By BlueShift | 2026-02-20

Researchers at the British Antarctic Survey (BAS) have developed a first-of-its kind artificial intelligence (AI) tool that uses satellite imagery to track and catalogue the lifecycle of icebergs.

The new system identifies individual icebergs, assigns each a unique identity and follows it over time, as the massive ice formations drift, melt and fragment across the world’s oceans.

This has not been previously possible at scale, BAS said February 5.

Satellites typically track upwards of 100 major calving events (larger than one square kilometer) annually. But the true number of smaller icebergs is likely in the many thousands, and possibly millions.

Icebergs more than 80km in length have broken away from Antarctic shelves, according to NASA. Smaller chunks of ice from glaciers are called bergy bits (less than five meters) and growlers (less than two meters).

Bergs release enormous volumes of freshwater when they melt on the open water, altering ocean currents and ecosystems. But scientists have long struggled to keep track of these floating behemoths once they break up.

BAS's new AI system automatically identifies and names individual icebergs at birth and tracks their sometimes decades-long journey.

Using satellite images, the tool captures the distinct shape of icebergs as they calve from glaciers and ice sheets on land.

Bergs typically come in two shapes: tabular, with steep sides and a flat top, and non-tabular, which come in a variety of shapes with domes and spires.

As they disintegrate over time, the machine performs a giant puzzle problem, linking "child" fragments back to the "parent" and creating detailed family trees.

Iceberg 'family trees'

The tool is a huge improvement on existing methods, where scientists pore over satellite images to visually identify and track only the largest icebergs, one by one, AFP reports.

The new AI system, tested using satellite observations over Greenland, provides "vital new information" for scientists, according to the British Antarctic Survey.

Knowing where these giant slabs of freshwater were melting into the ocean is especially crucial, it said.

"What's exciting is that this finally gives us the observations we've been missing," Ben Evans, a machine learning expert at the British Antarctic Survey, said in a statement.

"We've gone from tracking a few famous icebergs to building full family trees," he said, adding that for the first time, scientists can see where each fragment came from, where it goes and why that matters.

This use of AI also could be adapted to aid safe passage for navigators through treacherous polar regions littered with icebergs.

Tracking icebergs of all sizes is vital for safety at sea.

BAS remote-sensing expert Andrew Fleming says that while satellites can easily spot city-sized icebergs, it’s the smaller ones that endanger ships.

"The real risk is the ice you can barely see – fragments that lurk just below the surface. Hit one at speed, and you’re in serious trouble," he said.

In addition to working with the AI lab, Fleming is working toward an automated Antarctic iceberg monitoring system, also using satellite data.

For smaller icebergs, this would map likely ice density rather than individual blocks, helping vessels avoid hazardous waters, via a system modeled on one that already operates around Greenland.

Insights into glaciers

By analyzing millions of optical and radar satellite images collected between 2014 and 2022, glaciologists Chad Greene and Alex Gardner at NASA’s Jet Propulsion Laboratory have mapped changes in glacier flow on a global scale.

In an analysis published in November in the journal Science, they shared new insights on how glaciers in different regions respond to seasonal warming.

Glacier speed is measured by tracking the motion of deep cracks (crevasses) and surface debris in sequences of satellite images collected over time.

These provide unique glacial "fingerprints" that scientists track using an algorithm developed at JPL as part of the ITS_LIVE project.

The team used this technique to map glacier flow at high resolution globally, then analyzed subtle changes in glacier speed to understand how glaciers respond to warming that occurs between winter and summer.

"Earth has over 200,000 glaciers, and we’re watching all of them closely," said Gardner, the study’s coauthor. "It’s no surprise that with this much data, a pattern started to emerge."

In hotter areas of the planet, meanwhile, satellites are being used to detect areas that are dry and prone to wildfire outbreaks, in a development that will aid fire prevention.