STAT-LSTM: A multivariate spatiotemporal feature aggregation model for SPEI-based drought prediction

STAT-LSTM: A multivariate spatiotemporal feature aggregation model for SPEI-based drought prediction | 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 STAT-LSTM: A multivariate spatiotemporal feature aggregation model for SPEI-based drought prediction Ying Chen, Huanping Wu, Nengfu Xie, Xiaohe Liang, Lihua Jiang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5286493/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 Feb, 2025 Read the published version in Earth Science Informatics → Version 1 posted 9 You are reading this latest preprint version Abstract In recent decades, shifts in the spatiotemporal patterns of precipitation and extreme temperatures have contributed to more frequent droughts. These changes impact not only agricultural production but also food security, ecological sys- tems, and social stability. Advanced techniques such as machine learning and deep learning models outperform traditional models by improving meteorolog- ical drought prediction. Specifically, this study proposes a novel model named the multivariate feature aggregation-based temporal convolutional network for meteorological drought spatiotemporal prediction (STAT-LSTM). The method consists of three parts: a feature aggregation module, which aggregates multi- variate features to extract initial features; a self-attention-temporal convolutional network (SA-TCN), which extracts time series features and uses the self-attention module’s weighting mechanism to automatically capture global dependencies in the sequential data; and a long short-term memory network (LSTM), which cap- tures long-term dependencies. The performance of the STAT-LSTM model was assessed and compared via performance indicators (i.e., MAE, RMSE, and R 2 ). The results indicated that STAT-LSTM provided the most accurate SPEI pre- diction (MAE = 0.474, RMSE = 0.63, and R 2 = 0.613 for SPEI-3; MAE = 0.356, RMSE = 0.468, and R 2 = 0.748 for SPEI-6; MAE = 0.284, RMSE = 0.437, and R 2 = 0.813 for SPEI-9; and MAE = 0.182, RMSE = 0.267, and R2 = 0.934 for SPEI-12). Drought prediction Deep learning Temporal convolutional network Feature aggregation Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 25 Feb, 2025 Read the published version in Earth Science Informatics → Version 1 posted Editorial decision: Revision requested 23 Jan, 2025 Reviews received at journal 20 Jan, 2025 Reviewers agreed at journal 11 Jan, 2025 Reviews received at journal 23 Dec, 2024 Reviewers agreed at journal 01 Dec, 2024 Reviewers invited by journal 29 Oct, 2024 Editor assigned by journal 29 Oct, 2024 Submission checks completed at journal 21 Oct, 2024 First submitted to journal 18 Oct, 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-5286493","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":372206340,"identity":"250aca5a-dc36-4759-81c2-52b41264c2b0","order_by":0,"name":"Ying Chen","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Chen","suffix":""},{"id":372206341,"identity":"696fae1c-0367-45c9-9281-281a8369823a","order_by":1,"name":"Huanping Wu","email":"","orcid":"","institution":"China Meteorological Administration","correspondingAuthor":false,"prefix":"","firstName":"Huanping","middleName":"","lastName":"Wu","suffix":""},{"id":372206343,"identity":"4fb04531-590d-4ace-9fcd-9994bdb8648f","order_by":2,"name":"Nengfu Xie","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYDACZgjFz8DeQKIWyQaeAyRaJtkgkUCkUoPjzMcefNxRK2Fw8/HDDz/+MMibEzS9mS3dcOaZ4xIGt9OMJXvbGAx3NhDQws/MYybN23aszuB2DoMEbwNDgsEBAlrYmPm/Sf9tOwZ02Bnmn3/+EKEFaAubNGNbjYTBDSCDh40ILUC/mAG9cEBC8kyambVsm4ThBkJaDM4ffibxs61Ogu/44cc33/yxkSdoCxQchjEkiFMPBHVEqxwFo2AUjIIRCADSpzyf30V0iwAAAABJRU5ErkJggg==","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":true,"prefix":"","firstName":"Nengfu","middleName":"","lastName":"Xie","suffix":""},{"id":372206345,"identity":"d223ff01-f61b-46b4-8b9d-21b3c234b309","order_by":3,"name":"Xiaohe Liang","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xiaohe","middleName":"","lastName":"Liang","suffix":""},{"id":372206346,"identity":"b3c62ea2-a1bc-4c32-a705-8a850f202c9c","order_by":4,"name":"Lihua Jiang","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"prefix":"","firstName":"Lihua","middleName":"","lastName":"Jiang","suffix":""},{"id":372206348,"identity":"c7228537-08ec-4238-a71a-72ab697ccf65","order_by":5,"name":"Minghui Qiu","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"prefix":"","firstName":"Minghui","middleName":"","lastName":"Qiu","suffix":""},{"id":372206349,"identity":"4e839306-52f6-46d7-b80b-5846122b4f9b","order_by":6,"name":"Yonglei Li","email":"","orcid":"","institution":"Chinese Academy of Agricultural Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yonglei","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2024-10-18 05:08:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5286493/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5286493/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s12145-025-01813-0","type":"published","date":"2025-02-25T15:57:26+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":77622447,"identity":"2ae8e44d-734a-4f25-930e-59afe97c9437","added_by":"auto","created_at":"2025-03-03 16:06:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1758285,"visible":true,"origin":"","legend":"","description":"","filename":"revisedmanuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5286493/v1_covered_fa66b94c-c4ea-4903-8c0e-b704c77cfc28.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"STAT-LSTM: A multivariate spatiotemporal feature aggregation model for SPEI-based drought prediction","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":"earth-science-informatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"esin","sideBox":"Learn more about [Earth Science Informatics](http://link.springer.com/journal/12145)","snPcode":"12145","submissionUrl":"https://submission.nature.com/new-submission/12145/3","title":"Earth Science Informatics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Drought prediction, Deep learning, Temporal convolutional network, Feature aggregation","lastPublishedDoi":"10.21203/rs.3.rs-5286493/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5286493/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn recent decades, shifts in the spatiotemporal patterns of precipitation and extreme temperatures have contributed to more frequent droughts. These changes impact not only agricultural production but also food security, ecological sys- tems, and social stability. Advanced techniques such as machine learning and deep learning models outperform traditional models by improving meteorolog- ical drought prediction. Specifically, this study proposes a novel model named the multivariate feature aggregation-based temporal convolutional network for meteorological drought spatiotemporal prediction (STAT-LSTM). The method consists of three parts: a feature aggregation module, which aggregates multi- variate features to extract initial features; a self-attention-temporal convolutional network (SA-TCN), which extracts time series features and uses the self-attention module’s weighting mechanism to automatically capture global dependencies in the sequential data; and a long short-term memory network (LSTM), which cap- tures long-term dependencies. The performance of the STAT-LSTM model was assessed and compared via performance indicators (i.e., MAE, RMSE, and R\u003csup\u003e2\u003c/sup\u003e ). The results indicated that STAT-LSTM provided the most accurate SPEI pre- diction (MAE = 0.474, RMSE = 0.63, and R\u003csup\u003e2\u003c/sup\u003e = 0.613 for SPEI-3; MAE = 0.356, RMSE = 0.468, and R\u003csup\u003e2\u003c/sup\u003e = 0.748 for SPEI-6; MAE = 0.284, RMSE = 0.437, and R\u003csup\u003e2\u003c/sup\u003e = 0.813 for SPEI-9; and MAE = 0.182, RMSE = 0.267, and R2 = 0.934 for SPEI-12).\u003c/p\u003e","manuscriptTitle":"STAT-LSTM: A multivariate spatiotemporal feature aggregation model for SPEI-based drought prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-31 04:24:09","doi":"10.21203/rs.3.rs-5286493/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-01-24T01:00:15+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-20T11:00:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"95288040518551130229607451571380384777","date":"2025-01-11T18:50:37+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-23T07:38:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"116349358178056141174016075213047449108","date":"2024-12-01T05:53:17+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-10-29T10:57:20+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-10-29T10:56:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-10-21T05:22:43+00:00","index":"","fulltext":""},{"type":"submitted","content":"Earth Science Informatics","date":"2024-10-18T05:00:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"earth-science-informatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"esin","sideBox":"Learn more about [Earth Science Informatics](http://link.springer.com/journal/12145)","snPcode":"12145","submissionUrl":"https://submission.nature.com/new-submission/12145/3","title":"Earth Science Informatics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"fb503604-72fa-4e87-89c9-a242602077ea","owner":[],"postedDate":"October 31st, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-03-03T16:00:53+00:00","versionOfRecord":{"articleIdentity":"rs-5286493","link":"https://doi.org/10.1007/s12145-025-01813-0","journal":{"identity":"earth-science-informatics","isVorOnly":false,"title":"Earth Science Informatics"},"publishedOn":"2025-02-25 15:57:26","publishedOnDateReadable":"February 25th, 2025"},"versionCreatedAt":"2024-10-31 04:24:09","video":"","vorDoi":"10.1007/s12145-025-01813-0","vorDoiUrl":"https://doi.org/10.1007/s12145-025-01813-0","workflowStages":[]},"version":"v1","identity":"rs-5286493","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5286493","identity":"rs-5286493","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}