Extracting Antarctic ice shelf fracture depths using the linear cloth simulation filtering algorithm

• Developed LCSF-ice to automatically extract ice-shelf fracture depth and width from ICESat-2 photon-counting data. • Demonstrated high accuracy across Antarctic ice shelves with diverse and complex fracture morphologies. • Provided a new framework for three-dimensional fracture characterization and structural integrity assessment of ice shelves. Antarctic ice shelves play a critical role in regulating the mass balance of the ice sheet and influencing global sea level change. The formation and evolution of fractures serve as key precursors to ice shelf calving. While remote sensing technologies have enabled effective identification of fracture features in the horizontal dimension, the automated extraction of their depth remains a significant challenge. This study proposes a novel fracture depth extraction algorithm for ICESat-2 photon-counting data: the Linear Cloth Simulation Filtering for Ice Surface (LCSF-ice). The algorithm is inspired by the physical behavior of a cloth naturally draping under gravity, simulating its descent onto the ice surface and suspension along fracture edges. Through this process, LCSF-ice precisely delineates the physical boundary between the ice surface and fracture voids, enabling automatic retrieval of both fracture depth and width. To evaluate the algorithm’s performance under varying fracture complexities, we apply LCSF-ice to the Amery and Thwaites ice shelves. Results demonstrate the algorithm’s strong adaptability, even in scenarios involving intricate fracture morphologies. These findings highlight the potential of ICESat-2 photon data for detailed three-dimensional fracture characterization and provide a new framework for monitoring ice shelf structural integrity.