{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Zhang X"],"funding":["Anhui Jianzhu University","Nationale Geologiske Undersøgelser for Danmark og Grønland","National Natural Science Foundation of China"],"pagination":["113869"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12661306"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["28(11)"],"pubmed_abstract":["Accurate firn density estimation is essential for assessing glacier mass balance. However, the accuracy of existing firn densification models (FDMs) is limited by an incomplete understanding of firn compaction dynamics, particularly as firn structure alters in warming conditions. This study proposes the firn density prediction transformer (FDTransformer), a deep learning framework that combines firn densification physics to improve density estimation. By employing the transformer network with sequential self-attention mechanisms, the FDTransformer learns a nonlinear mapping between physical firn parameters input and observed firn density, enabling physics-to-density sequence transformation. These physically constrained parameters are estimated by applying physics-based FDMs. Evaluated usin"],"journal":["iScience"],"pubmed_title":["FDTransformer: A firn density prediction framework combining a self-attention transformer network with firn densification physics."],"pmcid":["PMC12661306"],"funding_grant_id":["2021QDZ07","42274028","41704023"],"pubmed_authors":["Liu L","Luo Z","Zhang X","Jiang H"],"additional_accession":[]},"is_claimable":false,"name":"FDTransformer: A firn density prediction framework combining a self-attention transformer network with firn densification physics.","description":"Accurate firn density estimation is essential for assessing glacier mass balance. However, the accuracy of existing firn densification models (FDMs) is limited by an incomplete understanding of firn compaction dynamics, particularly as firn structure alters in warming conditions. This study proposes the firn density prediction transformer (FDTransformer), a deep learning framework that combines firn densification physics to improve density estimation. By employing the transformer network with sequential self-attention mechanisms, the FDTransformer learns a nonlinear mapping between physical firn parameters input and observed firn density, enabling physics-to-density sequence transformation. These physically constrained parameters are estimated by applying physics-based FDMs. Evaluated usin","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Nov","modification":"2026-07-15T11:16:25.566Z","creation":"2026-07-04T03:11:56.079Z"},"accession":"S-EPMC12661306","cross_references":{"pubmed":["41323272"],"doi":["10.1016/j.isci.2025.113869"]}}