New AI model reveals the volume of the world's glaciers
How much ice is stored in the world's glaciers? And where exactly is it located? A new study led by Ca' Foscari University of Venice, in collaboration with the Institute of Polar Sciences of the National Research Council of Italy (CNR-ISP), provides an updated global map of glacier ice volume.
A groundbreaking study led by Ca' Foscari University of Venice has unveiled a new AI model, IceBoost v2.0, that accurately maps the volume of glaciers worldwide. Trained on over 7 million ice-thickness measurements, the model combines these observations with 26 physical and geometrical variables to reconstruct the ice thickness and volume of each glacier in the Randolph Glacier Inventory, a comprehensive global glacier inventory.
IceBoost v2.0 provides an improved estimate of the world's glaciers containing approximately 150,000 cubic kilometers of ice, which would result in a sea-level rise of 32.3 centimeters if melted completely. This figure aligns with previous estimates but offers a more realistic distribution of glacier ice, matching field observations up to 40% more accurately. For instance, the Geikie Plateau in eastern Greenland, known for its 2-kilometer-thick glacier, was previously underestimated by nearly half.
The new dataset offers valuable insights for glaciologists and climate models, helping researchers predict glacier behavior and their contribution to sea-level rise. It also guides future field campaigns and scientific investments, particularly in regions with limited observations, such as the Himalaya and Karakoram ranges, and major Patagonian ice fields.
These areas are crucial for freshwater management, as glaciers support rivers, ecosystems, agriculture, and local communities, providing water to around 1.9 billion people worldwide.
Machine-learning models like IceBoost v2.0 offer an alternative approach to understanding complex glacier systems, as they learn directly from data without relying on predefined physical descriptions. While the governing physics of glaciers is well understood, many parameters remain unknown or challenging to measure. As mid-latitude glaciers, including those in the Alps, are projected to disappear within the next few decades, the accurate representation of current glacier conditions becomes increasingly important.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.