The Greenland Ice Sheet is the second largest ice sheet in the world and it's melting rapidly, likely driving almost a third of global sea level rise. A new study shows clouds are playing a larger role in that process than scientists previously believed.

"Over the next 80 years, we could be dealing with another foot of sea level rise around the world," says Tristan L'Ecuyer, professor in the Department of Atmospheric and Oceanic Sciences at the University of Wisconsin-Madison and co-author of the study. "Parts of Miami and New York City are less than two feet above sea level; another foot of sea level rise and suddenly you have water in the city."

The study, published in Nature Communications and led by the University of Leuven in Belgium, shows that clouds are raising the temperature of the Greenland Ice Sheet by 2 to 3 degrees compared to cloudless skies and accounting for as much as 30 percent of the ice sheet melt.

Numerous statements in the Nobel Peace Prize-winning 2007 Intergovernmental Panel on Climate Change report address the need to better account for clouds in climate models, L'Ecuyer says. Arctic clouds are no exception, especially since climate models have not kept pace with the rate of melting actually observed on the Greenland Ice Sheet.

"With climate change at the back of our minds, and the disastrous consequences of a global sea level rise, we need to understand these processes to make more reliable projections for the future," says Kristof Van Tricht, the University of Leuven graduate student who led the study. "Clouds are more important for that purpose than we used to think."

But in order to better understand them, the right technology needed to be in place.

"Within the last 10 years, NASA launched two satellites that have just completely changed our view of what clouds look like around the planet," says L'Ecuyer, who is affiliated with the UW-Madison Space Science and Engineering Center, where satellite meteorology was born. "Once you know what the clouds look like, you know how much sunlight they're going to reflect and how much heat from Earth's surface they're going to keep in."

When it comes to heat, clouds essentially behave in two ways: They either cool the Earth's surface by reflecting sunlight back into space, or, like a thick blanket, they trap heat at the surface - the greenhouse effect of clouds. On Greenland, which is covered in bright, light-reflecting snow, clouds primarily act to trap heat.

Using the two satellites - CloudSat and CALIPSO - L'Ecuyer was able to take "X-ray images" of Greenland's clouds from space between 2007 and 2010 and determine their structure, how high they were in the atmosphere, their vertical thickness, and their composition (ice or liquid).

The Belgian team combined this data with ground-based observations, snow model simulations and climate model data to map the net effect of clouds. They learned that cloud cover prevents the ice that melts in the sunlight of day from refreezing at night.

"A snowpack is like a frozen sponge that melts during the day," says Van Tricht, who spent six weeks in Madison last year working with L'Ecuyer. "At night, clear skies make a large amount of meltwater in the sponge refreeze. When the sky is overcast, by contrast, the temperature remains too high and only some of the water refreezes. As a result, the sponge is saturated more quickly and excess meltwater drains away."

Researchers already know that while clouds can change the climate, the climate can also change clouds, a phenomenon known as cloud-climate feedback. L'Ecuyer is optimistic that the study - a good example of how satellites are helping us solve the complicated cloud-climate feedback problem - will improve future climate models, to help scientists and policymakers across the world adapt to climate change.

With a background in physics, L'Ecuyer is driven to study clouds by a desire to better understand how people and society are affected by the natural world. "Many of the countries most susceptible to sea level rise tend to be the poorest; they don't have the money to deal with it," he says. "This is something we have to get right if we want to predict the future."