1	Lake-effect Snow NOAA/NWS CLE Winter Weather Workshop 10 November 2005 Adapted from materials from Tom Hultquist(MQT), Dr. Greg Mann (DTX) and Tom Niziol (BUF) Robert LaPlante (CLE)
2	Overview and Objectives Lake-effect primer Forecasting Tools/Guidance Become familiar with the processes responsible for the formation of shallow intense convection Become familiar with tools available, and limitations of those tools.
3	Lake Effect Snow Characteristics Mesoscale Convective Snow Events Develop as polar/arctic air travels across warmer water, picks up heat and moisture, and is destabilized Occur from late fall through winter (though lake effect rain can occur from late summer through mid fall). Produce tremendous snowfall amounts and gradients of snowfall. Rarely produces numerous fatalities directly but are very disruptive to commerce and transportation.
4	Lake Effect Snow Characteristics Localized instability (lapse rate and BL depth) Fetch Wind direction and shear Cloud microphysics Synoptic (large)-scale forcing Orography/topography lake shape and orientation, lee-shore topography Upstream moisture / Upstream lakes Great Lakes aggregate (upscale/downscale forcing) Snow/ice cover on the lake
5	Mean Land vs. Lake Temperatures for Lake Erie at Buffalo, NY. from Eichenlaub (1979)
6	The vertical scale of convective lake effect snows (Great things come in little packages)
7	The vertical scale of convective lake effect snows (Great things come in little packages)
8
9	Conceptual Model of Lake-Effect Heat and moisture from lake + frictional convergence + upslope flow = clouds and lake-effect precipitation
10
11	Formation Regions Downwind of concave coastline, bays, etc.
12
13
14
15
16	Morphology of the eastern Great Lakes Shape – elliptical Orientation – major axis Size – smaller relative to the western Lakes Depth – Erie (shallow which cools fast) Type of bands Single and multiple band events are common Lake Huron often a source of multi-lake bands Mesoscale lake vorticies rare
17
18	Single Band Development Develop down the long fetch of elliptically shaped lakes. Associated with “deep” convective layer. Produce the most intense, heavy snowfalls. Tremendous latent heat release, producing a mesolow core. The deeper the clouds, the stronger the low (warm core). Accompanied by strong low level convergence zone. Horizontal thermal convergence below cloud base and diffluence aloft leads to strong mesoscale ascent within the snowband.
19	"Single Band Storm" Single Band Storm This storm off Lake Ontario produced a very narrow band of heavy snow. Convergence Zone Single band snows are often associated with very strong convergence zones with as much as a 90° wind shift on either side of the band.
20	Single Band Characteristics from Byrd et al. (1991) The snow band is self-sustaining and persistent. Within the band the temperature may be 2-3 C warmer than the surrounding environment.
21	Multiple Snowbands from NWS Marquette (1996) Weaker than single bands, shallower mixed layer Horizontal roll convection Occur when mean boundary layer wind is more normal to the long axis of the lake Little thermodynamic difference with environment Oriented parallel to the mean boundary layer wind direction
22
23	Multiple-Lake Bands May be single or multiple, commonly initiate off Lake Huron/Georgian Bay in northwest flow.
24	Meso-Vortex Occur on bowl-shaped bodies of water (Lakes Michigan, Huron and Superior). Develop under weak synoptic flow and strong lake-air temperature contrasts. (Arctic Highs). Do not, as a rule, produce significant snowfall amounts.
25	Instability and Moisture Lake-induced lapse rate: T_lake-T850>13 C gives absolute instability/vigorous heat and moisture transport, but does not address the depth of the mixed layer. Depth of instability: Defined by the capping subsidence inversion, combined with the lapse rate determines the depth to which convective clouds can grow. Difficult to get heavy snow if depth of mixed layer < 1.0 -1.5 km. Moisture: Impacts precipitation potential. Low RH: difficult to get condensation, clouds, and precipitation. High RH: produce cloud bands and snowfall much more quickly.
26	Example of the Capping Subsidence Inversion That Limits Convective Cloud Growth in Polar Airmass
27	Instability Lake Induced CAPE - better estimate of static stability over the lake. Surface air temp over the lake is calculated from an average of the water temperature and surface air temperature over land. Available within BUFKIT.
28
29	Wind Direction, Fetch and Shear Wind direction within the cloud layer determines the orientation and type of snow band. The longer the over-water fetch, the greater the overall flux of heat and moisture into the airmass. Small changes in wind direction can significantly change the fetch. Lake Erie: 250 deg wind--225 mi fetch 230 deg wind-- 80 mi fetch Directional wind shear through the cloud layer leads to disorganization of snow bands, especially on the eastern Great Lakes.
30	Favorable Fetches for Lake-Effect Snow from LaDue (1996)
31
32	Upstream Lakes Impact snowfall to the lee of downwind lakes, e.g., In northwest flow, Lake Huron snow bands re-form/intensify over Lake Ontario and Lake Erie. Upstream lakes may not always bring an increase in snowfall to downwind lakes!
33	Synoptic-Scale Forcing Cold advection may enhance lake-effect by increasing the instability Cyclonic vorticity advection aloft may enhance lake-effect by raising the capping inversion
34	Favorable Synoptic Setting for Eastern Lakes Niziol (1987) Sfc-850: Broad WSW trof extension from parent low over Maritimes. 500: Closed low south of James Bay; Deep (>3 km) layer of instability. Result is prolonged, unidirectional fetch over long axes of Erie, Ontario.
35	Lake-Effect vs. Lake-Enhanced Lake-effect: precipitation which results from cold polar air flowing over warm lake water after passage of a synoptic cyclone Lake-enhanced: the additional precipitation resulting from a boundary layer fetch over a lake during a synoptic cyclone. (e.g. Isentropic lift, seeder/feeder ) event.
36	Snow/Ice Cover on the Great Lakes Diminishes the lake-effect intensity. Lake Erie LES weakens after Lake Erie freezes usually from late January to early March A frozen lake doesn’t preclude a significant lake-effect event.
37
38	Orography/Topography Snowfall increases with elevation to the lee of the lakes In general, annual snowfall increases by 8-12 inches per 100 ft increase in elevation
39	10 yr Mean Annual Snowfall
40	"Forecasting Lake-Effect" Forecasting Lake-Effect Lake-effect snowstorms are difficult to observe and forecast for the following reasons: They are shallow systems (depth often < 3 km); and the lowest elevation radar scans overshoot the tops. The onset, intensity, orientation, and exact location are very sensitive to wind shear/direction and thermal stratification in the lower troposphere. Conventional rawinsondes measure profiles at times and locations which are not optimum for monitoring the atmosphere over the lakes. Operational models do not have sufficient resolution, microphysics etc. to resolve the scales of lake-effect snowbands.
41	Summary of Guidance Support Models are doing much better at predicting snow band location and movement. Models underforecast snowfall in BIG lake effect events. Also tend to under estimate the depth of the convective boundary layer in nearly all instances Models do not allow snow to fall far enough inland. Other tools such as BUFKIT provide added information on features not handled well by models. (eg. snow crystal microphysics).
42	Forecasting Guidance Available Guidance NCEP Model Suite CMC Model Suite (including HIMAP) Local Models (WsEta, WsWRF) etc… Caveats When it comes to LES, horizontal resolution is the greatest limiting factor in a models ability to make an accurate prediction. However, even with very high resolution, limitations in model physics, microphysics, and the use of parameterizations still result in significant errors at times. In addition, one must remember that a poor large scale forecast will almost always result in a poor high resolution model forecast from any model which utilizes that large scale output.
43	Forecasting Guidance Caveats High resolution models are better able to predict land/lake interactions and topographic forcing, as well as lake aggregate effects (given sufficient domain size). However, the ability of these models to depict individual snow bands is limited. Forecast soundings, stability progs, mass fields (wind, divergence), and omega can be of much greater value than things such as QPF. Although QPF can provide some indication of areas which may be impacted and the general threat level, it should rarely be taken at face value.
44	Forecasting Guidance Derived vs. Raw NWP Output
45	Forecasting Tools Current MQT decision tree was developed by Dockus, Wagenmaker, and others in the 80s, and revised by ARB/DTX staff in the early 90s. Lake-850mb delta-T Fetch Synoptic scale lift Inversion height APX decision tree is similar, but includes a check for precip type (ensure that heterogeneous nucleation is possible). These simple tools do not account for more recent research, and were not developed in the age of high resolution model data (input now can give different results). Can still be useful, though, since they are well tested and easily understood.
46	BUFKIT Enhancements Snow Growth Microphysics Favorable Temperature Range for Dendritic Snow Crystal Growth. High Relative Humidity Profile at Cloud Level. Significant Omega Field Provides Localized Lift. As a result, dendritic snow growth is maximized and snowfall rates are increased dramatically.
47
48
49
50
51
52	LES Summary Operational lake effect snow forecasting is a very exciting challenge. There is still a tremendous amount of knowledge we just don’t know. New model suites and other information are improving the forecasts. Key parameters to keep in mind: Degree of instability Environmental RH Depth of moist convection Shear & fetch Orography Meso & synoptic scale forcing Cloud microphysics
53
54	Panoramic View of a Lake Effect Snowband Over the Western End of Lake Ontario