IceBoost AI Model Refines Global Glacier Volume, Suggests 32 cm Sea‑Level Rise

Researchers at Italy's University Ca' Foscari Venezia, together with the National Research Council's Institute for Polar Sciences (CNR‑ISP), have released IceBoost v2.0, a machine‑learning model trained on more than seven million glacier‑thickness measurements and 26 physical and geomorphological variables. The model reconstructs glacier thickness point‑by‑point for the global glacier inventory, excluding the Antarctic and Greenland ice sheets, and improves volume estimates by up to 40% compared with previous methods.

IceBoost v2.0 estimates that the world's glaciers contain roughly 150 000 km³ of ice. If all this ice were to melt, global mean sea level would rise by about 32.3 cm. The higher‑resolution ice‑volume data will be incorporated into next‑generation glacier simulations and support the IPCC’s assessments of future glacier evolution, with implications for freshwater availability for an estimated 1.9 billion people.