You are here

Ice Cover Summary

This page is part of the project: Ice Cover


  • A sudden shift to less ice occurred in many lake locations, but some regions have experienced no change
  • The abrupt shift occurred at different times across the region highlighting the non-linearity of the lakes' ice cover response 
  • High ice cover years are still possible and may occur consecutively to low ice years
  • Ice cover dynamics vary within each lake so impacts are localized 

Ice cover on the Great Lakes impacts people and places in a variety of ways from shipping navigation, aquatic ecosystem health, recreation to lake levels and beyond.  With almost 50 years of ice cover data, there's a long enough record to perform analyses and investigate historical ice cover trends.  Here we present a summary of Great Lakes ice cover research and explain how practitioners  (those who use/need ice cover information for planning) can interpret these findings when planning for future ice cover conditions on the Great Lakes. 

Physical Ice Drivers

Ice formation is influenced by the physical proprieties of the lakes and complex temperature-driven dynamics, which creates large variability in ice conditions day-to-day, week-to-week, and year-to-year.  The geographic placement of individual lakes makes them more or less susceptible to cold temperatures (north versus south), and lake depth plays a role in determining the lake's heat capacity or temperature "memory."  Northern deep lakes, like Superior, are less sensitive to climate forcing because of their longer memory of water temperatures (Wang et al 2012).  Comparatively, Lake Erie is so shallow that ice can form rather quickly when temperatures are cold enough.  Once ice does form, it acts as a reflector for incoming solar radiation that prevents additional warming of the lake.  For the northern lakes, ice cover strongly influences the timing of surface layer warm-up (stratification) of the lakes the following spring (Austin and Coleman 2007).  Reduced winter ice cover can lead to greater lake warming, like observed for Lake Superior summer water temperatures (Austin and Coleman 2007).  Water temperatures going into fall determine the magnitude of evaporation from the lake surface, which has a cooling effect on the water leading to higher amounts of ice cover (Figure 1).  For example, if the previous winter experienced low amounts of ice cover (more solar warming), higher evaporation rates (strong cooling effect) during the fall would lead to increased ice cover the next winter (Lenters et al 2013).  Oppositely, cooler water temperatures during fall leads to lower evaporation rates (less cooling) thereby decreased ice cover.  Interestingly, an extreme of one setup (i.e., high ice cover one winter) can lead to the opposite extreme (i.e., low ice cover) the next year. 

Weekly variations in ice cover can be larger than the seasonal mean indicating large "swings" in ice conditions can occur on short time scales (Wang et al 2012).  These swings may be due to extreme weather events controlled by internal climate patterns such as AO and ENSO (Wang 2012).  The Arctic Oscillation (AO) is a measure of atmospheric pressure differences, which determine where and how air masses move over the region on a daily basis.  Negative phase AO events can lead to increased ice cover.  The El Nino-Southern Oscillation (ENSO) is a climate phenomenon representative of warmer than average sea surface temperatures over the equatorial Pacific for an extended period of time and high air surface pressure over the western Pacific, which can bring warmer than normal winter temperatures to the Great Lakes Region and reduced ice cover. 

All of these ice cover drivers operate with varying influence, duration, and predictability.

Figure 1: Illustration of the interactions of lake temperatures, lake evaporation, and ice cover from Lenters et al (2013)

Regional Ice Trends

The majority of Great Lakes ice cover research has focused on large-scale (lake-wide) analysis, such as calculating the average amount (%) of ice on the lake surface for a given year.  There are many reasons this measure (and others like it) may not be useful to practitioners, but we will start our discussion at the lake-wide scale and provide more localized information to follow.  The lakes have experienced less ice cover on average during the last 20-30 years compared to earlier years, prior to the 1990s.  However, there remains strong variability year-to-year - meaning years with very little to a lot of ice are still possible.  The decline of ice cover on the Great Lakes is best explained by a more sudden shift to less ice as opposed to a gradual decline (Mason et al 2016).  Although the reason for the shift is not completely understood, Lake Superior's and Huron's shifts coincide with a strong ENSO event occurring in 1998.  The unknowns surrounding the physical drivers behind each lake's relatively abrupt shift in ice conditions highlights the challenge in predicting ice cover in the future (whether that's one year or many years ahead).  In a warming world there is less potential for large amounts of ice cover, but there are many forces at play that can still usher in winters of extreme cold leading to unprecedented high seasonal mean ice cover like in 2013 and 2014.                      

Local Ice Trends

Lake ice typically forms first near the shore and in protected bay areas and last over the deepest portions of the lake.  The figure below indicates how many days ice cover exists on average per year (winter season) across the region.  Near shore ice typically lasts the longest and areas of white indicate insufficient data. 

Figure 2: Great Lakes Average Ice Duration (days/winter) map for days with at least 90% lake surface ice coverage


The magnitude and timing of the recent shift to less ice cover varied spatially and in time across the Region.  The figure below shows that ice cover has decreased the most in the north (i.e., Lake Superior, Northern Lake Michigan and Huron) and in coastal areas.  There was no significant ice cover decline observed for Green Bay, sections along Lake Michigan's southern shoreline near Chicago, IL, much of Lake Huron's Michigan coastline, and a large part of Lake Erie.           

Figure 3: Change in seasonal ice cover duration (days) from 1973-2013 adopted from LaFond (2016) that was based on the Mason et al (2016) article

Future of Great Lakes Ice

This discussion of future Great Lakes ice cover relies on past observations for context.  Ice cover observations for the Great Lakes clearly show one distinct shift from higher amounts of ice cover to lower amounts of ice cover.  Since we only have one historical shift to analyze, and the reasons behind that shift are not completely understood, evidence is lacking to support the idea that another shift will occur in the future or if one does that the direction of the shift would continue to be towards less ice cover.   

There is strong evidence to suggest the Great Lakes region will continue experiencing warming air temperatures into the future especially during winter (Prior et al 2014).  How the Great Lakes respond to this warming is less certain.  Some of the Great Lakes have experienced warmer summer water temperatures (Austin and Coleman 2007, Mason et al 2016), but the role of lake evaporation creates a negative feedback on ice cover (warmer lake temperatures leads to greater evaporation which leads to increased lake cooling and potentially more ice).  

Predicting specific weather drivers, like ENSO and the AO, that effect weekly to seasonal temperatures in the Great Lakes Region is also very challenging.  In the case of the AO, an extreme negative mode (cooler) can be followed by an extreme positive (warmer) mode, highlighting the large amount of variability in the climate system.  Some researchers claim the unprecedented warming over the Arctic has and will continue to effect weather over the northern hemisphere, including the Great Lakes Region.  In the chain of response is the AO more often existing in the negative (cooler, more ice) phase, but limited observational data and questionable research methods underline the uncertainty regarding the future (Barnes and Polvani 2015).  Even climate models, which all simulate arctic warming, are in disagreement about how atmospheric circulation may respond (Barnes and Polvani 2015).              

Surprisingly, the future of Great Lakes ice cover is not as straightforward as one would think, especially as temperatures continue to rise.  The future may hold another shift in ice cover but not necessarily in the downward direction.  There is still the possibility of years with very high ice cover, like recently experienced (2013/14).  Practitioners should prepare for increased variability - high ice cover years followed by low ice cover years, and vice versa.  In lake regions where ice cover has not declined (gray areas in Figure 3), it may be beneficial to learn how other practitioners adapted to the recent shift to lower ice cover.  Most certainly, ice will continue to form first where it always has - in protected areas near the shore - but it may not persist for as long.     

Where to Get Ice Cover (and related) Data    

NOAA Great Lakes Environmental Research Laboratory (GLERL) hosts a suite of ice cover observations available here.  Some of the products that may be of interest to practitioners are shown below.

Maps of current ice conditions (click image)


Historical Ice Cover (Total for all lakes plus plots for individual lakes - click image)


This year's daily ice cover plotted (see "Graphs of Percent Ice Cover By Month for the Current Season")


Surface Water Temperatures by Lake for individual years (click image)



Austin, Jay A., and Steven M. Colman. "Lake Superior summer water temperatures are increasing more rapidly than regional air temperatures: A positive ice-albedo feedback." GEOPHYSICAL RESEARCH LETTERS. 34 (2007).

Barnes, E.A. and L.M. Polvani, 2015: CMIP5 Projections of Arctic Amplification, of the North American/North Atlantic Circulation, and of Their Relationship. J. Climate, 28,5254–5271, 

Great Lakes Region El Niño Impacts and Outlook | November 2015

Kaye Lafond. "Tracking Changes in Great Lakes Temperature and Ice: New Approaches." NOAA Great Lakes Environmental Research Laboratory’s Blog.  2016. Accessed February 15, 2018.

Lenters, J. D., J. B. Anderton, P. Blanken, C. Spence, and A. E. Suyker, 2013: Assessing the Impacts of Climate Variability and Change on Great Lakes Evaporation. In: 2011 Project Reports. D. Brown, D. Bidwell, and L. Briley, eds. Available from the Great Lakes Integrated Sciences and Assessments (GLISA) Center:

Mason, Lacey A., et al. "Fine-scale spatial variation in ice cover and surface temperature trends across the surface of the Laurentian Great Lakes." 138.1 (2016): 71-83.

Pryor, S. C., D. Scavia, C. Downer, M. Gaden, L. Iverson, R. Nordstrom, J. Patz, and G. P. Robertson, 2014: Ch. 18: Midwest. Climate Change Impacts in the United States: The Third National Climate Assessment, J. M. Melillo, Terese (T.C.) Richmond, and G. W. Yohe, Eds., U.S. Global Change Research Program, 418-440. doi:10.7930/J0J1012N. On the Web:

Wang, Jia, et al. "Temporal and Spatial Variability of Great Lakes Ice Cover, 1973-2010." JOURNAL OF CLIMATE. 25 (2012): 1318-1329.