[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122221-en":3,"doc-seo-122221-105":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},122221,13056703019404,"Miles","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Tuning parameters of a sea ice model using machine learning","A new method is developed to tune sea-ice rheology parameters by combining (1) an evaluation metric that characterizes sea-ice deformation patterns and (2) a machine-learning workflow for parameter tuning. The approach is applied to the brittle Bingham–Maxwell rheology implemented in the next-generation sea-ice model neXtSIM, using RADARSAT Geophysical Processing System observations as reference. The metric encodes deformation statistics, spatial scaling structure functions, linear kinematic feature geometry, and added anisotropy and texture descriptors. An ensemble of perturbed-model runs trains an ML mapping from deformation descriptors to rheology parameters, enabling general tuning for arbitrary models and platforms.","Geosci. Model Dev., 18, 885–904, 2025  \n[https://doi.org/10.5194/gmd-18-885-2025](https://doi.org/10.5194/gmd-18-885-2025)[ ](https://doi.org/10.5194/gmd-18-885-2025)© Author(s) 2025 . This work is distributed under the Creative Commons Attribution 4 .0 License.  \nTuning parameters of a sea ice model using machine learning Anton Korosov, Yue Ying, and Einar Ólason  \nNansen Environmental and Remote Sensing Centre, Jahnebakken 3, 5007, Bergen, Norway Correspondence: Anton Korosov ([anton.korosov@nersc.no](anton.korosov@nersc.no))  \nReceived: 12 August 2024 – Discussion started: 27 August 2024  \nRevised: 9 December 2024 – Accepted: 17 December 2024 – Published: 14 February 2025  \nAbstract. We developed a new method for tuning sea ice rheology parameters, which consists of two components: anew metric for characterising sea ice deformation patterns and a machine learning (ML)-based approach for tuning rheology parameters. We applied the new method to tune the brittle Bingham–Maxwell rheology (BBM) parameterisation, which was implemented and used in the next-generation sea ice model (neXtSIM) . As a reference dataset, we used sea ice drift and deformation observations from the RADARSAT Geophysical Processing System (RGPS) .  \nThe metric characterises a field of sea ice deformation with a vector of values. It includes well-established descriptors such as the mean and standard deviation of deformation, the structure–function of the spatial scaling analysis, and the density and intersection of linear kinematic features (LKFs) . We added more descriptors to the metric that characterises the pattern of ice deformation, including image anisotropy and Haralick texture features. The developed metric can describe ice deformation from any model or satellite platform. In the parameter tuning method, we first run an ensemble of neXtSIM members with perturbed rheology parameters and then train a machine learning model using the simulated data. We provide the descriptors of ice deformation as input to the ML model and rheology parameters as targets.  \nWe apply the trained ML model to the descriptors computed from RGPS observations. The developed ML-based method is generic and can be used to tune the parameters of any model.  \nWe ran experiments with tens of members and found optimal values for four neXtSIM BBM parameters: scaling parameter for compressive strength ( P0 􀀙 5:1 kPa), cohesion at the reference scale (c ref 􀀙 1:2 MPa), internal friction angle tangent (􀀖 􀀙 0:7) and ice–atmosphere drag coefficient (CA 􀀙 0:00228) . A neXtSIM run with the optimal parameterisation produces maps of sea ice deformation visually in-  \ndistinguishable from RGPS observations. These parameters exhibit weak interannual drift related to changes in sea ice thickness and corresponding changes in ice deformation patterns.  \n1 Introduction  \nSea ice dynamics in highly compact ice result from the interaction between surface stress on the ice supplied by wind and ocean currents and the emerging internal stress in the ice. In sea ice models, the internal stress is calculated by a set of equations commonly referred to as rheology. Virtually all large-scale sea ice models used for sea ice forecasting and climate modelling use the so-called viscous–plastic (VP) rheology of Hibler (1979) or more numerically efficient derivatives thereof. Additionally, the elastic–plastic– anisotropic (EAP) approach was introduced by parameterising the anisotropy of the ice stress through interactions of diamond-shaped floes (e.g. Tsamados et al., 2013 ; Wilchinsky and Feltham, 2004) . The free parameters of the VP rheology have been estimated in various traditional sensitivity experiments (e.g. Panteleev et al., 2020, 2023), and their values are generally considered fixed by the community today.  \nA new branch of brittle rheologies has been proposed and extended by Girard et al. (2011), Dansereau et al. (2016) and Ólason et al. (2022), with the latest version, i.e. the brittle Bingham–Maxwell (BBM","cbCaiqZkqpMZnS9M","https://ap.wps.com/l/cbCaiqZkqpMZnS9M","pdf",9971676,1,20,"English","en",105,"# Abstract\n## Method overview\n## Deformation metric and descriptors\n## Machine-learning tuning workflow\n## Experiments and results\n# 1 Introduction\n## Sea-ice rheology background\n## Brittle Bingham–Maxwell rheology context\n## Motivation for parameter tuning","[{\"question\":\"What are the two main components of the proposed tuning method?\",\"answer\":\"The method combines a metric for characterizing sea-ice deformation patterns and a machine-learning approach that tunes rheology parameters based on those descriptors.\"},{\"question\":\"How is the deformation metric constructed for model tuning?\",\"answer\":\"It represents a deformation field using a vector of descriptors, including deformation statistics, spatial scaling structure functions, properties of linear kinematic features, and added anisotropy and Haralick texture features.\"},{\"question\":\"How are optimal BBM rheology parameters obtained in the experiments?\",\"answer\":\"An ensemble of neXtSIM members with perturbed BBM parameters generates training data, then a trained ML model predicts parameters using descriptors computed from RGPS observations.\"}]","Tuning parameters of a sea ice model using machine learning | PDF",1785809470,50,{"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},"tuning-parameters-of-a-sea-ice-model-using-machine-learning","",{"@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/tuning-parameters-of-a-sea-ice-model-using-machine-learning/122221/",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","2026-08-04",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},"What are the two main components of the proposed tuning method?","Question",{"text":75,"@type":76},"The method combines a metric for characterizing sea-ice deformation patterns and a machine-learning approach that tunes rheology parameters based on those descriptors.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the deformation metric constructed for model tuning?",{"text":80,"@type":76},"It represents a deformation field using a vector of descriptors, including deformation statistics, spatial scaling structure functions, properties of linear kinematic features, and added anisotropy and Haralick texture features.",{"name":82,"@type":73,"acceptedAnswer":83},"How are optimal BBM rheology parameters obtained in the experiments?",{"text":84,"@type":76},"An ensemble of neXtSIM members with perturbed BBM parameters generates training data, then a trained ML model predicts parameters using descriptors computed from RGPS observations.","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},[93,97,101,105,110,114,119,122,126,129,133],{"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":29,"slug":113},6,"Technology","technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":21,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":127,"show_sort_weight":21,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":106,"slug":136},19,"General","general"]