Ice Sheet
Ice sheets are massive bodies of land ice, and understanding their dynamics is crucial for predicting sea-level rise and its consequences. Current research focuses on developing computationally efficient models, such as graph neural networks (GNNs) and convolutional-transformer networks, to simulate ice sheet behavior and predict future changes under various climate scenarios. These advanced models improve accuracy and speed compared to traditional methods, enabling more comprehensive analyses of ice sheet mass balance, flow, and their impact on global sea levels. Improved data interpretation techniques, including automated annotation methods and advanced image analysis, are also enhancing the accuracy and efficiency of ice sheet research.
Papers
Graph Neural Network as Computationally Efficient Emulator of Ice-sheet and Sea-level System Model (ISSM)
Younghyun Koo, Maryam Rahnemoonfar
Graph Neural Networks for Emulation of Finite-Element Ice Dynamics in Greenland and Antarctic Ice Sheets
Younghyun Koo, Maryam Rahnemoonfar
Assessing Annotation Accuracy in Ice Sheets Using Quantitative Metrics
Bayu Adhi Tama, Vandana Janeja, Sanjay Purushotham