Ice has become (pardon the pun) something of a hot topic lately.

Professional and amateur scientists alike have studied the timing of seasonal ice formation on lakes and rivers for hundreds of years, and the patterns that have emerged from these studies provide a window into the progression of climate change. Overwhelmingly, the data show that lakes and rivers are freezing up later in the winter and their ice cover is melting earlier in the spring than in the past.

Take Lake Suwa, in Japan. Observations of the lake’s winter freeze-up date back to 1443, when Shinto priests began to record the timing of the freeze-up as part of their religious observation. From 1443 through 1683, the date when the lake first froze over shifted later in the year by an average of 0.19 days per decade. More recently, between 1923 and 2014, the freeze-up date has been delayed by an additional 4.6 days each decade. The once-rare occasions when Lake Suwa remained ice free all winter have also increased in frequency: there were only three ice free years between 1443 and 1700, but this has happened twelve times between 1950 and 2004 (and five times between 2005 and 2014!).

Or consider Finland’s Torne River, whose ice breakup dates have moved an average of 0.66 earlier each decade since 1867. Or Lake Mendota in Wisconsin, USA, which has lost around 29-35 days of average ice cover since 1855.

Worryingly, scientists don’t yet fully understand just what we’re losing when lakes and rivers become increasingly ice free: a 2016 synthesis study of under-ice ecology warned that “we are losing ice without a deep understanding of what ecological processes are at stake”. Increasingly, scientists are working to describe and measure the impact that winter ice cover has on the year-round ecology of freshwater systems.

To be sure, we do already know quite a bit about life under the ice. Looking at a frozen lake in the wintertime, with no outward signs of life, you’d be forgiven for thinking that winter represents something of a “pause button” for aquatic ecosystems—but nothing could be further from the truth. You can think of the world underneath lake and river ice as a unique, ephemeral ecosystem that appears each winter and disappears each year as the ice melts. Many aspects of this ephemeral winter ecosystem are well understood, but others aren’t; the connections between winter ecology and the rest of the year are particularly under-studied.

Let’s start with some of the most well known aspects of under-ice ecology: animal species, and the adaptations that let them survive a harsh winter under ice. As anyone who has gone ice fishing knows, fish are active under the ice in winter. They still swim around, catching and eating prey—just, for most species, at a slower pace. To save energy, many fish species slow their growth rate and avoid unnecessary movements during winter. Some other fish species prefer colder water, and they tend to be pretty active in the winter, as long as the lake has enough oxygen under the ice to support their activity (lakes with a lot of organic matter can be low on oxygen; that’s why clearer cold lakes are sometimes more associated with coldwater fish species).

Other animal species have their own unique adaptations. Some amphibians and reptiles slow their metabolism and partially bury themselves in mud at the bottom of the lake. Many of these animals are aided by their ability to absorb oxygen through their skin. And some alter their body chemistry to essentially flood their veins with antifreeze, keeping their tissues and organs safe until spring arrives to thaw them out—some species can even let their heart stop beating in the winter, only to thaw out and awaken in the spring!

Some mammals have to deal with lake ice, too. Beavers might be one of the most fascinating examples—they spend the entire winter in lakeside shelters that they’ve constructed out of sticks and mud. The interior of these lodges are mostly above the surface of the lake, but the entrances are underwater. Once the lake ices over, beavers spend most of their time eating branches that they stockpiled in the lodge during the fall, though they also swim out under the ice occasionally in search of other food.

As fascinating as animal adaptations are, the under-ice lives of smaller life forms—plants, plankton, and algaes—may be even more interesting, because they are far less well understood. A major conclusion of the Hampton et al. 2016 study is that photosynthetic algae remain highly active under ice, especially in lakes without much snow cover to block sunlight through the ice. Many of these photosynthesizers actually hang onto the underside of the ice, maximizing their access to sunlight. These algae are foundational to lake food webs, and might play a key role in feeding new zooplankton, insects, and fish that hatch before many new spring plants and algae have had a chance to grow.

On the other hand, other studies point out that longer ice-free periods can result in a longer growing season, leading to overall greater productivity in lakes (O’Beirne et al. 2017, Creed et al. 2018). Of course, this longer ice-free growing period may come with unforeseen consequences, such as greater input of terrestrial plant matter, greater year-round input of water from precipitation and runoff, and changes in lake chemistry.

The number of unknowns in under-ice ecology, combined with the urgency of studying a system that is quickly shrinking, mean that we can probably look forward to many more studies of life under lake ice in the years to come.