MVPinn: Integrating Milne-Eddington Inversion with Physics-Informed Neural Networks for GST/NIRIS Observations

MVPinn: Integrating Milne-Eddington Inversion with Physics-Informed Neural Networks for GST/NIRIS Observations | 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 MVPinn: Integrating Milne-Eddington Inversion with Physics-Informed Neural Networks for GST/NIRIS Observations Qin Li, Bo Shen, Haodi Jiang, Vasyl Yurchyshyn, Taylor Baildon, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7475340/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 7 You are reading this latest preprint version Abstract We introduce MVPinn, a Physics-Informed Neural Network (PINN) approach tailored for solving the Milne–Eddington (ME) inversion problem, specifically applied to spectropolarimetric observations from the Big Bear Solar Observatory's Near-InfraRed Imaging Spectropolarimeter (BBSO/NIRIS) at the Fe I 1.56 µm lines. Traditional ME inversion methods, though widely used, are computationally intensive, sensitive to noise, and often struggle to accurately capture complex profile asymmetries resulting from gradients in magnetic field strength, orientation, and line-of-sight velocities. By embedding the ME radiative transfer equations directly into the neural network training as physics-informed constraints, our MVPinn method robustly and efficiently retrieves magnetic field parameters, significantly outperforming traditional inversion methods in accuracy, noise resilience, and the ability to handle asymmetric and weak polarization signals. After training, MVPinn infers one magnetogram in about 15 seconds, compared to generally a few tens of minutes required by ME inversion on high-resolution spectropolarimetric data. Quantitative comparisons demonstrate excellent agreement with well-established magnetic field measurements from the SDO/HMI and Hinode/SOT-SP instruments, with correlation coefficients of approximately 90%. In particular, MVPINN aligns better with Hinode/SOT-SP data, indicating some saturation of HMI data at high magnetic strengths. We further analyze the physical significance of profile asymmetries and the limitations inherent in the ME model assumption. Our results illustrate the potential of physics-informed machine learning methods in high-spatial-temporal solar observations, preparing for more sophisticated, near-real-time magnetic field analysis essential for current and next-generation solar telescopes. Magnetogram Solar photosphere Spectropolarimetry Radiative transfer equation High resolution spectroscopy Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 17 Oct, 2025 Reviews received at journal 14 Oct, 2025 Reviewers agreed at journal 08 Sep, 2025 Reviewers invited by journal 08 Sep, 2025 Editor assigned by journal 01 Sep, 2025 Submission checks completed at journal 01 Sep, 2025 First submitted to journal 27 Aug, 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-7475340","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":512226589,"identity":"22ef4e46-be2c-4a22-8f0e-75c75de2bf78","order_by":0,"name":"Qin Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAArklEQVRIiWNgGAWjYDACZiB+YGDDwNgM4rERqyXBIA2ohZlYLSCQwHAYqpkYLXzHeQ8wJBScz2Nu5z/A8KHsMGEtkof5EoAOu10MchjjjHNEaDE4zGMA0pLYCNTCzNtGvJZzEC1/SdByAKKFkRgtkkAtBxIMkkFaDA72nEsnrIXv/BnDBx/+2CVu7D/48MGPMmvCWhgOgBEDg2EDlEGUFjCQJ075KBgFo2AUjEQAAFAHN8gP/BX4AAAAAElFTkSuQmCC","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Qin","middleName":"","lastName":"Li","suffix":""},{"id":512226590,"identity":"1b1013ce-52cb-48ea-8477-8b3ecd995a72","order_by":1,"name":"Bo Shen","email":"","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Shen","suffix":""},{"id":512226591,"identity":"0ba6dd5f-d662-4ca8-bcdc-2c1d1452c0d8","order_by":2,"name":"Haodi Jiang","email":"","orcid":"","institution":"Sam Houston State University","correspondingAuthor":false,"prefix":"","firstName":"Haodi","middleName":"","lastName":"Jiang","suffix":""},{"id":512226592,"identity":"3a518af9-2c49-4e39-8b75-a5a174d18b1b","order_by":3,"name":"Vasyl Yurchyshyn","email":"","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Vasyl","middleName":"","lastName":"Yurchyshyn","suffix":""},{"id":512226593,"identity":"74219485-81dd-40f7-abf9-05c8a994cf56","order_by":4,"name":"Taylor Baildon","email":"","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Taylor","middleName":"","lastName":"Baildon","suffix":""},{"id":512226594,"identity":"b7d116c3-018c-47ea-af46-744077dbf83e","order_by":5,"name":"Kangwoo Yi","email":"","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Kangwoo","middleName":"","lastName":"Yi","suffix":""},{"id":512226595,"identity":"1b1bacc5-c1ea-4622-8de0-236ac9eb2460","order_by":6,"name":"Wenda Cao","email":"","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Wenda","middleName":"","lastName":"Cao","suffix":""},{"id":512226596,"identity":"2267e7db-cfee-450c-b3a7-a012ab1843da","order_by":7,"name":"Haimin Wang","email":"","orcid":"","institution":"New Jersey Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Haimin","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-08-28 01:53:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7475340/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7475340/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91318505,"identity":"09c3dc3a-b46b-4666-b77d-983e7d0c7107","added_by":"auto","created_at":"2025-09-15 08:40:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8054051,"visible":true,"origin":"","legend":"","description":"","filename":"MVPinn0.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7475340/v1_covered_c0742dc3-3c93-494c-8d7c-1db568db47e7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MVPinn: Integrating Milne-Eddington Inversion with Physics-Informed Neural Networks for GST/NIRIS Observations","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"solar-physics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sola","sideBox":"Learn more about [Solar Physics](http://link.springer.com/journal/11207)","snPcode":"11207","submissionUrl":"https://submission.nature.com/new-submission/11207/3","title":"Solar Physics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Magnetogram, Solar photosphere, Spectropolarimetry, Radiative transfer equation, High resolution spectroscopy","lastPublishedDoi":"10.21203/rs.3.rs-7475340/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7475340/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"We introduce MVPinn, a Physics-Informed Neural Network (PINN) approach tailored for solving the Milne–Eddington (ME) inversion problem, specifically applied to spectropolarimetric observations from the Big Bear Solar Observatory's Near-InfraRed Imaging Spectropolarimeter (BBSO/NIRIS) at the Fe I 1.56 µm lines. Traditional ME inversion methods, though widely used, are computationally intensive, sensitive to noise, and often struggle to accurately capture complex profile asymmetries resulting from gradients in magnetic field strength, orientation, and line-of-sight velocities. By embedding the ME radiative transfer equations directly into the neural network training as physics-informed constraints, our MVPinn method robustly and efficiently retrieves magnetic field parameters, significantly outperforming traditional inversion methods in accuracy, noise resilience, and the ability to handle asymmetric and weak polarization signals. After training, MVPinn infers one magnetogram in about 15 seconds, compared to generally a few tens of minutes required by ME inversion on high-resolution spectropolarimetric data. Quantitative comparisons demonstrate excellent agreement with well-established magnetic field measurements from the SDO/HMI and Hinode/SOT-SP instruments, with correlation coefficients of approximately 90\\%. In particular, MVPINN aligns better with Hinode/SOT-SP data, indicating some saturation of HMI data at high magnetic strengths. We further analyze the physical significance of profile asymmetries and the limitations inherent in the ME model assumption. Our results illustrate the potential of physics-informed machine learning methods in high-spatial-temporal solar observations, preparing for more sophisticated, near-real-time magnetic field analysis essential for current and next-generation solar telescopes.","manuscriptTitle":"MVPinn: Integrating Milne-Eddington Inversion with Physics-Informed Neural Networks for GST/NIRIS Observations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-15 08:31:58","doi":"10.21203/rs.3.rs-7475340/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-17T09:59:42+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-14T17:31:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"183027893694502431531508782874530643163","date":"2025-09-09T03:58:33+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-08T14:56:46+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-01T06:12:06+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-01T06:09:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Solar Physics","date":"2025-08-28T01:42:47+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"solar-physics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sola","sideBox":"Learn more about [Solar Physics](http://link.springer.com/journal/11207)","snPcode":"11207","submissionUrl":"https://submission.nature.com/new-submission/11207/3","title":"Solar Physics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"4efb07ab-cf1d-4047-8179-985061d62aaa","owner":[],"postedDate":"September 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-04-07T11:25:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-15 08:31:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7475340","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7475340","identity":"rs-7475340","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}