Good evening folks. If you are a fan of snow around here… the European models are your allies. Both the operational run and the ensembles are showing low pressure taking a track later this week that would deliver a nice accumulating snow to our part of the world.

Here is the European…


The European Ensembles…


The 18z run of the GFS is trying it’s best to trend toward what the above models are showing. We shall see what happens in the next few days… but we are on waffle watch as this is a system where some will go overboard with each flip and flop from the models.

In the short term… I am still interested in the possibility of snow late Tuesday into early Wednesday.  We also have a clipper working in tonight with some light snow and flurries into early Sunday.

 In one of Friday’s posts… our buddy MJ from the NWS in Louisville talked about some research on winter storms that impact Kentucky. He was kind enough to email me some information on this and I wanted to share this with you guys…

We have been working with St. Louis University since at least 1995 when I started interning at NWS Louisville.  Our first research with them was basically heavy precipitation and flash flooding.  Much of the research was done with Dr. Jim Moore and Ted Funk.  Sadly, Dr. Moore passed away a few years ago, but Dr. Chuck Graves has been keeping the research institute alive.

The link below is an NWA poster that was presented last October at the National Weather Association Annual Meeting in Arizona that highlights the research.  Make sure your computer has Adobe Reader for this.

Click Me

In a nutshell here is how things work.  The period of 1980 through 2010 was used to find snow storms that affect our CWA.  Any snow storm that produced 2+ inches of snow was selected.  A few systems were dismissed (less than 5 in the 30 year data set), but I am not sure exactly which ones.  They were most likely not used because they were either hybrids, did not fit the 5 data sets, or were anomalous events.  In the end, they had 81 cases out of the 30 years of data (not bad).  The only thing that I’ve noted is that we should have more clipper events in there.  However, many of the problems were based on the spotty co-op snowfall data.  Co-op snowfall data is much better since the late 90s…and now they are utilizing the public reports.  Thus, the number of clipper cases should increase as each year passes.  There is talk of getting an upslope category in there, but I am not sure what the status of that is.  Since the upslope events only affect the northeast 1/3 to 1/2 of the forecast area…perhaps that is why they have not been included yet.  So after these storms were selected, they went back at ran the NARR data sets and came up with composites at 850, 500, 300 hPa, and frontogenesis features.  They came up with 6 types of events.

1A/B: Southwesterly flow aloft events.
2: Deep trough aloft
3: Closed Low Aloft (East Coast Storm)
4: Polar Vortex (PV) over the Northern States
5: Clipper Systems

Types 1B and 2 were the most common types of set ups that brought major snows to the OV (1B: 24 cases, Type 2: 23 cases).  The above poster has snap shots of all the cases, but is quite limited.  We have a more comprehensive presentation in powerpoint, but I will need to ask and see if that can shared outside the office.

So when I commented this afternoon about the 14/12Z Euro solution being a match…it appeared to me quickly that we could be looking at a type 1A/1B set up.  Many times, I simply look at those 4 levels when looking at the GFS and Euro and look for the pattern setup.  I don’t really give a lot of attention to the surface solutions until we’re within 3-4 days of the event.  So far this winter season (Fall 2010 to Now), the GFS and Euro have had pretty good verification scores from 500 hPa and up.  Thus, one can see potential set ups some 6-10 days out.  So far, this research has held up well this year and almost all of our events have fell into one of these categories.  Some forecasters will mention these composites in the forecast discussion from time to time….so now you have an idea of what we we’re talking about.  I am going to see if we can get this up on our website (though it may be on there already).  This is ongoing research, so the number of cases will increase over time as systems that affect us are added into the composites.

I have done composite work before while I was at Florida State.  While working at the Tallahassee office, we did a sea-breeze composite study showing how the different flow regimes resulted in different areas of convection.  Simply looking at the flow near the surface, there were about 8 different types of wind flow that produced different areas of afternoon convection.  Many times, we could simply ignore the large scale model guidance (GFS, NAM, NGM) and simply use our local MM5 (now WRF) and our morning balloon soundings to do the forecast for the next couple of days.

-MJ

That’s some good stuff and we certainly appreciate it. I will have a full update coming later tonight so be sure to check back. Take care.