Symposium Theme and Topics
Artificial intelligence and machine learning are transforming how we observe, model, and predict the behaviour of the cryosphere. By combining data-driven approaches with physical understanding, AI enables new ways to accelerate forward and inverse models, to fuse diverse remote sensing and in situ observations, and to quantify uncertainty in projections of ice dynamics and sea-level rise. This symposium invites contributions that explore the development, evaluation and application of AI methods across glaciology, from physics-informed machine learning that embeds conservation laws into neural surrogates, to deep learning for automated interpretation of optical, radar and lidar imagery; from Bayesian and ML-based inversions for subglacial mapping to data-driven downscaling of climate and ocean forcings at ice margins. We welcome work on emulators and hybrid models that accelerate probabilistic ensemble projections, AI-assisted parameter tuning and model coupling, and studies that integrate sparse observations with prior physical knowledge to improve resolution and uncertainty characterization. Both methodological advances and demonstrative applications, including operational monitoring, process-level insights, and cross-disciplinary integration, are encouraged. Contributions addressing model evaluation, interpretability, data and code sharing, and the societal implications of AI-enabled cryospheric sciences are particularly welcome.
We seek papers and presentations on any timely topic related to Artificial Intelligence in Glaciology. Key focus areas include (but are not limited to):
– Physics-Informed Machine Learning for Ice Processes
– Integrating physical laws and conservation principles into ML models
– Emulation and acceleration of forward and inverse models using neural networks
– ML-based emulators for probabilistic sea level rise projections
– Data-driven discoveries broadly defined
– Remote Sensing and Image Analysis for Glacier and ice sheet Monitoring
– ML for interpreting optical, radar, and lidar imagery
– Automated rift, supraglacial lakes, calving front, grounding lines and iceberg detection
– Time-series analysis for monitoring changes in snow and ice cover
– AI for Subglacial and Bedrock Mapping
– Bayesian and ML-based inversion of gravity, radar, and seismic data
– Enhancing resolution and uncertainty characterization of subglacial topography and internal layer annotation
– Integration of sparse observations with prior knowledge using ML
– ML-aided climate and ocean forcing of ice sheets
– Data-driven downscaling of climate and ocean reanalyses to ice sheet margins
– Learning spatiotemporal patterns of melting and ice-ocean interaction
– AI for coupling ice sheet models with climate models or ocean models
– Other use of AI in glaciology
– AI-based visualization for science communication
– Custom LLMs to support glaciologists.
