By Sarah Anderson
Medill Reports

“Tropical glacier” —  the term sounds like an oxymoron and, due to climate change, it might become one.

These bodies of ice nestle in the mountain ranges of tropical regions, providing a major source of freshwater and tourism revenue. However, studies predict most tropical glaciers will disappear within the next 10 years.

“It’s hugely important to understand the rate at which these glaciers are melting,” said Alice Doughty, a lecturer of Earth and climate sciences at the University of Maine.

Doughty and Meredith Kelly, a professor of Earth sciences at Dartmouth College, are developing a model to investigate tropical glacier melt in places such as the Rwenzori Mountains in Uganda and the Sierra Nevada del Cocuy in Colombia. They presented their findings at the virtual 2021 Comer Climate Conference, an annual event usually held in southwestern Wisconsin.

Glaciers sand down the rock beneath them as they melt, acting “sort of like bulldozers,” Kelly said. The debris piles up, depositing a series of ridge-like features called moraines at the glacier’s retreating boundaries. Kelly examines satellite images of a glacier site to identify moraines the glacier left behind, then analyzes samples of rock to determine when the moraines were created. By measuring a type of beryllium atom that accumulates in the rock as it is exposed to Earth’s atmosphere, Kelly can approximate how long ago the rock was freed from the ice’s hold to form the moraine. Collectively, this information allows her to generate a map of the size and shape of the glacier at a specific time in the past.

Doughty then works to develop a computer model to simulate how climate variables interact to produce the glacier. She tries to “grow the glacier,” adjusting temperature, precipitation and other inputs until the glacier output matches Kelly’s map.

Data from any nearby weather stations provide a useful starting point; observing whether current climate conditions yield the modern glacier helps her evaluate the model. When the simulation is optimized, Kelly and Doughty will be able to use it to predict the effect of climate change on glacial melt.

“Once we calibrate the model, we can just as easily make things warmer,” Doughty said. “And so we can have estimates like in the Rwenzori, one degree of warming and those glaciers are gone.”

Other scientists are interested in developing similar models for the melting of sea ice, “a really big part of the climate system,” said Ed Brook, a professor of Earth, ocean and atmospheric sciences at Oregon State University.

Sea ice helps insulate the ocean from heat and gases in the atmosphere and contributes to sea level rise when it melts, but it doesn’t leave behind the same physical record as glaciers. While seasonal sea ice melting can be tracked using IP25, an organic molecule produced by algae that grow along the receding ice edge, the presence of permanent sea ice in the past has remained elusive.

“We don’t have a very good method for reconstructing how much sea ice there was at any particular time,” Brook said.

Frank Pavia, a postdoctoral researcher in geological and planetary sciences at the California Institute of Technology, presented his research at the Comer Climate Conference, exploring a new way to monitor this more stable sea ice cover. His method relies on interplanetary dust particles, the solar system’s version of dust that rains down on Earth from outer space.

Interplanetary dust particles deposit a light type of helium atom onto the sea floor. If the surface of the ocean is blocked by ice, however, the particles (and the helium) can’t enter the water. Pavia is examining whether the amount of helium in the ocean floor can be used as a measure of sea ice cover. To account for any differences in helium levels due to changes in how fast atoms settle to the sea floor, he also measures a special thorium atom that is produced inside the ocean and sinks to the bottom at a constant rate, regardless of ice cover.

To test his method, Pavia acquired samples of the floor of the Arctic Ocean from the Last Glacial Maximum, one period of well-characterized sea ice cover in the Arctic. The age of the samples had been previously determined by measuring a radioactive form of carbon in the shells of tiny marine organisms that indicates when they were alive.

When sea ice cover was thick, Pavia detected high amounts of thorium but low levels of helium, demonstrating that while atoms were efficiently burying in the sea floor, helium couldn’t access the ocean due to the sea ice. When there was no sea ice cover, he measured similar signals for both thorium and helium, revealing that the helium atoms were successfully deposited into the ice-free ocean. Pavia is also interested in seeing if the period of melting in between gives a surge in helium, corresponding to an influx of interplanetary dust particles that accumulated on top of the ice over time.

While accurate measurements will require that the dust particles are evenly distributed in the ocean floor rather than concentrated in specific pockets, the approach “has a lot of promise,” Brook said. The prospect of detecting nonseasonal sea ice melting — an event that leaves very few fingerprints — via a helium spike could be a major advantage of the method, he said.

“It’s potentially very important, because if the pulse of particles was the signature of melting a bunch of permanent ice, then you would have a sign of that process, which would be very hard to see other ways,” Brook said.

After further validation, Pavia plans to use his method to reconstruct poorly understood sea ice patterns during past periods of warming in the Arctic. Like Kelly, he aims to provide a map that other researchers can use to test and refine models that simulate sea ice melting as climate change progresses.

“The hope is to help improve the projections of sea ice coverage into the future in the Arctic,” Pavia said.

Sarah Anderson is a health, environment and science reporter at Medill and a Ph.D. chemist.  Follow her on Twitter @seanderson63.