For the first time, a team of climate and water experts has predicted patterns of declining ice cover in northern freshwater lakes for the next few decades.

Not a lake here and a few more over there — that’s been done, of course. The new work, published last week in the journal Nature Climate Change, looks at ice-loss scenarios for some 1.4 million lakes across our hemisphere.

And you needn’t be an ice-fishing enthusiast or lake-hopping snowmobiler to appreciate the bad news in their findings: “An extensive loss of lake ice will occur within the next generation.”

The focus here is not only on the relatively few, more southerly lakes that may stop freezing altogether as the globe continues to warm. It also includes the very large set of lakes where ice formation will continue, but in a reduced and unreliable way — starting later, ending sooner, with unpredictable cycles of thaw and refreeze along the way. One example, from the research announcement prepared at the University of Wisconsin-Madison Center for Limnology:

From 1862 through the winter of 1996, southern Wisconsin’s Lake Geneva, a prime destination for regional ice fishermen, froze over each year. But, since 1997, the lake has had four ice-free winters. Lakes from Alaska to Germany and Japan are experiencing similar trends.

Right now, according to the paper, about 14,800 lakes are experiencing such “intermittent” ice cover. How that number will grow depends upon how much the earth continues to warm:

• If the overall rise in global temperature is held to 2° Celsius, the increasingly unlikely ambition of the Paris accords, it would likely reach 35,300.
• If the current trend toward 3.2° continues, it would probably be 57,600.
• If the increase reaches 8°, the extreme but not implausible scenario in current policy discussions, the number of lakes with intermittent ice cover would swell to 230,400.

These figures are likely conservative, because researchers weren’t able to factor in an obvious feedback loop — that an open-water lake amplifies local warming in a way that an ice-covered lake does not. Therefore, they say, these calculations probably understate the likely rates of ice loss by at least a slight degree.

Ice as infrastructure

In a land of lacustrine plenty like Minnesota, we may think first of solid, season-long lake ice as a desirable recreational resource for fishing, pond hockey, snowmobiling, maybe paraskating or iceboating. In other places, recreational worth is less important than the ice cover’s role as critical infrastructure — the ice roads that carry Arctic peoples from town to town, and home to hospital — or as a cultural and economic as well as ecological cornerstone.

Changes in ice cover can drive negative changes in fish populations and the health of aquatic vegetation (not to mention patterns of algal bloom) as well as undesirable shifts in water chemistry and nutrient content. Moreover, the paper observes:

Loss of lake ice could contribute to decreased availability of freshwater owing to increased evaporation rates. Ecosystem services that are highly susceptible to loss of ice include transportation via ice roads, hydroelectric production, and provisioning.

Winter ice fishing can provide sustenance and economic growth to local communities (for example, ice fishing in Lake Peipsi, Estonia, can account for up to 40% of the annual fish harvest). In the USA, $178 million was spent on ice fishing equipment alone in 2011.

In addition, lake ice embodies the winter identity of many cultures, as evidenced by the abundance of winter ice festivals each year. For example, 25,000 people participated in an ice fishing competition in Finland, and up to 100,000 people attended one of the last remaining traditional ice fishing events in remote China, where the first fish caught is deemed auspicious and fetches tens of thousands of dollars.

Many other examples of cultural and socioeconomic consequences of losing lake ice are apparent, but these impacts have not yet been quantified or synthesized.

One broad-brush estimate from the paper: At the lower level of continued warming, the 2° scenario, some undesirable effects of reduced ice cover would be felt by 394 million people in 50 countries of the northern hemisphere. Under the 8° scenario, the number climbs to 656 million.

Simple process, complex system

Although the mechanism for lake-ice formation is  simple and unchanging — “vertical heat transfer” from water to air, until the surface temperature falls to freezing — the rate is influenced by a combination of many factors.

Some of these are obvious (mean air temperature, lake depth, prevailing precipitation and wind patterns). Others, not so much (elevation above sea level, which drives atmospheric pressure and evaporation rates; patterns of cloud cover; and, especially, complexity of shoreline, which influences wind speed over the water).

In modeling the relative strength of these factors, the research team had a large and welcome advantage in that so many communities around the world have kept such good records of freeze up and ice out for so long. An interesting if outlying example is Lake Suwa in Japan, where followers of the Shinto tradition believe their god crosses the ice in winter to visit his wife; their data on formation of a walkable ice ridge go back to the year 1443.

From these records, laid alongside mapping and meteorological data, the team discovered that the factors of air temperature, water depth, surface elevation and shoreline shape had the strongest correlations with ice-formation patterns. This led to a modeling exercise that, when applied to a sample of 513 lakes, correctly predicted their recent patterns of ice formation about 95 percent of the time.

Of  those factors, air temperature is the one most directly affected by global warming, and therefore it’s the more southerly ice-forming lakes where climate-driven losses will show up first and most strongly. (However, the paper notes, some lakes at very high latitudes may experience an additional influence — the amplification of climate change by continuing loss of Arctic sea ice.)

Apart from noting that “some lakes in historically cold regions, such as the northern United States, Norway and Sweden, are already beginning to lose annual lake ice,” the paper does not offer much in the way of regional detail.

However, one of the principal authors is John Magnuson of the University of Wisconsin at Madison, where he’s the emeritus director of the limnology program and has an office window looking out on an iconic water body familiar to many Minnesotans: Lake Mendota.

According to the school’s announcement of the research, this winter brought Magnuson an exceptionally rare sight:

He watched as the lake froze and thawed twice in December and January. Meanwhile, nearby and smaller Lake Wingra froze in early December and has maintained its ice. Mendota’s “fits and starts freezing,” he says, illustrates how climate change is expected to affect northern lakes.

And this is how the trend is likely to proceed, Magnuson explained:

“Lake Mendota’s not going to suddenly have no ice and then have no ice again the next year,” he says. “It’s going to have increasing proportions of years with no ice. It will have good winters for ice activity and winters with no ice activity. This is going to be a gradual process.”

It means millions of people will find themselves awaiting the big winter freeze they have come to depend on and turn to each other to ask: “Winter is coming … right?”

***

The full Nature Climate Change paper, “Widespread loss of lake ice around the Northern Hemisphere in a warming world,” can be read here without charge.