A Super-Resolution Framework for Downscaling Machine Learning Weather Prediction toward 1-km air temperature

A Super-Resolution Framework for Downscaling Machine Learning Weather Prediction toward 1-km air temperature | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A Super-Resolution Framework for Downscaling Machine Learning Weather Prediction toward 1-km air temperature Hyebin Park, Seonyoung Park, Daehyun Kang, Jeong-Hwan Kim This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7658869/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 26 Jan, 2026 Read the published version in npj Climate and Atmospheric Science → Version 1 posted 10 You are reading this latest preprint version Abstract Artificial intelligence has improved the accuracy and efficiency of weather forecasting, surpassing traditional numerical weather prediction models. However, the coarse spatial resolution of global weather forecasting systems limits their ability to capture fine-scale surface heterogeneity and localized extremes, particularly in regions with complex terrain or urban heat island effects. Here, we introduce SR-Weather, a deep learning-based super-resolution framework that converts coarse 0.25° forecasts into 1-km surface air temperature fields using MODIS-derived temperature targets and high-resolution auxiliary inputs. SR-Weather outperforms existing super-resolution methods by explicitly incorporating spatial context, such as topography, impervious surface fraction, and seasonal climatology maps of air temperature. When SR-Weather was applied to the FuXi global weather forecast, the 7-day forecast error in South Korea decreased by more than 20%, which was comparable to the 1-day forecast error from low-resolution prediction using simple spatial interpolation. In addition, SR-Weather effectively reconstructs missing pixels in MODIS-derived air temperature maps under heavy cloud contamination by leveraging auxiliary variables and climatologically smoothed fields. Although validated over South Korea, the method is region-agnostic and readily generalizable because of the global availability of MODIS inputs and minimal auxiliary data requirements. These results indicate that SR-Weather is a scalable and high-fidelity tool for enhancing machine learning-based weather forecasts at fine spatial scales. Earth and environmental sciences/Climate sciences Earth and environmental sciences/Environmental sciences Full Text Additional Declarations No competing interests reported. Supplementary Files SupplementaryATSuperresolution.docx Cite Share Download PDF Status: Published Journal Publication published 26 Jan, 2026 Read the published version in npj Climate and Atmospheric Science → Version 1 posted Editorial decision: Revision requested 17 Oct, 2025 Reviews received at journal 17 Oct, 2025 Reviews received at journal 09 Oct, 2025 Reviewers agreed at journal 01 Oct, 2025 Reviewers agreed at journal 01 Oct, 2025 Reviewers agreed at journal 01 Oct, 2025 Reviewers invited by journal 26 Sep, 2025 Editor assigned by journal 25 Sep, 2025 Submission checks completed at journal 25 Sep, 2025 First submitted to journal 19 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7658869","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":526150046,"identity":"132f4860-098f-4277-867e-4f8689676735","order_by":0,"name":"Hyebin Park","email":"","orcid":"","institution":"Seoul National University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Hyebin","middleName":"","lastName":"Park","suffix":""},{"id":526150047,"identity":"1c9cb82f-4c24-45f2-b2b6-d456a5d381a1","order_by":1,"name":"Seonyoung Park","email":"","orcid":"","institution":"Seoul National University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Seonyoung","middleName":"","lastName":"Park","suffix":""},{"id":526150048,"identity":"8ea02b94-216c-4071-99ab-7093d3a77fe2","order_by":2,"name":"Daehyun Kang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYDCCMyCiIoEHyk0gVssZoBY2krQwtgFVEq2F78zhZw+/zkuTkZ/fwPbhA0NaPkEtkmfbzI1lt+XwGBxjYJ45gyHHsoGQFoPzDGbSktsqeAzYGJiZeRgqDAjaYnCe/Zu05JwKHvk2oJY/RGk522Mm+bEhh4cB6DBmBoYcwlokz5wpk2Y4lgb0S2IzY49BGmEtfGfSt0n+qEm2l28+fJjhR0UyYS0gwAyJesYGoDuJ0gBU+4NIhaNgFIyCUTBCAQDLHzWCDFRc5gAAAABJRU5ErkJggg==","orcid":"","institution":"Korea Institute of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Daehyun","middleName":"","lastName":"Kang","suffix":""},{"id":526150049,"identity":"ea4474e5-a45b-409a-b948-dba5ed6b8574","order_by":3,"name":"Jeong-Hwan Kim","email":"","orcid":"","institution":"Korea Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Jeong-Hwan","middleName":"","lastName":"Kim","suffix":""}],"badges":[],"createdAt":"2025-09-19 13:08:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7658869/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7658869/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41612-026-01328-5","type":"published","date":"2026-01-26T15:57:44+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93175564,"identity":"cfefc31f-681a-46ac-9ff9-e680684ede42","added_by":"auto","created_at":"2025-10-09 20:54:56","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2455989,"visible":true,"origin":"","legend":"","description":"","filename":"ArticleFileATSuperresolution.docx","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/428e885cde0e9a14882a4f69.docx"},{"id":93174728,"identity":"45b6b1b2-ed61-4758-b3a2-e361e085972d","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6962,"visible":true,"origin":"","legend":"","description":"","filename":"81a9366afb854120ac92e158e79aa293.json","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/9c02776817f5f45d7aa6910f.json"},{"id":93174739,"identity":"30b1150f-5bcc-4a16-98d0-0ac87a860fb6","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":634898,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryATSuperresolution.docx","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/cadd559dd446a22104d532d1.docx"},{"id":93175199,"identity":"ea6a2cc5-8194-4aeb-b9ff-964598dc3084","added_by":"auto","created_at":"2025-10-09 20:46:56","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115433,"visible":true,"origin":"","legend":"","description":"","filename":"81a9366afb854120ac92e158e79aa2931enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/5cf80766d6f85c682d452907.xml"},{"id":93174729,"identity":"f9267c0b-1cfc-460c-a4c8-372b89ab5842","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":362325,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/829f51d4760bd8169c5ab761.png"},{"id":93174732,"identity":"44451bb6-a414-46e2-8033-f8eb5c10bfca","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1005656,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/f1ffec3b4e4c4184e64066b5.png"},{"id":93174731,"identity":"0f95f41d-05d4-41c8-8e46-387719182270","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108736,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/a8daf97fa9c69da0ed45cd33.png"},{"id":93175200,"identity":"f20959ec-c7af-4393-9f1d-6dc1e60da8e1","added_by":"auto","created_at":"2025-10-09 20:46:56","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":182922,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/5514ed42b79b10b610ec8621.png"},{"id":93175201,"identity":"d243944e-5b82-422d-aa2d-54d2fc28dd7c","added_by":"auto","created_at":"2025-10-09 20:46:56","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":563133,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/245e2622742d06c23de5d1af.jpeg"},{"id":93175707,"identity":"c47aecfc-dc38-4b97-8fdb-efdec8fa570a","added_by":"auto","created_at":"2025-10-09 21:02:56","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":99882,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/cc5312ab177dde376ddb0014.png"},{"id":93174734,"identity":"c1968edc-318d-4830-930b-fb67d5c442b8","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79263,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/d356b6b1cdb85df9b36d4755.png"},{"id":93175563,"identity":"5ba833c2-a3cb-4a9d-a58d-3332b6fffc34","added_by":"auto","created_at":"2025-10-09 20:54:56","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":189439,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/008972591afc1f8cedff1255.png"},{"id":93174743,"identity":"537ff342-2b23-4591-a0c6-9bf95c3d6743","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41035,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/72c000457b2ecfbf76b3e822.png"},{"id":93175204,"identity":"80beecba-f699-4f6f-a2ca-aa4c98e3d900","added_by":"auto","created_at":"2025-10-09 20:46:56","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29865,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/f42c1ac75d5efe9908c4b66a.png"},{"id":93175207,"identity":"76d7c7bd-f778-47bb-a177-2d929eeb3e9b","added_by":"auto","created_at":"2025-10-09 20:46:56","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231351,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/70999f389ab0b115562c70f0.png"},{"id":93174736,"identity":"118d673b-5a31-43f4-8d23-079b01548ff8","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34588,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/ee3b5d16b265c55952d88993.png"},{"id":93174746,"identity":"6a75c867-57e4-48a9-8f48-31a22f0f64a4","added_by":"auto","created_at":"2025-10-09 20:38:57","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":113220,"visible":true,"origin":"","legend":"","description":"","filename":"81a9366afb854120ac92e158e79aa2931structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/88041593db5cbd339a8fdfae.xml"},{"id":93174744,"identity":"2797bbf2-2300-42f7-bb40-b161d4ce2804","added_by":"auto","created_at":"2025-10-09 20:38:56","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125102,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/723ffbac36728f836bb38afe.html"},{"id":101690396,"identity":"1fb49954-2f53-4aab-8a68-42d9ac0b33f7","added_by":"auto","created_at":"2026-02-02 16:00:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1426282,"visible":true,"origin":"","legend":"","description":"","filename":"ArticleFileATSuperresolution.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1_covered_ccb851b8-88a0-4553-978b-362f63edd3f9.pdf"},{"id":93175562,"identity":"9f6f96f8-7d26-4d48-9a10-581204357620","added_by":"auto","created_at":"2025-10-09 20:54:56","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":634898,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryATSuperresolution.docx","url":"https://assets-eu.researchsquare.com/files/rs-7658869/v1/58f865a3aa13900f99b6a623.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Super-Resolution Framework for Downscaling Machine Learning Weather Prediction toward 1-km air temperature","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-climate-and-atmospheric-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjclimatsci","sideBox":"Learn more about [npj Climate and Atmospheric Science](http://www.nature.com/npjclimatsci/)","snPcode":"41612","submissionUrl":"https://submission.springernature.com/new-submission/41612/3","title":"npj Climate and Atmospheric Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7658869/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7658869/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eArtificial intelligence has improved the accuracy and efficiency of weather forecasting, surpassing traditional numerical weather prediction models. However, the coarse spatial resolution of global weather forecasting systems limits their ability to capture fine-scale surface heterogeneity and localized extremes, particularly in regions with complex terrain or urban heat island effects. Here, we introduce SR-Weather, a deep learning-based super-resolution framework that converts coarse 0.25\u0026deg; forecasts into 1-km surface air temperature fields using MODIS-derived temperature targets and high-resolution auxiliary inputs. SR-Weather outperforms existing super-resolution methods by explicitly incorporating spatial context, such as topography, impervious surface fraction, and seasonal climatology maps of air temperature. When SR-Weather was applied to the FuXi global weather forecast, the 7-day forecast error in South Korea decreased by more than 20%, which was comparable to the 1-day forecast error from low-resolution prediction using simple spatial interpolation. In addition, SR-Weather effectively reconstructs missing pixels in MODIS-derived air temperature maps under heavy cloud contamination by leveraging auxiliary variables and climatologically smoothed fields. Although validated over South Korea, the method is region-agnostic and readily generalizable because of the global availability of MODIS inputs and minimal auxiliary data requirements. These results indicate that SR-Weather is a scalable and high-fidelity tool for enhancing machine learning-based weather forecasts at fine spatial scales.\u003c/p\u003e","manuscriptTitle":"A Super-Resolution Framework for Downscaling Machine Learning Weather Prediction toward 1-km air temperature","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-09 20:38:51","doi":"10.21203/rs.3.rs-7658869/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-17T15:50:28+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-17T14:05:22+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-10T01:08:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"302202111128546130475967415603053603305","date":"2025-10-01T16:13:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"99170056095842556638985489220145917705","date":"2025-10-01T15:43:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"83387428825756137408326067887984807302","date":"2025-10-01T15:19:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-26T14:14:13+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-25T14:34:52+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-25T14:13:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Climate and Atmospheric Science","date":"2025-09-19T12:54:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-climate-and-atmospheric-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjclimatsci","sideBox":"Learn more about [npj Climate and Atmospheric Science](http://www.nature.com/npjclimatsci/)","snPcode":"41612","submissionUrl":"https://submission.springernature.com/new-submission/41612/3","title":"npj Climate and Atmospheric Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"cf6991ca-732c-4f1e-808b-161f5dde9c62","owner":[],"postedDate":"October 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":55923552,"name":"Earth and environmental sciences/Climate sciences"},{"id":55923553,"name":"Earth and environmental sciences/Environmental sciences"}],"tags":[],"updatedAt":"2026-02-02T15:59:53+00:00","versionOfRecord":{"articleIdentity":"rs-7658869","link":"https://doi.org/10.1038/s41612-026-01328-5","journal":{"identity":"npj-climate-and-atmospheric-science","isVorOnly":false,"title":"npj Climate and Atmospheric Science"},"publishedOn":"2026-01-26 15:57:44","publishedOnDateReadable":"January 26th, 2026"},"versionCreatedAt":"2025-10-09 20:38:51","video":"","vorDoi":"10.1038/s41612-026-01328-5","vorDoiUrl":"https://doi.org/10.1038/s41612-026-01328-5","workflowStages":[]},"version":"v1","identity":"rs-7658869","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7658869","identity":"rs-7658869","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}