"Lake-effect" precipitation is a common event in the state of Michigan, occurring most often during the late autumn and early winter months. This form of precipitation is the result of rapid warming and moistening of cP (continental polar) air masses that regularly move southward from central Canada (Kunkel, 2000). The "moistening" occurs when water from the Great Lakes is evaporated into the cP air mass, causing an increase in the dew point temperature. The "warming" of the air occurs as a result of the temperature contrast between the cP air mass and the lake surface. The unfrozen waters are relatively warm when compared with the temperature of the wintertime air mass. Therefore, the temperature of the air that comes into contact with the water increases. The warmed air expands and become less dense, which causes it to rise. This is an "unstable" situation. As the air rises, the temperature decreases until it reaches the dew point, which is the temperature at which the air becomes saturated.. Ice crystals or water droplets will then begin to collect until the force of gravity pulls them down. The result is "lake-effect" precipitation. When the cP air mass is very cold, as is often the case between December and February, the precipitation falls as snow. During late autumn, however, the polar air mass may be warm enough for the precipitation to fall in the form of rain.
"Lake-effect" precipitation can cause substantial intensification of snowfall amounts in very narrow bands, often referred to as "snow belts," along the leeward (downwind) shores of the Great Lakes. The prevailing wind direction in the Great Lakes region is westerly; therefore, most "lake-effect" precipitation events occur to the east of the lakes. Consider, for example, the following examples:
(1) Detroit, Michigan lies on the western (upwind) shore of Lake Erie and averages about 42 inches of snow per year; however, Buffalo, New York lies on the eastern (downwind) shore of Lake Erie and averages about 92 inches of snow per year (Kunkel, 2000).
(2) Toronto lies on the northwestern (upwind) shore of Lake Ontario and averages about 54 inches of snow per year; however, Syracuse is located along the southeast (downwind) shore of Lake Ontario and averages about 109 inches of snow per year (Kunkel, 2000).
(3) Escanaba, Michigan lies on the northwest (upwind) shore of Lake Michigan and averages about 50 inches of snow per year; however, Kalamazoo, Michigan, located much further south but approximately 40 miles east (downwind) of the Lake Michigan shoreline, averages about 74 inches of snow per year (Eichenlaub, 1990).
(4) Milwaukee, Wisconsin lies on the western shore (upwind) of Lake Michigan and received an average of 52.6 inches of snow for the period 1971-2000; however, Muskegon, Michigan, located along the eastern shore (downwind) of Lake Michigan, received an average of 106.0 inches of snow for the same time period.
An interesting feature of "lake-effect" is that the heaviest bands of snow do not usually occur along the immediate shoreline, but tend to fall several miles inland. Snowfall accumulations are enhanced inland because the air experiences more uplift when it is forced over hills and higher terrain. Notable areas of enhancement in the state of Michigan due to elevation include the interior of the northern portion of the Lower Peninsula and the higher terrain of the northern portion of the Upper Peninsula. A comparison of snowfall amounts between Gaylord, Michigan and Charlevoix, Michigan can be used to illustrate this phenomenon: Gaylord is located several miles east of the Lake Michigan shoreline in the elevated interior of the northern Lower Peninsula. The station is 1,350 feet above sea level. An average of 149.6 inches of snow fell at Gaylord for the period 1971-2000. Charlevoix is located to the west of Gaylord but immediately along the eastern shoreline of Lake Michigan. The station sits at an elevation of 593 feet above sea level. An average of 117.8 inches of snow fell at Charlevoix for the period 1971-2000, which is almost 3 feet less than what was observed at Gaylord. Higher elevation therefore produced an intensification of snowfall accumulations at the Gaylord station.
Unique geographic features and the shape of the shoreline also play major roles in the substantial intensification of snowfall amounts in very narrow bands. There are four notable examples of this in Michigan that can be found in the following areas: the Keweenaw Peninsula, the "thumb" area, the Leelanau Peninsula, and the southwest portion of the Lower Peninsula.
One does not need to stare at a map of the Keweenaw Peninsula for very long to realize that this is one of the most unique geographic features in the Great Lakes region. Located in the Upper Peninsula, the Keweenaw Peninsula shoots straight out into Lake Superior and then curves to the east at its northern end. Thus, it is practically surrounded by the waters of Lake Superior. This leaves it "vulnerable" to massive "lake-effect" snowstorms that can hit from almost any direction. Whether the wind blows from the west, northwest, north, northeast, or east, the Keweenaw will experience "lake-effect" snow if the conditions are right. No other locations in the Great Lakes region (except possibly certain islands) are downwind from so many directions. Several stations within the Keweenaw Peninsula average well over 200 inches of snowfall per year! In fact, the largest seasonal snowfall ever recorded in Michigan was recorded in the Keweenaw. A State Highway Snow Station at Delaware received a total of 391.9 inches of snow during the 1978-1979 winter season (Eichenlaub, 1990).
The Leelanau Peninsula is a fairly small peninsula located in the northern portion of the Lower Peninsula. As such, it is downwind of Lake Michigan when the winds are flowing out of the southwest, west, northwest, and north. This provides many opportunities for "lake-effect" snowfall during the winter months. In fact, average snowfalls in this peninsula are one of the highest for a Lower Peninsula location. For example, the average annual snowfall at Maple City, located in the center of Leelanau County, was 151.3 inches for the period 1971-2000. This is quite astonishing, considering that few other region in the Lower Peninsula receives this amount!
The "thumb" is probably the most recognizable feature of the state of Michigan because it gives the appearance that the Lower Peninsula is shaped like a hand. Not only is the "thumb" an interesting geographic feature, but it is also an interesting region to study for "lake-effect" snow. This peninsula is located in the eastern portion of Michigan north of Detroit. It sticks out into Lake Huron like a sore thumb (pun intended!) leaving it "vulnerable" to lake-effect snowstorms when the winds are blowing from the north or northeast. Snowfall amounts in the "thumb," however, are not as dramatic as what is found in western portions of the Lower Peninsula. Harbor Beach, located in Huron County, averaged only 67.0 inches of snow over the period of 1971-2000. Nevertheless, this is greater that what was observed in areas south and east of the "thumb." For example, Saginaw, located to the southeast of the "thumb" in Saginaw County, averaged only 45.7 inches of snow over the period of 1971-2000. This was over 20 inches less that what was observed at Harbor Beach. The snow belts are not as intense in the "thumb" area because the prevailing wind direction in Michigan is westerly. A westerly wind over the "thumb" is upwind of Lake Huron. Thus, the "thumb" does not have as many opportunities for "lake-effect" snow. Nonetheless, there are enough instances of easterly and northerly winds to create a significant enhancement of snowfall amounts in the "thumb" area.
In the southwest portion of the Lower Peninsula of Michigan, there is a narrow strip of land in which the average annual snowfall exceeds 100 inches (Eichenlaub, 1990). This snow belt is centered at 10 to 15 miles inland from the shoreline of Lake Michigan. The northern extent of this snow belt is slightly to the southeast of Hesperia in Oceana County. The southern extent is located slightly to the northeast of South Haven. It should be noted, however, that the total area effected by "lake-effect" is much wider in extent than this narrow band. Surrounding the narrow 100 inch band are areas that average 80-100 inches of snow per year (such as Allegan in Allegan County and Bloomingdale in VanBuren County). There are also areas in southwest lower Michigan that average 70-80 inches of snow per year (such as Grand Rapids in Kent County and Kalamazoo in Kalamazoo County). The cause of the narrow 100 inch snow belt is not immediately obvious. There are no major areas of high terrain in southwest Michigan, and there are no peninsulas that "stick out" into the Lake. What, then, is causing an increase in snowfall totals? A close inspection of a map of Lake Michigan reveals a "bulge" in the southern portion of the lake. The southern part of Lake Michigan is several miles wider than what is found further north. Thus, the cold, polar winds are taking significantly longer to cross the lake. This allows for more evaporation and more warming of the air leading to more uplift. As a result, the increases in snowfall accumulations in southwest lower Michigan can be quite dramatic. White-out conditions are often experienced by travelers on US-131 between Kalamazoo and Grand Rapids, and on US-31 in and around the Holland area.
Finally, it should be noted that instability over the Great Lakes can substantially increase snowfall amounts received from synoptic scale low pressure systems. When a cyclonic storm system moves over the Great Lakes region during the cold, winter months, it often deposits large amounts of snow on all sides of the lakes, both downwind and upwind. However, significant accumulations of snow can be added to regions downwind of the lakes. Thus, a storm that is already producing excessive amounts of snow can drop substantially greater amounts when it strikes locations downwind of all five Great Lakes. This is referred to as "lake-enhanced" snow.
Note: The 1971-2000 snowfall climatic normals presented in this section were taken from the web site of the Midwest Regional Climatic Center.
The findings presented in this paragraph are based on Eichenlaub (1986) and Eichenlaub, et. al. (1990). A well-documented climatic change has occurred over the Great Lakes region involving "lake-effect" snow. A sharp increase in snowfall amounts was observed over the second half of the 20th century at many stations located along and near the downwind shores of the Great Lakes. A gradual increase in snowfall began to occur at Calumet-Houghton (located on the Keweenaw Peninsula) after 1940. In fact, by the early 1980’s, ten-year moving snowfall averages for this station were approximately 100 inches greater than what was observed during the first half of the 20th century. It has been noted, however, that there was a site change for this station in 1948. This implies that some of the snowfall increase was due to a lack of site homogeneity. However, the records from other stations located in the Michigan "snow belts" are consistent with the trend observed at Calumet-Houghton. A large increase in snowfall also occurred at Muskegon, a station located near the eastern shore of Lake Michigan. Prior to 1950, ten-year moving snowfall averages at this station were around 60 inches. However, these averages gradually increased during the 1950’s. Ten-year moving averages at Muskegon gradually increased to between 100-130 inches during the next two and a half decades. In contrast, stations outside of the "lake-effect" snow belts did not see dramatic increases in snowfall. The Michigan state capitol of Lansing, for example, showed very little increase in snowfall amounts between 1940 and the early 1980’s. The dramatic increases in snowfall observed at many stations along and near the lee shores of the Great Lakes were attributed to "an increase in lake effect snow associated with colder winters from the 1950’s to the late 1970’s, but may also have an anthropogenic contribution in the form of heat and pollution from increased urbanization and industrialization" (Eichenlaub et. al., 1990).
After 1980, Michigan escaped the extremely cold weather pattern that dominated during the 1970’s. However, snowfall amounts remained high in the "lake-effect" zones. This can be demonstrated by examining climatological data from Grand Rapids, Michigan. I compiled annual wintertime (December through February) mean temperature averages for all winters from the 1979-80 season through the 2010-11 season. Additionally, I compiled annual wintertime (December through February) snowfall totals for all winters from the 1979-80 season through the 2010-11 season. The mean wintertime temperatures and annual winter-month snowfall totals are presented on the following graph. .
Please note that the snowfall totals appear to be low because they are not the totals for the entire winter season; they are the totals for only the three months from December through February. The graph indicates that the overall trend of snowfall at Grand Rapids is upward (p-value = 0.000). This is occurring despite the fact that there was no significant decrease or increase in temperatures over the period (p-value = 0.668).
[About the author: Robert J. Ruhf received his Ph.D. in Science Education from the Mallinson Institute for Science Education at Western Michigan University in December 2006. He received a Master of Arts degree in Geography with a concentration in Environmental and Resource Analysis from Western Michigan University, a Bachelor of Science degree in Meteorology from Central Michigan University, and a Bachelor of Arts degree in Communications from Cornerstone University. He currently works for Science and Mathematics Program Improvement (SAMPI) at Western Michigan University.]
References
Eichenlaub, V. (1986). Some Recent Climatic Fluctuations in Michigan. Michigan Earth Scientist 22, 3-9.
Eichenlaub, V., J. Harman, F. Nurnberger, & H. Stolle. (1990). The Climatic Atlas of Michigan. Notre Dame, Indiana: University of Notre Dame Press.
Kunkel, K., N. Westcott, and D. Kristovich. (2000). Climate Change and Lake Effect Snow. In Preparing for a Changing Climate: The Potential Consequences of Climate Variability and Change. A Report of the Great Lakes Regional Assessment Group for the U.S. Global Change Research Program. Ann Arbor: University of Michigan.