A gradient-boosted tree framework to model the ice thickness of the World's glaciers (IceBoost v1)

Abstract. Knowledge of glacier ice volumes is crucial for constraining future sea level potential, evaluating freshwater resources, and assessing impacts on societies, from regional to global. Motivated by the disparity in existing ice volume estimates, we present IceBoost, a global Machine Learning framework to model individual glacier ice thickness distributions. IceBoost is an ensemble of two gradient-boosted trees trained with 3.7 million globally-available ice thickness measurements and an array of 34 numerical features. The model error is similar to existing models outside polar regions and up to 30–40 % lower at high latitudes. Providing supervision by exposing the model to available glacier thickness measurements reduces the error by up to a factor 2 to 3. A feature ranking analysis reveals that geodetic information are the most informative variables, while ice velocity can improve the model performance by 6 % at high latitudes. A major feature of IceBoost is a capability to generalize outside the training domain, i.e. producing meaningful ice thickness distributions in all regions of the World, including in the ice sheet peripheries.