Presentation: 2025 ND EPSCoR Annual conference 

October 21, 2025, NDSU Memorial Union, Fargo, North Dakota

Continental-Scale, High-Order, High-Spatial-Resolution Ice Flow Modeling Based on Graphics Processing Units (GPUs)

Quantifying the uncertainties in future Antarctic ice sheet evolution requires running high-fidelity simulation ensembles. Conventional central processing units (CPUs) limit the time; as a result, studies have employed lower-fidelity models, i.e., stress balance approximations lower than the 3-D Blatter-Pattyn higher-order model and spatial resolutions equal to or greater than 1 km near grounding lines, to keep computational resources manageable when running simulation ensembles of continental-scale Antarctica forward in time. These studies partially quantify the uncertainties in future Antarctic ice sheet evolution. Ice flow predictions are the most computationally expensive part of ice sheet simulations in terms of computer memory and execution time. Here, we develop an accelerated numerical graphics processing unit (GPU) implementation for the Ice-sheet and Sea-level System Model’s (ISSM’s) ice flow solver and compare it with a standard CPU implementation. Our metrics include accuracy, performance, and scalability. The overarching goal is to enable running high-fidelity simulation ensembles for uncertainty quantification of future Antarctic ice sheet evolution, which has been previously impossible due to computational costs. These findings are urgently needed as the Antarctic Ice Sheet loses mass at an increasing rate and will significantly benefit process-oriented and sea-level projection studies over the coming decades.