[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83923-en":3,"doc-seo-83923-105":29,"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":20,"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":13,"seo_description":14,"update_tm":27,"read_time":28},83923,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Probing Geospatial SSL Representations with Environmental Signals","Self-supervised learning (SSL) aims to learn transferable geospatial representations, yet most benchmarks judge quality only through downstream task metrics. This work evaluates representation contents directly by testing whether SSL embeddings of satellite imagery preserve statistical associations with co-varying environmental variables from ERA5 reanalysis. Probing targets include temperature, precipitation, surface solar radiation, surface pressure, and volumetric soil water. Complementary intrinsic geometry metrics relate representation accessibility to downstream performance. The study also releases ERA5 co-located annotations for future physically grounded evaluation of geospatial foundation models.","Probing Geospatial SSL Representations with Environmental Signals  \nRohita Mocharla 1 ,2  \n1Johns Hopkins Applied Physics Laboratory  \nLaurel, MD [nmochan1@jhu.edu](nmochan1@jhu.edu)  \nVishal M. Patel2  \n2Johns Hopkins University Baltimore, MD  \n[vpatel36@jhu.edu](vpatel36@jhu.edu)  \narXiv :2607 .05207v 1 [ cs .CV] 6 Jul 2026  \nAbstract  \nSelf-supervised learning (SSL) is designed to learn generic, transferable representations rather than representations optimized for a single task. Most geospatial benchmarks evaluate representations solely through downstream tasks, providing limited insight into the information encoded within the representation itself. We ask a different question: do SSL representations of satellite imagery preserve statistical associations with environmental variables that co-vary with the imaging process? To answer this question, we probe SSL representations using co-located ERA5 reanalysis variables [12], a global dataset of physically consistent environmental variables, including temperature, precipitation, surface solar radiation, surface pressure, and volumetric soil water. These variables are physically related to the spectral reflectance and radar backscatter recorded by Sentinel-1 and Sentinel-2, making them meaningful evaluation targets despite not being used during SSL pretraining. We complement this probing analysis with intrinsic representation metrics to characterize representation geometry and investigate how these properties relate to downstream performance and the encoding of environmental signals. Using DINO [3], MAE [11], and MoCo [10] models trained under identical conditions, we show that representation-level metrics distinguish models with similar downstream benchmark performance, providing complementary information beyond task-driven benchmarks. We further find that the linear accessibility of environmental signals is associated with performance on environmentally dependent tasks in the PANGAEA benchmark [21]. Finally, we release ERA5 [12] annotations co-located with the SSL4EO [36] dataset to enable physically grounded representation evaluation for future geospatial foundation models.  \n1. Introduction  \nThe goal of self-supervised learning (SSL) is to learn generalizable representations that transfer across tasks rather than  \nfitting to a specific label set. Existing geospatial benchmarks primarily evaluate learned representations through downstream task performance. While downstream evaluation measures transfer to specific tasks, it provides an incomplete view of representation quality because no benchmark can exhaustively evaluate the broad range of downstream applications SSL representations are intended to support [20] . As SSL representations are designed for unknown future tasks, representation-level diagnostics provide complementary insight beyond any fixed downstream benchmark. Recent work has shown that even when models achieve similar downstream performance, they can organize information differently in latent space, suggesting that the mechanisms underlying their downstream performance are functionally different [24] . This motivates evaluation protocols that characterize representations beyond task utility.  \nIn this work, we introduce a representation-driven evaluation protocol grounded in the physical processes underlying Earth observation imagery. Rather than measuring representations solely by task performance, we probe whether they maintain statistical associations with environmental variables that co-vary with satellite observations. Specifically, we use ERA5 reanalysis variables [12]—a global atmospheric reanalysis dataset that combines physical weather models with observational data—to estimate environmental factors such as temperature, precipitation, surface solar radiation, surface pressure, and volumetric soil water. These physically grounded variables are used as probing targets because they directly influence the spectral reflectance and radar backscatter obse","cbCaioKMdcLBG9bm","https://ap.wps.com/l/cbCaioKMdcLBG9bm","pdf",3918767,1,16,"English","en",105,"# Abstract\n# Introduction\n## Problem with downstream-only geospatial SSL benchmarks\n## Representation-driven, physically grounded evaluation protocol\n## ERA5 probing and intrinsic representation diagnostics\n## Key contributions","[{\"question\":\"What problem does the paper address with existing geospatial SSL benchmarks?\",\"answer\":\"It argues that most benchmarks evaluate SSL representations only via downstream task performance, which gives limited visibility into what environmental information is encoded inside the representation itself.\"},{\"question\":\"How does ERA5 probing work in the proposed evaluation protocol?\",\"answer\":\"The method uses ERA5 reanalysis variables co-located with satellite imagery as probing targets, testing linear and nonlinear accessibility of environmental factors such as temperature, precipitation, surface solar radiation, surface pressure, and soil water.\"},{\"question\":\"What intrinsic representation metrics are used, and why?\",\"answer\":\"It uses metrics like alignment, uniformity, effective rank (eRank), and off-diagonal covariance (offDiag) to characterize representation geometry independently of downstream tasks, enabling comparison across SSL models.\"}]",1784191463,40,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"probing-geospatial-ssl-representations-with-environmental-signals","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/probing-geospatial-ssl-representations-with-environmental-signals/83923/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the paper address with existing geospatial SSL benchmarks?","Question",{"text":75,"@type":76},"It argues that most benchmarks evaluate SSL representations only via downstream task performance, which gives limited visibility into what environmental information is encoded inside the representation itself.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does ERA5 probing work in the proposed evaluation protocol?",{"text":80,"@type":76},"The method uses ERA5 reanalysis variables co-located with satellite imagery as probing targets, testing linear and nonlinear accessibility of environmental factors such as temperature, precipitation, surface solar radiation, surface pressure, and soil water.",{"name":82,"@type":73,"acceptedAnswer":83},"What intrinsic representation metrics are used, and why?",{"text":84,"@type":76},"It uses metrics like alignment, uniformity, effective rank (eRank), and off-diagonal covariance (offDiag) to characterize representation geometry independently of downstream tasks, enabling comparison across SSL models.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":28,"slug":118},7,"Healthcare","healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]