Optimisation of Machine Learning ARIMA Forecasts for Modelling Future Variations in Earth’s Surface Phenomena Using Sparse Non-Destructive Time Series Data

Optimisation of Machine Learning ARIMA Forecasts for Modelling Future Variations in Earth’s Surface Phenomena Using Sparse Non-Destructive Time Series Data | 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 Optimisation of Machine Learning ARIMA Forecasts for Modelling Future Variations in Earth’s Surface Phenomena Using Sparse Non-Destructive Time Series Data Opeyemi Ajibola-James This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9642824/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Data-poor regions often struggle to meet the multiple predictor requirements for multivariate machine learning (ML) regression models using in situ or radar satellite observations of Earth’s surface phenomena (ESP). Consequently, univariate ML autoregressive integrated moving average (ARIMA) models, consistent with non-destructive testing (NDT) goals, offer a relative advantage. However, the comparative accuracy of traditional ESP forecast variants (“Forecast”, “Lo95”, and “Hi95”) and their optimisation via a derivative “Hybrid” forecast remains under-researched, particularly when using sparse time-series satellite data. This study utilized three sets (36, 48, and 60 monthly epochs) of sparse sea surface salinity (SSS) data from Soil Moisture Active Passive (SMAP) satellite products (2016–2021). Traditional SSS forecast variants yielded RMSE of 0.5810–2.3283 psu and MAPE of 1.3613–6.9260%. Conversely, optimised “Hybrid” forecasts achieved superior RMSE (0.3133–0.7813psu) and MAPE (0.6773–1.3643%). Paired hypothesis tests confirmed significant differences between the best traditional and Hybrid MAPE values ( p-values : 0.004576–0.01343). Results demonstrate that the innovative “Hybrid” approach consistently and significantly outperforms traditional variants across varying data sparsity levels, reducing forecast error by 50.25–80.30%. These results establish a statistically valid, NDT-compliant, and cost-effective framework, providing a reliable early warning decision-support tool for the global monitoring of harmful ESP anomalies. Oceanography Physical Geography Earth’s surface phenomenon sea surface salinity machine learning arima variations modelling time series forecasts optimisation Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted 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-9642824","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":636219731,"identity":"91ed5b1d-093b-4cde-9904-2c76da270cf6","order_by":0,"name":"Opeyemi Ajibola-James","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-3012-7569","institution":"Geo Inheritance Ltd","correspondingAuthor":true,"prefix":"","firstName":"Opeyemi","middleName":"","lastName":"Ajibola-James","suffix":""}],"badges":[],"createdAt":"2026-05-07 13:03:13","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9642824/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9642824/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108807043,"identity":"227ed568-f5d4-4a26-8c25-2bfd9a0cf09d","added_by":"auto","created_at":"2026-05-08 15:29:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":381175,"visible":true,"origin":"","legend":"","description":"","filename":"OptimisationofMLARIMAESPforecastsDLIndabav1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9642824/v1_covered_59384a46-8390-44f4-a59d-6e096aae88bf.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eOptimisation of Machine Learning ARIMA Forecasts for Modelling Future Variations in Earth’s Surface Phenomena Using Sparse Non-Destructive Time Series Data\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Earth’s surface phenomenon, sea surface salinity, machine learning arima, variations modelling, time series forecasts optimisation","lastPublishedDoi":"10.21203/rs.3.rs-9642824/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9642824/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eData-poor regions often struggle to meet the multiple predictor requirements for multivariate machine learning (ML) regression models using in situ or radar satellite observations of Earth\u0026rsquo;s surface phenomena (ESP). Consequently, univariate ML autoregressive integrated moving average (ARIMA) models, consistent with non-destructive testing (NDT) goals, offer a relative advantage. However, the comparative accuracy of traditional ESP forecast variants (\u0026ldquo;Forecast\u0026rdquo;, \u0026ldquo;Lo95\u0026rdquo;, and \u0026ldquo;Hi95\u0026rdquo;) and their optimisation via a derivative \u0026ldquo;Hybrid\u0026rdquo; forecast remains under-researched, particularly when using sparse time-series satellite data. This study utilized three sets (36, 48, and 60 monthly epochs) of sparse sea surface salinity (SSS) data from Soil Moisture Active Passive (SMAP) satellite products (2016\u0026ndash;2021). Traditional SSS forecast variants yielded RMSE of 0.5810\u0026ndash;2.3283 psu and MAPE of 1.3613\u0026ndash;6.9260%. Conversely, optimised \u0026ldquo;Hybrid\u0026rdquo; forecasts achieved superior RMSE (0.3133\u0026ndash;0.7813psu) and MAPE (0.6773\u0026ndash;1.3643%). Paired hypothesis tests confirmed significant differences between the best traditional and Hybrid MAPE values (\u003cem\u003ep-values\u003c/em\u003e: 0.004576\u0026ndash;0.01343). Results demonstrate that the innovative \u0026ldquo;Hybrid\u0026rdquo; approach consistently and significantly outperforms traditional variants across varying data sparsity levels, reducing forecast error by 50.25\u0026ndash;80.30%. These results establish a statistically valid, NDT-compliant, and cost-effective framework, providing a reliable early warning decision-support tool for the global monitoring of harmful ESP anomalies.\u003c/p\u003e","manuscriptTitle":"Optimisation of Machine Learning ARIMA Forecasts for Modelling Future Variations in Earth’s Surface Phenomena Using Sparse Non-Destructive Time Series Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-08 05:04:06","doi":"10.21203/rs.3.rs-9642824/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3d79f149-7cc2-41fd-9d89-85ff589af93b","owner":[],"postedDate":"May 8th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":67716875,"name":"Oceanography"},{"id":67716876,"name":"Physical Geography"}],"tags":[],"updatedAt":"2026-05-08T05:04:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-08 05:04:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9642824","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9642824","identity":"rs-9642824","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}