[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127006-en":3,"doc-seo-127006-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127006,2336474466412,"Ezra","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","UNCERTAINTY-ENABLED MACHINE LEARNING FOR EMULATION OF REGIONAL SEA-LEVEL CHANGE CAUSED BY THE ANTARCTIC ICE SHEET","Sea-level change projections across climate scenarios often require forward simulations of the Earth’s gravitational, rotational, and deformational (GRD) response to ice-mass change, which is computationally expensive. This work builds neural-network emulators for sea-level change at 27 coastal locations driven by Antarctic Ice Sheet mass change through the 21st century. Emulators are trained on data from a static sea-level equation solver and ISMIP6-2100 simulations. Calibrated prediction intervals quantify uncertainty via linear-regression postprocessing of nonlinear model outputs, yielding accuracy competitive with baseline emulators and about 100× training-time speedups versus the numerical solver.","arXiv:2406.17729v1 [[physics. ao-ph](physics. ao-ph)] 21 Jun 2024  \nUNCERTAINTY-ENABLED MACHINE LEARNING FOR EMULATION OF REGIONAL SEA-LEVEL CHANGE CAUSED BY THE ANTARCTIC  \nICE SHEET  \nMyungsoo Yoo∗  \nUniversity of Missouri [mym4v@mail.missouri.edu](mym4v@mail.missouri.edu)  \nSophie Coulson  \nUniversity of New Hampshire [sophie.coulson@unh.edu](sophie.coulson@unh.edu)  \nGiri Gopalan  \nLos Alamos National Laboratory [ggopalan@lanl.gov](ggopalan@lanl.gov)  \nMatthew J. Hoffman  \nLos Alamos National Laboratory [mhoffman@lanl.gov](mhoffman@lanl.gov)  \nHolly Kyeore Han  \nLos Alamos National Laboratory (now at Jet Propulsion Laboratory) [kyeore.han@jpl.nasa.gov](kyeore.han@jpl.nasa.gov)  \nChristopher K. Wikle  \nUniversity of Missouri  \n[wiklec@missouri.edu](wiklec@missouri.edu)  \nTrevor Hillebrand  \nLos Alamos National Laboratory  \n[trhille@lanl.gov](trhille@lanl.gov)  \nABSTRACT  \nProjecting sea-level change in various climate-change scenarios typically involves running forward simulations of the Earth’s gravitational, rotational and deformational (GRD) response to ice mass change, which requires high computational cost and time. Here we build neural-network emulators of sea-level change at 27 coastal locations, due to the GRD effects associated with future Antarctic Ice Sheet mass change over the 21st century. The emulators are based on datasets produced using a numerical solver for the static sea-level equation and published ISMIP6-2100 ice-sheet model simulations referenced in the IPCC AR6 report. We show that the neural-network emulators have an accuracy that is competitive with baseline machine learning emulators. In order to quantify uncertainty, we derive well-calibrated prediction intervals for simulated sea-level change via a linear regression postprocessing technique that uses (nonlinear) machine learning model outputs, a technique that has previously been applied to numerical climate models. We also demonstrate substantial gains  \n∗Author of correspondence  \nin computational efficiency: a feedforward neural-network emulator exhibits on the order of 100 times speedup in comparison to the numerical sea-level equation solver that is used for training.  \nKeywords Regional sea level change · Emulator · Machine learning  \n1 Introduction  \nSea-level change is a major impact of climate change that threatens coastal communities and infrastructure (Fox-Kemperet al., 2021 ; Hauer et al., 2020) . Global mean sea level (GMSL) rose by 0.20 m over the last century, and the recent Intergovernmental Panel on Climate Change report (IPCC AR6; Fox-Kemper et al. (2021)) indicates that it is \"virtually certain\" that GMSL will continue to rise through 2100, regardless of anthropogenic greenhouse gas emissions scenario, due to the likelihood of all major processes contributing to GMSL continuing to operate. Sea-level change varies spatially, and relative sea level (RSL, i.e., ocean depth) can differ by up to a factor of two relative to GMSL across regions of the globe (Fox-Kemper et al., 2021 ; Kopp et al., 2015) . Variations in RSL are caused by steric effects (changes in ocean density and salinity) and dynamic sea level changes, as well as gravitational, rotational, and deformational (GRD) effects (Slangen et al., 2014 ; Kopp et al., 2015 ; Gregory et al., 2019 ; Hamlington et al., 2020 ; Fox-Kemperet al., 2021) . Unlike the other two factors, GRD effects are induced by changes in surface mass loading such as ice sheets, glaciers, and terrestrial water storage (reservoirs, lakes, groundwater) . That is, mass redistribution on the Earth’s surface perturbs the gravitational field and rotation potential of the Earth. In addition, the solid Earth deforms over time in response to the changing surface loads. These complex processes yield spatially and temporally varying sea-level change referred to as \"sea-level fingerprints\" unique to the geometry of ice mass change (Farrell and Clark, 1976 ; Mitrovica et al., 2011) . Future sea-level change and un","cbCaimTyJk1wC3tH","https://ap.wps.com/l/cbCaimTyJk1wC3tH","pdf",5963096,1,31,"English","en",105,"# Abstract\n# Introduction\n## Motivation: spatial variability of sea-level change\n## GRD effects and sea-level fingerprints\n## Antarctic Ice Sheet projections and probabilistic frameworks\n## Need for fast emulators of time-evolving fingerprints","[{\"question\":\"Why are neural-network emulators needed for projecting regional sea-level change?\",\"answer\":\"Forward GRD response simulations to ice-mass change are computationally costly, while probabilistic regional projections require evaluating tens of thousands of samples. Emulators provide fast-to-evaluate alternatives without repeatedly running the expensive solver.\"},{\"question\":\"What data and physical processes underpin the proposed emulators?\",\"answer\":\"The emulators target sea-level change at 27 coastal locations caused by GRD effects from Antarctic Ice Sheet mass change. Training data come from a static sea-level equation solver and ISMIP6-2100 ice-sheet model simulations referenced in IPCC AR6.\"},{\"question\":\"How does the method quantify uncertainty in sea-level projections?\",\"answer\":\"A linear regression postprocessing step uses nonlinear machine-learning model outputs to derive well-calibrated prediction intervals for simulated sea-level change.\"}]","UNCERTAINTY-ENABLED MACHINE LEARNING FOR EMULATION OF REGIONAL SEA-LEVEL CHANGE CAUSED BY THE ANTARCTIC ICE SHEET | PDF",1785936237,78,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"uncertainty-enabled-machine-learning-for-emulation-of-regional-sea-level-change-caused-by-the-antarctic-ice-sheet","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/uncertainty-enabled-machine-learning-for-emulation-of-regional-sea-level-change-caused-by-the-antarctic-ice-sheet/127006/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are neural-network emulators needed for projecting regional sea-level change?","Question",{"text":75,"@type":76},"Forward GRD response simulations to ice-mass change are computationally costly, while probabilistic regional projections require evaluating tens of thousands of samples. Emulators provide fast-to-evaluate alternatives without repeatedly running the expensive solver.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What data and physical processes underpin the proposed emulators?",{"text":80,"@type":76},"The emulators target sea-level change at 27 coastal locations caused by GRD effects from Antarctic Ice Sheet mass change. Training data come from a static sea-level equation solver and ISMIP6-2100 ice-sheet model simulations referenced in IPCC AR6.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the method quantify uncertainty in sea-level projections?",{"text":84,"@type":76},"A linear regression postprocessing step uses nonlinear machine-learning model outputs to derive well-calibrated prediction intervals for simulated sea-level change.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]