[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-127972-105":59,"doc-detail-127972-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","monitoring-supraglacial-lake-dynamics-in-antarctica-with-machine-learning","Monitoring Supraglacial Lake Dynamics in Antarctica with Machine Learning","","The work addresses the need for continuous monitoring of supraglacial lake dynamics in Antarctica to better quantify impacts on ice shelf stability under climate change. It links meltwater accumulation to mechanisms that may accelerate ice discharge, including surface runoff, basal meltwater injection, and hydrofracture. Using Sentinel-1 and Sentinel-2 processing, neural-network lake detection is trained separately for each sensor and produces bi-weekly maximum lake extent layers and anomaly/recurrence products since 2014/15.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/monitoring-supraglacial-lake-dynamics-in-antarctica-with-machine-learning/127972/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/monitoring-supraglacial-lake-dynamics-in-antarctica-with-machine-learning/127972.png","ImageObject",300,407,{"name":92,"@type":93},"Ezra","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-19","2026-08-05",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",13,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"Why is monitoring supraglacial lake dynamics important in Antarctica?","Question",{"text":113,"@type":114},"It helps explain how meltwater accumulation can affect ice shelf stability and potentially drive accelerated ice discharge through runoff, basal sliding, or hydrofracture.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How are Sentinel-1 and Sentinel-2 data prepared for the machine learning workflow?",{"text":118,"@type":114},"Sentinel-1 IW data undergo calibration, speckle filtering, and orthorectification, while Sentinel-2 L1C is processed to L2A using the ESA SNAP toolbox on an HPC cluster.",{"name":120,"@type":111,"acceptedAnswer":121},"What outputs are produced from the lake detections over time?",{"text":122,"@type":114},"The method merges lake predictions into bi-weekly maximum lake extent layers, then automatically generates statistics, anomalies relative to a long-term mean (2014–2024), and recurrence maps.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},127972,1785943478,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":8,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":130,"read_time":81},2336474466412,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","Celia A. Baumhoer 1,*, Jonas Koehler 1, Stef Lhermitte2,3, Bert Wouters2 and Andreas J. Dietz 1  \n1German Aerospace Center (DLR), Earth Observation Center (EOC), German Remote Sensing DataCenter (DFD), Wessling, Germany, *[celia.baumhoer@dlr.de](celia.baumhoer@dlr.de)  \n[2](2 Delft University of Technology)[ Delft University of Technology](2 Delft University of Technology), [Department of Geoscience & Remote Sensing](Department of Geoscience & Remote Sensing), Delft, Netherlands, 3KU Leuven, Department of Earth & Environmental Sciences, Leuven, Belgium  \nMonitoring supraglacial lake dynamics in Antarctica with machine learning  \nMotivation  \nMonitoring supraglacial lake dynamics is important for understanding their impact on ice shelf stability and possible changes in the context of climate change. Supraglacial meltwater accumulation can drive accelerated ice discharge through various processes: surface runoff leading to ice thinning, basal meltwater injection causing basal sliding, and hydrofracture triggering ice shelf collapse and subsequent glacier acceleration. Furthermore, the increased presence of supraglacial lakes around the Antarctic margin can enhance melting due to the low albedo of lakes, which increases solar radiation absorption. Continuous monitoring is therefore essential for improving our understanding of the seasonal variations in the extent of supraglacial lakes, their effects on ice shelf stability, and the surface mass balance of ice sheets.  \nSentinel-1  \nSentinel-2  \nInternal Data Access  \nPre-Processing  \nPre-processing for Sentinel-1 IW data includes calibration, speckle filtering and orthorectification. Sentinel-2 L1C data is processed to L2A data. Pre-processing is performed on a HPC cluster with the ESA SNAP Toolbox.  \nDetection with DL  \nLake  \nTraining  \nNetwork  \nNeural  \nContinous search for new files over Antarctic supraglacial lakes during austral summer months (Nov-Mar)  \nMerging Lake Predictions  \nAll available supraglacial lake predictions for the first and second half of the month are merged to a bi-weekly maximum lake extent layer. Especially in cloudy conditions, Sentinel-1 SAR data provide important additional information.  \nAnalysing  \nDynamics  \nLake  \n1  \n3  \n4  \n5  \n6  \nBased on the time series since summer 2014/15 bi-weekly statistics on maximum lake extents, anomalies based on the long-term mean (2014-2024), and recurrence maps are automatically generated.  \n\n| Outlook\u003Cbr>Lakes4Antarctic [eoc.dlr.de](eoc.dlr.de)\u003Cbr>The combination of Sentinel-1 and Sentinel- 2 data enables the generation of bi-weekly supraglacial lake extents since 2014/15 . The presented processing workflow will be included in the automated processing pipeline to provide a seasonally updated dataset on supraglacial lake dynamics. Moreover, the derived and evaluated mapping products of the Lakes4Antarctica project will be used for modelling (e.g. hydrological routing) . | a. |\n| --- | --- |\n|  |  |\n\nThe neural network architecture is based on a U-Net with an additional ASPP (Atrous Spatial Pyramid Pooling) module for multiscale feature extraction. The architecture was trained separately for Sentinel-1 (S1) and Sentinel-2 (S2) data. Training with S2 was performed on 33 scene subsets from 15 different S2 scenes distributed along the Antarctic coastline based on four layers (green, red, SWI, AWEInsh). Training for S1 is only based on the channel with HH polarization over nine coastal regions. Accuracies based on random validation points have a Kappa coefficient of 0.925 (0.87) and a F1-score of 0.93 (0.87) for S1 and S2, respectively.  \nPine Island Shelf  \nMoscow  \nUniversity Ice Shelf  \nRayner Glacier, Enderby Land  \nBeaver Glacier, Amundsen Bay  \nSupraglacial Lake Recurrence  \n- >15  \n-7  \n-1  \nNickerson Ice Shelf  \n2  \nReferences  \n[1] M. Dirscherl, A. J. Dietz, and C. Kuenzer, . Seasonal evolution of Antarctic supraglacial lakes in 2015-2021 and links to environmental controls. The Cryosphere, 15, 5205-5226, [ht","cbCaibxI8jLxckAE","https://ap.wps.com/l/cbCaibxI8jLxckAE","pdf",2673512,"English","# Motivation\n# Data Sources and Pre-processing\n## Sentinel-1\n## Sentinel-2\n# Deep Learning Detection\n## U-Net with ASPP\n## Training and Validation\n# Lake Predictions and Time-Series Products\n## Merging Bi-weekly Extents\n## Anomalies and Recurrence Maps\n# Outlook","[{\"question\":\"Why is monitoring supraglacial lake dynamics important in Antarctica?\",\"answer\":\"It helps explain how meltwater accumulation can affect ice shelf stability and potentially drive accelerated ice discharge through runoff, basal sliding, or hydrofracture.\"},{\"question\":\"How are Sentinel-1 and Sentinel-2 data prepared for the machine learning workflow?\",\"answer\":\"Sentinel-1 IW data undergo calibration, speckle filtering, and orthorectification, while Sentinel-2 L1C is processed to L2A using the ESA SNAP toolbox on an HPC cluster.\"},{\"question\":\"What outputs are produced from the lake detections over time?\",\"answer\":\"The method merges lake predictions into bi-weekly maximum lake extent layers, then automatically generates statistics, anomalies relative to a long-term mean (2014–2024), and recurrence maps.\"}]","Monitoring Supraglacial Lake Dynamics in Antarctica with Machine Learning | PDF"]