AI Assists in Tracking Icebergs by Analyzing Radar Data

Scientists are employing machine learning to analyze satellite radar data for the identification of icebergs in the Southern Ocean surrounding Antarctica. This approach aims to enhance comprehension of their life cycle and environmental repercussions.

While icebergs might appear distant and exotic, as depicted in movies like Titanic, they can significantly impact us unexpectedly. Just last week, the world witnessed the detachment of the largest iceberg, A23a, more than twice the size of Greater London, after being grounded for nearly three decades. It is now drifting north in the Antarctic Ocean. Simultaneously, numerous smaller icebergs are continually breaking away from the Antarctic ice shelf and drifting into the open sea.

The Ecological Impact of Melting Icebergs on Ocean Dynamics and Global Sea Levels

The presence of numerous icebergs extends beyond posing a threat to shipping. As they gradually melt over decades, these icebergs release cold fresh water and nutrients, causing shifts in local ecology, complex ocean circulation dynamics, sea ice breakup, and even global sea levels.

Identifying and tracking these icebergs, which move chaotically in large numbers, has been challenging. To address this, scientists, supported by the Alan Turing Institute, have utilized Synthetic Aperture Radar (SAR) from the European Space Agency’s Sentinel-1 satellites. These satellites can scan icebergs day and night, regardless of weather conditions.

While radar data itself is not new, the innovation lies in employing an unsupervised AI algorithm to analyze readings collected from October 2019 to September 2020. This analysis identified nearly 30,000 icebergs, each measuring approximately 1 km² (0.4 miles²) or less, in the Amundsen Sea Embayment in West Antarctica near the calving front of Thwaites Glacier.

Striving for an Antarctic Digital Twin to Deepen Insights into Ocean, Ice, and Atmosphere Dynamics

The goal is to accurately detect and track icebergs, eventually creating a digital twin of the Antarctic sea. This digital model aims to enhance understanding of the intricate physics governing the interactions between the ocean, ice, and atmosphere.

Ben Evans from the British Antarctic Survey (BAS) AI Lab stated, “The technology we used to develop this tool is already used quite commonly for medical imaging, and so we are excited to apply the same technology to the complex features seen in SAR satellite images of the polar oceans. The method we used is as accurate as the other alternative iceberg-detection methods and outperforms most, without the need for human input. This means it can be easily scaled up beyond our study area and even provide near real-time monitoring.”

Read the original article: New Atlas

Read more: The Model Identifies Glaciers Risking Collapse Because of Climate Change