Application of deep learning to estimate blue and fin whale call density in the southern California Current Ecosystem

Application of deep learning to estimate blue and fin whale call density in the southern California Current Ecosystem | 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 Application of deep learning to estimate blue and fin whale call density in the southern California Current Ecosystem Michaela N. Alksne, Marie A. Roch, Kaitlin E. Frasier, John A. Hildebrand, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9203950/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Blue (Balaenoptera musculus) and fin whales (Balaenoptera physalus) are dominant contributors to low-frequency ocean soundscapes, yet reliably extracting their calls from long-term passive acoustic recordings is methodologically challenging. Here, we train a multi-class deep-learning detector to identify five principal blue and fin whale call types (A, B, D, 20 Hz, and 40 Hz) from low-frequency spectrograms using a Faster R-CNN architecture combined with iterative human review, hard-negative mining, and multi-platform training on California Cooperative Oceanic Fisheries (henceforth, CalCOFI) sonobuoy and High-frequency Acoustic Recording Package recordings from the southern California Current Ecosystem. The detector was evaluated on four independent test datasets spanning multiple years, seasons, and recording platforms and then deployed on CalCOFI sonobuoy recordings collected quarterly over two decades (2004--2024). The final model achieved consistently high mean precision and recall for most call types (e.g., A: 0.71/0.71; B: 0.83/0.59; D: 0.79/0.84; 20 Hz: 0.87/0.74), while 40 Hz calls remained challenging (0.42/0.69), primarily due to confusion with spectrally overlapping humpback whale downsweeps. Detections were post-processed using call-specific characteristics and received-level thresholds and normalized by recording effort and detection area to derive standardized indices of call density (calls/h*1000km2) with uncertainty estimates. Densities were aggregated annually and show call-specific differences between inshore and offshore habitats and interannual variability associated with periods of anomalous oceanographic conditions. Inter-call interval analyses suggested seasonal stability in blue whale song, high variability in blue and fin whale social calls, and seasonal and interannual variability in fin whale song repetition rates. This study is among the first to use deep-learning to estimate baleen whale call density from decades of passive acoustic recordings in a complex soundscape. Biological sciences/Ecology Earth and environmental sciences/Ecology Earth and environmental sciences/Ocean sciences Full Text Additional Declarations No competing interests reported. Supplementary Files Alksneetalsupplemental.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 09 Apr, 2026 Reviewers invited by journal 03 Apr, 2026 Editor assigned by journal 03 Apr, 2026 Submission checks completed at journal 31 Mar, 2026 First submitted to journal 31 Mar, 2026 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-9203950","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":619016748,"identity":"e11cac7f-f5e9-4620-a388-f213a4e8a3c8","order_by":0,"name":"Michaela N. Alksne","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYDACZgbGgw1Amh/E4WFgYGwgqIWdgQGsRbKBaC38UC0GB4jVotvM/ODgjD935IxvJD/+8IbBRnbDAQJazA6zGRzc2PbM2OxGmpnkHIY0YyK0MBgcfNhwOHHbjRw2Zh6Gw4lEaGH/cPDBn8P1m2fkMH/mYfhPjBYeg4Mb2A4nGEjkMEjzMBwgSkvBwZlthw1nnHkG9ItBsvFMglrOH9/4sOfPYXn+dlCIVdjJ9hHSggYMSFM+CkbBKBgFowAHAABz10uw+4iwwgAAAABJRU5ErkJggg==","orcid":"","institution":"University of California San Diego","correspondingAuthor":true,"prefix":"","firstName":"Michaela","middleName":"N.","lastName":"Alksne","suffix":""},{"id":619016749,"identity":"a4c86f6d-f584-44e1-9133-51362b8c2cb8","order_by":1,"name":"Marie A. Roch","email":"","orcid":"","institution":"San Diego State University","correspondingAuthor":false,"prefix":"","firstName":"Marie","middleName":"A.","lastName":"Roch","suffix":""},{"id":619016750,"identity":"68623711-cc2c-4a2d-b962-040e8b1b369a","order_by":2,"name":"Kaitlin E. Frasier","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"Kaitlin","middleName":"E.","lastName":"Frasier","suffix":""},{"id":619016751,"identity":"97261f89-5858-4b91-83eb-ad7d100a5596","order_by":3,"name":"John A. Hildebrand","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"A.","lastName":"Hildebrand","suffix":""},{"id":619016752,"identity":"6b85ac63-709f-4653-bb24-edf43bf72bff","order_by":4,"name":"Shane Andres","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"Shane","middleName":"","lastName":"Andres","suffix":""},{"id":619016753,"identity":"cac14621-cae6-49db-826f-6b6a35e696bf","order_by":5,"name":"Dolapo Adesanya","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"Dolapo","middleName":"","lastName":"Adesanya","suffix":""},{"id":619016754,"identity":"870843e6-1bda-4fb0-aeae-07aeec3405d0","order_by":6,"name":"Lauren M. Baggett","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"Lauren","middleName":"M.","lastName":"Baggett","suffix":""},{"id":619016755,"identity":"ca0d08e7-d421-4c58-98bc-35a3f2344471","order_by":7,"name":"Joshua M. Jones","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"M.","lastName":"Jones","suffix":""},{"id":619016756,"identity":"ce477a6f-d3d8-4e08-827d-ba767b757ec8","order_by":8,"name":"Xiaobai Liu","email":"","orcid":"","institution":"San Diego State University","correspondingAuthor":false,"prefix":"","firstName":"Xiaobai","middleName":"","lastName":"Liu","suffix":""},{"id":619016757,"identity":"250c3532-95ea-4f85-89af-6fd6b184e46b","order_by":9,"name":"Ana Širović","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"","lastName":"Širović","suffix":""},{"id":619016758,"identity":"ae5effd9-bec0-45c3-8b74-0c93ad4eb72a","order_by":10,"name":"Xixin Zhang","email":"","orcid":"","institution":"San Diego State University","correspondingAuthor":false,"prefix":"","firstName":"Xixin","middleName":"","lastName":"Zhang","suffix":""},{"id":619016759,"identity":"db52962e-b964-4b4e-80c7-5eb3f71cec97","order_by":11,"name":"Joshua Zingale","email":"","orcid":"","institution":"San Diego State University","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"","lastName":"Zingale","suffix":""},{"id":619016760,"identity":"1b9d80f5-7bd0-4c99-9da5-5a6591784a31","order_by":12,"name":"Simone Baumann-Pickering","email":"","orcid":"","institution":"University of California San Diego","correspondingAuthor":false,"prefix":"","firstName":"Simone","middleName":"","lastName":"Baumann-Pickering","suffix":""}],"badges":[],"createdAt":"2026-03-23 19:08:35","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9203950/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9203950/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106724356,"identity":"0fd46791-fd80-4fc8-999c-b3e2def2fac3","added_by":"auto","created_at":"2026-04-12 18:27:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":18587329,"visible":true,"origin":"","legend":"","description":"","filename":"Alksneetal2026.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9203950/v1_covered_965c5217-a96c-4d38-894f-299c13eb0142.pdf"},{"id":106478542,"identity":"86c56b8b-463e-4d27-bef1-f1f1c8cf0d4e","added_by":"auto","created_at":"2026-04-09 04:04:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":9666081,"visible":true,"origin":"","legend":"","description":"","filename":"Alksneetalsupplemental.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9203950/v1/da74dea5fc2bc887523c7acf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Application of deep learning to estimate blue and fin whale call density in the southern California Current Ecosystem","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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9203950/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9203950/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Blue (Balaenoptera musculus) and fin whales (Balaenoptera physalus) are dominant contributors to low-frequency ocean soundscapes, yet reliably extracting their calls from long-term passive acoustic recordings is methodologically challenging. Here, we train a multi-class deep-learning detector to identify five principal blue and fin whale call types (A, B, D, 20 Hz, and 40 Hz) from low-frequency spectrograms using a Faster R-CNN architecture combined with iterative human review, hard-negative mining, and multi-platform training on California Cooperative Oceanic Fisheries (henceforth, CalCOFI) sonobuoy and High-frequency Acoustic Recording Package recordings from the southern California Current Ecosystem. The detector was evaluated on four independent test datasets spanning multiple years, seasons, and recording platforms and then deployed on CalCOFI sonobuoy recordings collected quarterly over two decades (2004--2024). The final model achieved consistently high mean precision and recall for most call types (e.g., A: 0.71/0.71; B: 0.83/0.59; D: 0.79/0.84; 20 Hz: 0.87/0.74), while 40 Hz calls remained challenging (0.42/0.69), primarily due to confusion with spectrally overlapping humpback whale downsweeps. Detections were post-processed using call-specific characteristics and received-level thresholds and normalized by recording effort and detection area to derive standardized indices of call density (calls/h*1000km2) with uncertainty estimates. Densities were aggregated annually and show call-specific differences between inshore and offshore habitats and interannual variability associated with periods of anomalous oceanographic conditions. Inter-call interval analyses suggested seasonal stability in blue whale song, high variability in blue and fin whale social calls, and seasonal and interannual variability in fin whale song repetition rates. This study is among the first to use deep-learning to estimate baleen whale call density from decades of passive acoustic recordings in a complex soundscape.","manuscriptTitle":"Application of deep learning to estimate blue and fin whale call density in the southern California Current Ecosystem","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-09 04:04:22","doi":"10.21203/rs.3.rs-9203950/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"176855539329278433538689751167049990136","date":"2026-04-09T12:55:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-03T14:04:19+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-03T06:00:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-31T17:35:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-31T17:32:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"43f4743b-31a3-4fe1-b713-164b625e6ae6","owner":[],"postedDate":"April 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65869156,"name":"Biological sciences/Ecology"},{"id":65869157,"name":"Earth and environmental sciences/Ecology"},{"id":65869158,"name":"Earth and environmental sciences/Ocean sciences"}],"tags":[],"updatedAt":"2026-04-09T04:04:22+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-09 04:04:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9203950","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9203950","identity":"rs-9203950","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}