Researchers at Stanford University have used machine learning to analyze high-resolution satellite and aerial observations of ice movements in Antarctica, revealing key physical processes governing the large-scale movement of the Antarctic ice sheet. The study, published in Science, found that existing models lack critical complexity necessary for accurately predicting the Antarctic ice sheet’s dynamics and mass loss.
The researchers focused on five of Antarctica’s ice shelves, where they found that the parts closest to the continent are being compressed, but farther away from the continent exhibit anisotropic behavior. This means that the ice sheet movement is not uniform in all directions, but varies with the angular direction.
The study’s findings suggest that current constitutive models used to simulate ice sheet movement are not accurately capturing the complex behavior of the ice sheet. The researchers believe that the techniques used – combining observational data and established physical laws with deep learning – could be used to reveal the physics of other natural processes with extensive observational data.
The discovery is significant because it reveals that ice sheet movement in Antarctica is not as uniform as previously thought, but rather exhibits complex behavior due to factors such as crevasses, air and water filled cavities, and changes in basal friction. This information will be crucial for predicting the future evolution of Antarctica’s ice sheet and the impact on sea levels.
The study’s authors hope that their methods will assist with additional scientific discoveries and lead to new collaborations within the Earth science community. They also believe that the techniques used can help reveal new insights into the physics of other natural processes, such as those in Greenland and other parts of the world.
Source: https://scitechdaily.com/stanford-scientists-just-found-a-missing-piece-in-antarcticas-ice-puzzle