New AI Model IceBoost v2.0 Enhances Global Glacier Volume Estimates
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, has introduced IceBoost v2.0, a machine-learning model designed to provide updated global maps of glacier ice volume. Developed by physicist Niccolò Maffezzoli, the model utilizes over 7 million ice-thickness measurements from glaciers worldwide, combined with 26 physical and geometrical variables, to accurately estimate the ice thickness and volume of glaciers listed in the Randolph Glacier Inventory. This model offers a more precise distribution of glacier ice thickness, improving accuracy by up to 40% compared to previous estimates. The study, published in Scientific Data, highlights the model's ability to predict glacier evolution and its contribution to sea-level rise, which is crucial for future climate models.