Downscaling Taiwan Precipitation with a Residual Deep Learning Approach

Downscaling Taiwan Precipitation with a Residual Deep Learning Approach | 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 Research Article Downscaling Taiwan Precipitation with a Residual Deep Learning Approach Li-Huan Hsu, Chou-Chun Chiang, Kuan-Ling Lin, Hsin-Hung Lin, Jung-Lien Chu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3849748/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 May, 2024 Read the published version in Geoscience Letters → Version 1 posted 5 You are reading this latest preprint version Abstract In response to the growing demand for high-resolution rainfall data to support disaster prevention in Taiwan, this study presents an innovative approach for downscaling precipitation data. We employed a hierarchical architecture of Multi-Scale Residual Networks (MSRN) to downscale rainfall from a coarse 0.25-degree resolution to a finely 0.0125-degree resolution, representing a substantial challenge due to a resolution increase of over 20 times. Our results demonstrate that the hierarchical MSRN outperforms both the one-step MSRN and linear interpolation methods when reconstructing high-resolution daily rainfall. It surpasses the linear interpolation method by 15.1% and 9.1% in terms of mean absolute error and root mean square error, respectively. Furthermore, the hierarchical MSRN excels in accurately reproducing high-resolution rainfall for various rainfall thresholds, displaying minimal biases. The threat score (TS) highlights the hierarchical MSRN's capability to replicate extreme rainfall events, achieving TS scores exceeding 0.54 and 0.46 at rainfall thresholds of 350 and 500 mm per day, outperforming alternative methods. This method is also applying on an operational global model, the ECMWF’s daily rainfall forecasts over Taiwan. The evaluation results indicate that our approach is effective at improving rainfall forecasts for thresholds greater than 100 mm per day, with more significant improvement for the1-3-day lead forecast. This approach also offers a realistic visual representation of fine-grained rainfall distribution, showing promise for making significant contributions to disaster preparedness and weather forecasting in Taiwan. Downscaling super resolution deep learning residual network precipitation forecasting Full Text Cite Share Download PDF Status: Published Journal Publication published 09 May, 2024 Read the published version in Geoscience Letters → Version 1 posted Editorial decision: Major Revision 15 Feb, 2024 Reviewers agreed at journal 16 Jan, 2024 Reviewers invited by journal 11 Jan, 2024 Editor invited by journal 11 Jan, 2024 First submitted to journal 08 Jan, 2024 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-3849748","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":266666299,"identity":"0babc311-0c94-47c2-b620-becd961af705","order_by":0,"name":"Li-Huan Hsu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYBACPgkeIFkBxAeI1cIG1nIGpiUBLMjYQFALYxtJWqR7D3/mnXdHnu/2AbYHP38w2PVLJLA/5sGnReZcmjTvtmeGM88lsBv2JDAkz5yRwNiMV4tEjhkz77bDCQZnQI4EajE4c4CxOQe/FuPPvHMgWiT/EKnFQJq3AaJFGmiLncHxBoJazCTnHDtsOPMMY5u0TJpEgmR7Y+PsP3i08AMd9uFNzWF5vjPMxyTf2NjY8zMzH/g4A48WJACODYnEBgLRggnsSVM+CkbBKBgFIwEAADvwRiYV8Ey6AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-3979-9142","institution":"National Science and Technology Center for Disaster Reduction","correspondingAuthor":true,"prefix":"","firstName":"Li-Huan","middleName":"","lastName":"Hsu","suffix":""},{"id":266666300,"identity":"af48c949-2363-4d50-8b7b-3b7e6782b0f7","order_by":1,"name":"Chou-Chun Chiang","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction","correspondingAuthor":false,"prefix":"","firstName":"Chou-Chun","middleName":"","lastName":"Chiang","suffix":""},{"id":266666301,"identity":"6f0c5334-aa82-4c8d-9200-dcc02ea61e60","order_by":2,"name":"Kuan-Ling Lin","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction","correspondingAuthor":false,"prefix":"","firstName":"Kuan-Ling","middleName":"","lastName":"Lin","suffix":""},{"id":266666302,"identity":"529b56a4-64a0-4edd-9399-0cc3c260a94b","order_by":3,"name":"Hsin-Hung Lin","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction","correspondingAuthor":false,"prefix":"","firstName":"Hsin-Hung","middleName":"","lastName":"Lin","suffix":""},{"id":266666303,"identity":"77af231f-ea44-4bb9-86dc-f457ff9b6720","order_by":4,"name":"Jung-Lien Chu","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction","correspondingAuthor":false,"prefix":"","firstName":"Jung-Lien","middleName":"","lastName":"Chu","suffix":""},{"id":266666304,"identity":"d33cd455-7170-46df-88a8-c7c2f2e08e30","order_by":5,"name":"Yi-Chiang Yu","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction","correspondingAuthor":false,"prefix":"","firstName":"Yi-Chiang","middleName":"","lastName":"Yu","suffix":""},{"id":266666305,"identity":"ae45b8dc-a546-4b46-bb5b-37c09b0f45e0","order_by":6,"name":"Chin-Shyurng Fahn","email":"","orcid":"","institution":"National Taiwan University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Chin-Shyurng","middleName":"","lastName":"Fahn","suffix":""}],"badges":[],"createdAt":"2024-01-10 08:46:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3849748/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3849748/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s40562-024-00340-y","type":"published","date":"2024-05-09T21:18:27+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":56488631,"identity":"bdcce62a-667f-4d28-ac13-db933d233b4d","added_by":"auto","created_at":"2024-05-14 21:33:04","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1366417,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3849748/v1_covered_dfe2b3df-3d09-4d2f-9e26-8de04194a29b.pdf"}],"financialInterests":"","formattedTitle":"Downscaling Taiwan Precipitation with a Residual Deep Learning Approach","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":true,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"geoscience-letters","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gosl","sideBox":"Learn more about [Geoscience Letters](https://geoscienceletters.springeropen.com/)","snPcode":"40562","submissionUrl":"https://submission.springernature.com/new-submission/40562/3","title":"Geoscience Letters","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Downscaling, super resolution, deep learning, residual network, precipitation forecasting","lastPublishedDoi":"10.21203/rs.3.rs-3849748/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3849748/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn response to the growing demand for high-resolution rainfall data to support disaster prevention in Taiwan, this study presents an innovative approach for downscaling precipitation data. We employed a hierarchical architecture of Multi-Scale Residual Networks (MSRN) to downscale rainfall from a coarse 0.25-degree resolution to a finely 0.0125-degree resolution, representing a substantial challenge due to a resolution increase of over 20 times. Our results demonstrate that the hierarchical MSRN outperforms both the one-step MSRN and linear interpolation methods when reconstructing high-resolution daily rainfall. It surpasses the linear interpolation method by 15.1% and 9.1% in terms of mean absolute error and root mean square error, respectively. Furthermore, the hierarchical MSRN excels in accurately reproducing high-resolution rainfall for various rainfall thresholds, displaying minimal biases. The threat score (TS) highlights the hierarchical MSRN's capability to replicate extreme rainfall events, achieving TS scores exceeding 0.54 and 0.46 at rainfall thresholds of 350 and 500 mm per day, outperforming alternative methods. This method is also applying on an operational global model, the ECMWF’s daily rainfall forecasts over Taiwan. The evaluation results indicate that our approach is effective at improving rainfall forecasts for thresholds greater than 100 mm per day, with more significant improvement for the1-3-day lead forecast. This approach also offers a realistic visual representation of fine-grained rainfall distribution, showing promise for making significant contributions to disaster preparedness and weather forecasting in Taiwan.\u003c/p\u003e","manuscriptTitle":"Downscaling Taiwan Precipitation with a Residual Deep Learning Approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-15 04:34:17","doi":"10.21203/rs.3.rs-3849748/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major Revision","date":"2024-02-15T13:11:09+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2024-01-16T12:36:34+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-01-11T19:33:25+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Geoscience Letters","date":"2024-01-11T08:12:43+00:00","index":"","fulltext":""},{"type":"submitted","content":"Geoscience Letters","date":"2024-01-09T01:11:10+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"geoscience-letters","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gosl","sideBox":"Learn more about [Geoscience Letters](https://geoscienceletters.springeropen.com/)","snPcode":"40562","submissionUrl":"https://submission.springernature.com/new-submission/40562/3","title":"Geoscience Letters","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"aa1e38fb-5e83-4970-9add-18fe58bbcecd","owner":[],"postedDate":"January 15th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-05-14T21:31:50+00:00","versionOfRecord":{"articleIdentity":"rs-3849748","link":"https://doi.org/10.1186/s40562-024-00340-y","journal":{"identity":"geoscience-letters","isVorOnly":false,"title":"Geoscience Letters"},"publishedOn":"2024-05-09 21:18:27","publishedOnDateReadable":"May 9th, 2024"},"versionCreatedAt":"2024-01-15 04:34:17","video":"","vorDoi":"10.1186/s40562-024-00340-y","vorDoiUrl":"https://doi.org/10.1186/s40562-024-00340-y","workflowStages":[]},"version":"v1","identity":"rs-3849748","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3849748","identity":"rs-3849748","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}