The Inland Extent of Lake-Effect Snow (LES) Bands: Project Update Joe VillaniNOAA/NWS, Albany, NYMichael L. Jurewicz, Sr.NOAA/NWS, Binghamton, NY

U.S./Canada Great Lakes Meteorology WorkshopApril 16, 2013

Outline Goals Methodology Results

– Forecast application Latest improvement efforts

Future Work

Goals Single out the atmospheric

ingredients typically most influential on the inland extent (IE) of LES bands

Use this information to formulate predictive techniques in an operational setting

Methodology Investigate LES events over an

extended period (2006-2009 across Central/Eastern NY)

Pick a number of locations both inside and just outside of LES bands at specified time intervals (0000, 0600, 1200, and 1800 UTC)

Select parameters and see which correlate best to IE distances– 12 km NAM initial-hour soundings

Example of Data Points

Points in and near the LES band

BUF soundingALY sounding

Parameters1) Mixed layer (ML) wind Avg. direction/speed (deg/kt)2) Ambient low level moisture

Surface dewpoint (°C); Max ML dewpoint depression (TdD) (°C)

3) Snow band width/length

>= 15 dBZ contour (n mi)

4) Niziol instability class Lake–air T(°C) at 700/850 hPa 5) Capping inversion Inversion height: top of ML (m)6) Vertical wind sheara. magnitude (0-1, 0-3 km)

Vector difference between wind at top and bottom of layer (kt) (from Storm Prediction Center)

6) Vertical wind shearb. direction/speed

Estimated values between surface and top of ML (deg/kt)

7) Low-level convergence

From 0-hour 12km NAM

8) Multi-lake connection?

Satellite data

Correlations Best IE parameters: (ALY) [BGM]

– Multi-Lake connection (MLC, Yes or No) = [0.73] (0.59)

– 850 mb lake-air differentials = (-0.64)

[-0.63]– Capping Inversion Hgts = (0.53)– 0-1 km speed shear = (0.44) [0.34]– 0-3 km speed shear = [0.03]

Bottom line: Lake-lake influence + Elevated ML + Cond/Mdt Instab + Strong 0-1 km flow = Deep IE (Type A)

Favorable IE (Type A)Example Sounding Corresponding Radar

Unfavorable IE (Type B)Example Sounding Corresponding Radar

Another Recent Type BRadar Image (22 Jan 13) IE Application Output

IE Application Example * Prior statistics / correlations boil down to a multi-element regression equation

* Produces an IE estimate for Lk Ont bands

IE Application Results For LES events that impacted

Central NY (WFO BGM), results were fairly similar to those from Eastern NY (WFO ALY)– W flow events (250-280 wind vector)

Mostly single-band cases– Usually good IE estimates (within 10-20

miles)– Tendency to under-estimate for deep inland

bands– NW flow events (290-330 wind

vector) Mix of single and multi-bands

– Under-estimated at times

Under-Estimate Source ? Which term(s) of the IE equation

were most problematic?– It would seem those dealing with

moisture availability A simple Y/N for MLC likely not adequate

– How do we improve?

“Effective Fetch” This concept was designed to

better represent moisture contributions from upstream lakes – Tries to simulate the additive effects

of long fetch distances

IE Term Adaptation

Choice of upstream lakes now available

Example from 6-7 Dec 2010 in WFO BGM’s Area

IE Output: 80 nm

Visible Satellite Image* Effective fetch across Georgian Bay/Lake Ontario was nearly 200 nmi* Modifying the MLC term improved the IE output substantially

Anticipating MLC Not necessarily intuitive in

forecast mode What tools can be provided to

help operational meteorologists in this regard?– Trajectories– Analogs / Composites– Simulated Satellite / Radar products

Forward Trajectory

Using Synoptic Analogs/Composite Plots

* Can help forecasters identify favorable patterns when extensive IE might be expected* Could conceivably stratify by flow regime / type of event

The Future Better visualization

– Efforts are underway to add a graphical component to the current AWIPS IE application

IE application could conceivably be adapted to other portions of the Great Lakes region

Do high-resolution models reasonably simulate inland extent ?

Questions?Joe.Villani@noaa.govMichael.Jurewicz@noaa.gov