[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126418-en":3,"doc-seo-126418-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126418,962085564807,"Aurelia","https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8",8,"Research & Report","Estimating the variability of deep-ocean particle flux collected by sediment traps using satellite data and machine learning","The gravitational pump underpins ocean carbon sequestration by exporting sinking organic matter from the surface to the deep ocean. Deep sediment trap records capture sequestered particle fluxes, yet short-term spatio-temporal variability disrupts direct attribution to surface sources. This study proposes U-NetSST-SSH, a machine-learning method that predicts sediment-trap particle catchment area at 3000 m depth using only surface satellite data. Trained on realistic CROCO Lagrangian simulations, it improves catchment prediction over a conventional fixed box. Applied to 20 years of satellite observations, it yields a stronger link between PAP deep particle fluxes and surface chlorophyll-a, while remaining sensitive to unobserved local deep dynamics, supporting improved understanding of the biological carbon pump.","Biogeosciences, 22, 4309–4331, 2025  \n[https://doi.org/10.5194/bg-22-4309-2025](https://doi.org/10.5194/bg-22-4309-2025)[ ](https://doi.org/10.5194/bg-22-4309-2025)© Author(s) 2025 . This work is distributed under the Creative Commons Attribution 4 .0 License.  \nEstimating the variability of deep-ocean particle flux collected by sediment traps using satellite data and machine learning  \nThéo Picard 1 , Chelsey A. Baker2 , Jonathan Gula3,4 , Ronan Fablet5 , Laurent Mémery 1 , and Richard Lampitt2  \n1Univ Brest, CNRS, IRD, Ifremer, Laboratoire des Sciences de l'Environnement Marin (LEMAR), IUEM, Plouzané, France  \n2National Oceanography Centre, Southampton, UK  \n3Univ Brest, CNRS, IRD, Ifremer, Laboratoire d'Océanographie Physique et Spatiale (LOPS), IUEM, Plouzané, France  \n4Institut Universitaire de France (IUF), Paris, France  \n5IMT Atlantique, Lab-STICC, Plouzané, France Correspondence: Théo Picard ([theo.picard@univ-brest.fr](theo.picard@univ-brest.fr))  \nReceived: 22 October 2024 – Discussion started: 5 December 2024  \nRevised: 16 May 2025 – Accepted: 10 June 2025 – Published: 1 September 2025  \nAbstract. The gravitational pump plays a key role in the ocean carbon cycle by exporting sinking organic carbon from the surface to the deep ocean. Deep sediment trap time series provide unique measurements of this sequestered carbon flux. Sinking particles are influenced by physical shortterm spatio-temporal variability, which inhibits the establishment of a direct link to their surface origin. In this study, we present a novel machine learning tool, designated as U-NetSST􀀀SSH, which is capable of predicting the catchment area of particles captured by sediment traps moored at a depth of 3000 m above the Porcupine Abyssal Plain (PAP) based solely on surface data. The machine learning tool was trained and evaluated using Lagrangian experiments in a realistic CROCO numerical simulation. The conventional approach of assuming a static 100–200 km box over the sediment trap location only yields an average prediction for 􀀘 25 % of the source region, whilst U-NetSST􀀀SSH predicts 􀀘 50 % . U-NetSST􀀀SSH was then applied to satellite observations to create a 20-year catchment area dataset, which demonstrates a stronger correlation between the PAP site deep particle fluxes and surface chlorophyll-a concentration compared with the conventional approach. However, predictions remain highly sensitive to the local deep dynamics which are not observed in surface ocean dynamics. The improved identification of the particle source region for deepocean sediment traps can facilitate a more comprehensive understanding of the mechanisms driving the export of particles from the surface to the deep ocean, a key component of the biological carbon pump.  \n1 Introduction  \nThe biological carbon pump (BCP) is one mechanism that sequesters carbon from the atmosphere into the deep ocean. The BCP plays a key role in the climate system as, without it, the atmospheric CO 2 concentrations would be about twice those observed today (Parekh et al., 2006 ; Kwon et al., 2009) . Furthermore, the BCP is a crucial source of food resources in the deep ocean (Grabowski et al., 2019) . However, despite the considerable importance of the BCP, its driving mechanisms are poorly understood (Le Moigne, 2019) . Given that climate-change-driven perturbations may have widescale implications for the BCP, it is of utmost importance to improve our understanding of this topic (Kwon et al., 2009 ; Passowand Carlson, 2012 ; Palevsky and Nicholson, 2018 ; Henson et al., 2022 ; Wilson et al., 2022) .  \nOne of the main processes contributing to the export of the BCP is the export of organic particles from the surface to the deep ocean, which sink due to their excess density (Siegelet al., 2016 ; Durkin et al., 2016 ; Le Moigne, 2019) . This is known as the gravitational pump (Boyd et al., 2019 ; Siegelet al., 2023) . This is a complex process modulated, on the one hand, by phytoplankton net primary pr","cbCaimXCAarPTM0S","https://ap.wps.com/l/cbCaimXCAarPTM0S","pdf",9941602,6,1,23,"English","en",105,"# Abstract\n# Introduction\n## Biological carbon pump and gravitational pump\n## Sediment trap observations and the catchment area problem\n## Motivation and study objective","[{\"question\":\"What problem does the study address regarding sediment trap measurements?\",\"answer\":\"Sediment traps sample a fixed location, while deep particle fluxes vary due to local dynamics, making it difficult to link measured fluxes to surface sources.\"},{\"question\":\"How does U-NetSST-SSH differ from the conventional approach?\",\"answer\":\"U-NetSST-SSH predicts the particle catchment area using surface satellite data, whereas the conventional method assumes a static 100–200 km box and provides more limited average source identification.\"},{\"question\":\"What data and conditions are used to evaluate and apply the model?\",\"answer\":\"The tool is trained and tested using Lagrangian experiments in a CROCO numerical simulation, then applied to 20 years of satellite observations for the Porcupine Abyssal Plain site.\"}]","Estimating the variability of deep-ocean particle flux collected by sediment traps using satellite data and machine learning | PDF",1785904948,58,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"estimating-the-variability-of-deep-ocean-particle-flux-collected-by-sediment-traps-using-satellite-data-and-machine-learning","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/estimating-the-variability-of-deep-ocean-particle-flux-collected-by-sediment-traps-using-satellite-data-and-machine-learning/126418/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What problem does the study address regarding sediment trap measurements?","Question",{"text":77,"@type":78},"Sediment traps sample a fixed location, while deep particle fluxes vary due to local dynamics, making it difficult to link measured fluxes to surface sources.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How does U-NetSST-SSH differ from the conventional approach?",{"text":82,"@type":78},"U-NetSST-SSH predicts the particle catchment area using surface satellite data, whereas the conventional method assumes a static 100–200 km box and provides more limited average source identification.",{"name":84,"@type":75,"acceptedAnswer":85},"What data and conditions are used to evaluate and apply the model?",{"text":86,"@type":78},"The tool is trained and tested using Lagrangian experiments in a CROCO numerical simulation, then applied to 20 years of satellite observations for the Porcupine Abyssal Plain site.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":108,"slug":139},19,"General","general"]