Forecasting the Correct Path of a Lake Effect Band:
Determining Accurate Snow Totals


If you live in Syracuse, NY, lake effect snow is no stranger. It's not uncommon to see 3 to 4 feet of snow in one event. These extraordinary totals have won Syracuse the 'Golden Snowball Award' every year since the 2002-2003 season, marking them as the city to receive the most snow in all of Upstate New York.(11)

While snow can be inviting to winter enthusiasts, forecasting it can be frustrating. Thomas Niziol states, "a wind shift as small as 10 degrees can cause the heavy snow area to move a lateral distance greater than the width of the band".(4) This might cause unexpected heavy snow to fall in a city or town. For example, a band of lake effect forms to the East of Lake Ontario due to the strong winds from a low pressure system located to the north of Syracuse. As the low moves further to the Northeast and winds shift to the northwest. This causes the band to sink south and move into the Syracuse area, southeast of Lake Ontario. If you don't time the low's movement correctly or if low moves on a different path than forecasted, then wind speeds and/or direction will change. This will ultimately lead to the band moving and snow totals to be off. Areas to the north of Syracuse may see little if any snow and Syracuse will see several feet.

These circumstances occurred during a record breaking lake event, which started on December 9th, 1995. Intense bands formed to the east of Lake Ontario causing 40 inches of snow to pile up in Lewis and Jefferson counties, both north of Syracuse. Two days later the lake bands were still holding strong, however the winds made a slight shift, causing the band to move south. Areas in Oswego county and southern Lewis counties now were getting hit hard and picked up 19 inches.(6)

Lake effect snow develops when cold air passes over warmer lake water. The water adds heat and moisture to the lower levels of atmosphere, causing air to rise into colder temperatures aloft. Ultimately, ice crystals will form. In order for the snow to fall, the air must pass at least 80 km across the lake surface.(3,7)

There are a number of factors that contribute to the development of lake effect snow. These include: temperatures differences between the lake and air, lake instability, additional energy (latent heat), and a deep arctic air mass with low relative humidity. Two other key ingredients are wind speed and direction. (3)

Forecasting incorrect snow totals happens quite often in Syracuse not only because of ever-changing weather conditions, but also of how often the city experiences lake effect. Syracuse is positioned in Lake Ontario drainage basin, which is an area downwind of Lake Ontario.(8) Another name for this is the snowbelt region. These areas are most prone to heavy snow during the lake effect season, which usually starts in the late fall and continues through the early winter before Lake Ontario has had a chance to freeze.(10)

The National Weather Service does an excellent job keeping the public informed about lake effect events. Lake effect watches mean conditions are likely for lake snow to develop. These are issued up to 36 hours before an event. When a lake effect snow warning is issued, either more than seven inches of snow is expected in a 12 hour period or more than nine inches in a 24 hour period. A lake effect advisories means anywhere between four to seven inches of snow is expected in a 12 hour period.

While using National Weather Service alerts to keep you aware, it's always best to do your own forecasting. An excellent way to start is by looking at weather charts. You need to determine if conditions are likely for lake effect. Look at atmosphere temperatures at 850 millibar chart and subtract it from and temperature of the lake. If the difference is 13 degrees of larger, conditions are very good for lake effect snow.(4)

It's also beneficial to look at wind speed in different levels of the atmosphere, starting from the surface up to 700 millibars. Small directional wind shear is needed for a large lake snow event. Many forecasters suggest using ETA models, which have become a reliable source for forecasting lake effect development and movement. NGM models are more detailed with a 80-km horizontal resolution, however this model takes information from less layers of the atmosphere compared to ETA models.(5)

Another step in forecasting lake effect is using BUFKIT. According to National Weather Service, BUFKIT is "a forecast profile visualization and analysis tool kit", in which you download to import information. It was developed in the mid-90s to help meteorologists forecast lake effect events. There are detailed, hourly soundings from various cities where you can determine wind speeds, height inversions and shear, and amount of moisture for all levels of atmosphere.(1)

There are also maps that show movement of lake bands. The movement, along with low and mid level wind speed and direction, is displayed each hour. Going through each hour of the event, you can gather information to help you determine if the band will move north, south or stay stationary.(1)

Short term forecast are critical in forecasting as well. December 18, 1985 a low pressure passed to the north of Lake Erie causing an intense band of lake effect to form off the lake. This was expected, however the band formed three to four hours earlier due to changing weather conditions. By the time the band was supposed to form it was already between 50 to 200 km wide.(4)

This is why you should update your forecasts every hour with current conditions. To do this use data from satellites, Doppler radar and snow spotter forecasts. It's key to use both the satellite and radar together because sometimes bands are too small or are located too low in the atmosphere for radar beams to detect.(9)

With the exception of forecasting correctly, one way forecasters can make a more accurate forecast is using caution when forecasting totals. Give a range opposed to exact amounts. One example Thomas Niziol uses is "1 to 3 inches of snow, but 6 inches or more tonight in persistent squalls".(4)

Following these steps will lead to a more accurate snow total forecast, which is a plus for the forecaster. It also can help save lives and property. If people know their town or city will see several feet of snow in a 24 hours period, they know to stay inside and off the roads.


(1) Badini, William; Hoggart, Bradley; Mann, Greg; Sousounis, Peter, Wagenmaker, Richard; Young, George. Forecasting during the Lake-ICE/SNOWBANDS Field Experiments: December 1999.
014%3C095 5%3AFDTLIS%3E2.0.CO%3B2#I1520-0434-14-6955HOKE1%20oppsed

(2) Jackson, Scott and Nolan, Sean. Using BUFKIT to Display and Analyze Meteorlgoical Data: 13 February 2005.
sunday/ courses/jackson_bufkit.ppt+what+is+BUFKIT&hl=en&c=clnk&cd=6&gl=us

(3) National Oceanic & Atmospheric Administration. Question of the Month: 16 January 2002.

(4) Niziol, Thomas. Operational Forecasting of Lake Effect Snowfall in Western and Central New York: 14 September 1987.

(5) Niziol, Thomas and Mahoney, Edward. The Use of High Resolution Hourly Forecast Soundings for the Prediction of Lake Effect Snow.

(6) Northeast Regional Climate Center, Cornell University. More Cold and Snow for the Northeast.

(7) Ostro, Steve. Historic Snowfall for the Niagara Frontier: 12 October 2006.

(8) The American Heritage Dictionary of the English Language, Third Edition. Lake Effect snow: 1992.

(9) University Corporation for Atmospheric Research. Topics in Lake Effect Forecasting: 2005.

(10) Wikipedia. Snowbelt.

(11) Wikipedia. Golden Snowball Award.