Stanford researchers have combined machine learning with high-resolution satellite and airplane observations to understand the physics behind large-scale ice movements in Antarctica. The results show that current models are missing key complexity needed to accurately predict the dynamics and mass loss of the Antarctic ice sheet now and in the future.

Stanford researchers have uncovered a critical missing piece in the complex puzzle of Antarctica’s ice movement, using machine learning and high-resolution satellite data. Their findings, published in Science,

The British Antarctic Survey (BAS) has released the most detailed map yet of what Antarctica looks like when you strip away its ubiquitous cover of ice and snow. Derived from 60 years of data, it will help scientists understand ice flows better.