{"corpus_id":227124856,"paper_sha":"c40714cbb62a4daeb3653e1d2bcff90c362bbbec","doi":"10.1038/s41467-020-20779-9","arxiv_id":null,"pmid":33547299,"pmcid":"7865057","mag_id":null,"dblp_id":null,"acl_id":null,"title":"Deep learning to infer eddy heat fluxes from sea surface height patterns of mesoscale turbulence","year":2019,"publication_date":"2019-11-11","venue":"Nature Communications","journal":{"name":"Nature Communications","pages":null,"volume":"12"},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle"],"pubmed_pub_types":["Journal Article"],"s2_fields_of_study":["Medicine","Physics","Computer Science","Environmental Science"],"reference_count":81,"citation_count":55,"influential_citation_count":2,"is_open_access":true,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":5,"s2_open_access_pdf_url":"https://www.nature.com/articles/s41467-020-20779-9.pdf","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/c40714cbb62a4daeb3653e1d2bcff90c362bbbec","s2_open_access_license":"CCBY","s2_open_access_status":"GREEN","pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Mesoscale eddies have strong signatures in sea surface height (SSH) anomalies that are measured globally through satellite altimetry. However, monitoring the transport of heat associated with these eddies and its impact on the global ocean circulation remains difficult as it requires simultaneous observations of upper-ocean velocity fields and interior temperature and density properties. Here we demonstrate that for quasigeostrophic baroclinic turbulence the eddy patterns in SSH snapshots alone contain sufficient information to estimate the eddy heat fluxes. We use simulations of baroclinic turbulence for the supervised learning of a deep Convolutional Neural Network (CNN) to predict up to 64% of eddy heat flux variance. CNNs also significantly outperform other conventional data-driven techniques. Our results suggest that deep CNNs could provide an effective pathway towards an operational monitoring of eddy heat fluxes using satellite altimetry and other remote sensing products. Eddy heat fluxes crucially affect large-scale oceanic currents but are challenging to monitor on a global scale. Here the authors develop a Deep Learning model to predict the eddy heat fluxes from sea surface height data only, bypassing the need for simultaneous observations of the deep ocean.","claims":[{"public_id":"cl_185e6bf8698158fc57ad13dd4da51cbb","status":"active","text":"A supervised deep convolutional neural network predicts up to 64% of eddy heat flux variance from simulated baroclinic turbulence data.","confidence":0.98,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_185e6bf8698158fc57ad13dd4da51cbb"},{"public_id":"cl_0142ec7eccc2b9037f0c733080065899","status":"active","text":"Deep convolutional neural networks could enable operational monitoring of eddy heat fluxes using satellite altimetry and other remote sensing products.","confidence":0.89,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_0142ec7eccc2b9037f0c733080065899"},{"public_id":"cl_20264c29b1a7ad087a98e43ab57603d9","status":"active","text":"Eddy patterns in sea surface height snapshots alone contain sufficient information to estimate eddy heat fluxes in quasigeostrophic baroclinic turbulence.","confidence":0.96,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_20264c29b1a7ad087a98e43ab57603d9"},{"public_id":"cl_720f32117cbae2a684a1cf0627131291","status":"active","text":"The deep convolutional neural network significantly outperforms other conventional data-driven techniques.","confidence":0.93,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_720f32117cbae2a684a1cf0627131291"}],"concepts":[{"public_id":"co_1732836e1c448f427e3e2ba9ef853be4","status":"active","name":"simulations of baroclinic turbulence","description":"Numerical simulations of baroclinic turbulence used to generate training and evaluation data.","types":["dataset"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_1732836e1c448f427e3e2ba9ef853be4"},{"public_id":"co_6a42a5b7373b74cf0a3d6cd8e7d838b0","status":"active","name":"deep convolutional neural network","description":"A supervised deep learning model based on convolutional neural network architecture.","types":["method"],"aliases":["deep CNN","CNN"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_6a42a5b7373b74cf0a3d6cd8e7d838b0"},{"public_id":"co_8230a0a25a492da3868af0cec109e7f3","status":"active","name":"sea surface height patterns","description":"Spatial variations in ocean surface height used as input signals in this work.","types":["observation"],"aliases":["SSH patterns"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8230a0a25a492da3868af0cec109e7f3"},{"public_id":"co_8ee0dd02cc8c7e9e274998b9f27bf871","status":"active","name":"mesoscale turbulence","description":"Ocean turbulence at mesoscale spatial scales associated with eddy activity.","types":["phenomenon"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8ee0dd02cc8c7e9e274998b9f27bf871"},{"public_id":"co_a19f9f4cc6bcaece71d56ead7cd01621","status":"active","name":"remote sensing products","description":"Observational data products derived from remote sensing instruments.","types":["data source"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_a19f9f4cc6bcaece71d56ead7cd01621"},{"public_id":"co_b6e7cdca72a26856d1054d62f1183e6d","status":"active","name":"sea surface height snapshots","description":"Instantaneous maps or images of sea surface height anomalies.","types":["observation"],"aliases":["SSH snapshots"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_b6e7cdca72a26856d1054d62f1183e6d"},{"public_id":"co_c7d58f2a2ff93b6b2167c464c4dda804","status":"active","name":"quasigeostrophic baroclinic turbulence","description":"A fluid-dynamical turbulence regime modeled with quasigeostrophic and baroclinic dynamics.","types":["physical regime"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_c7d58f2a2ff93b6b2167c464c4dda804"},{"public_id":"co_d14d58c6da3dda260723b87841481518","status":"active","name":"eddy heat flux variance","description":"The fraction of variability in eddy heat fluxes explained by the model.","types":["evaluation metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_d14d58c6da3dda260723b87841481518"},{"public_id":"co_d8a20ce6ad64e6ac632c79017561cd86","status":"active","name":"satellite altimetry","description":"Satellite-based measurement of sea surface height used as a remote sensing source.","types":["measurement technique"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_d8a20ce6ad64e6ac632c79017561cd86"},{"public_id":"co_e6b5bac5d980aa67d9d46aefd3a13e18","status":"active","name":"conventional data-driven techniques","description":"Non-deep-learning predictive methods used as comparison baselines.","types":["baseline"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_e6b5bac5d980aa67d9d46aefd3a13e18"},{"public_id":"co_f18527c7037a5413f82eedd520eaaf03","status":"active","name":"eddy heat fluxes","description":"The transport of heat by ocean eddies.","types":["oceanographic quantity"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_f18527c7037a5413f82eedd520eaaf03"}],"external_ids":{"DOI":"10.1038/s41467-020-20779-9","ArXiv":null,"PubMed":33547299,"PubMedCentral":"7865057","MAG":null,"DBLP":null,"ACL":null},"open_access":{"is_open_access":true,"pdf_url":"https://www.nature.com/articles/s41467-020-20779-9.pdf","landing_url":"https://www.semanticscholar.org/paper/c40714cbb62a4daeb3653e1d2bcff90c362bbbec","source":"semantic_scholar","pdf_url_source":"semantic_scholar_open_access_pdf","license":"CCBY","status":"GREEN","reason":null},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":false,"full_text_source":null,"count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":5},"paper_id":637102,"paper_uid":"0441a7e5-8233-46b5-b7d2-717eb66a6a37","canonical_identity":{"paper_id":637102,"paper_uid":"0441a7e5-8233-46b5-b7d2-717eb66a6a37","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/227124856"}