow ratios will be 10-12:1 in the heaviest band.

Further northwest, it looks as though snow may accumulate as far north as central IA and into extreme southern WI.

Central Kansas will get a nice burst of snow much of tonight. This area should easily get 6"+. Looks like it's centered near the Hutchinson area.

This is a lot better than ice though, isn't it guys?

 

[HAltschule]

The Capital District is expecting a wide variety of snow accumulation. Sleet is expcected to mix in to some extend and keep the snow accumulations down but we should still see 5-9" of snow before the changeover.

The bullseye for this storm should be the deformation zone that's expected to develop to the NW of Albany. In these areas where all snow occurs, 18-24" is expected with higher amounts in the 'Dacks and the 'Skills (Adriondacks and Catskills).

NWS Boston's Walt Drag has a nice discussion about a possible gravity wave screwing around with the pcp totals.

If the storm stays a little further east, then we could stay with all snow. If that happens, then it seems that we could easily exceed a foot of snow. Stay tuned.

 

[Mike Peregrine]

EAX says 5-7" for metro Kansas City in this briefing at 3:30 pm. Looks like heaviest snowfalls cutting through the middle of the state, from Butler across toward St. Louis ... in this area they're expecting 8"+. This system turned into quite a little snow maker. It's no wonder with a nicely saturated plume of tropical moisture flowing.

I'll take this over ice any day of the week. Not sure how I feel about this ON TOP of ice, though -