[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122579-en":3,"doc-seo-122579-105":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":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":27,"seo_description":14,"update_tm":28,"read_time":29},122579,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","Knowledge-Guided Machine Learning for Operational Flood Forecasting","Knowledge-guided machine learning framework for operational hydrologic forecasting at the catchment scale is presented, built around a Factorized Hierarchical Neural Network (FHNN) with inverse and forward models. The inverse model encodes the current catchment state from observed precipitation, temperature, and streamflow, while the forward model predicts streamflow from that state. FHNN integrates real-time observations via inference-based data integration more efficiently than computationally intensive data assimilation. Comparisons on CAMELS-US and National Weather Service operational flood forecasts show FHNN outperforms an autoregressive LSTM and matches or exceeds NWS expert-derived skill after 12–18 hours.","RESEARCH ARTICLE  \n10.1029/2024WR039064  \nKey Points:  \n• Our approach builds hierarchically related, multiscale catchment states via an inverse model that integrates observed information  \n• Our model generally outperforms an expert human forecaster using a physics‐based model 12–18+ hours after forecast generation  \n• Our method outperforms a state‐of‐the‐ art alternative (Autoregressive LSTM), especially in arid catchments where other methods struggle  \nSupporting Information:  \nSupporting Information may be found in the online version of this article.  \nCorrespondence to:  \nZ. McEachran,  \n[mceac015@umn.edu](mceac015@umn.edu)  \n[Citation:](Citation:)  \nMcEachran, Z., Ghosh, R., Renganathan, A., Sharma, S., Lindsay, K., Steinbach, M., et al. (2025) . Knowledge‐guided machine learning for operational flood forecasting. Water Resources Research, 61, e2024WR039064. [https://doi.org/10.1029/](https://doi.org/10.1029/)[ ](https://doi.org/10.1029/)2024WR039064  \nReceived 29 SEP 2024  \nAccepted 3 OCT 2025  \n© 2025. The Author(s) . This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.  \nThis is an open access article under the terms of the Creative Commons Attribution‐NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.  \nKnowledge‐Guided Machine Learning for Operational Flood Forecasting  \nZac McEachran1 , Rahul Ghosh2, Arvind Renganathan2 , Somya Sharma2, Kelly Lindsay2, Michael Steinbach2 , John Nieber3 , Christopher Duffy4, and Vipin Kumar2  \n1University of Minnesota Climate Adaptation Partnership, University of Minnesota, St. Paul, MN, USA, 2Department of Computer Science and Engineering, University of Minnesota, Minneapolis, MN, USA, 3Department of Bioproducts and Biosystems Engineering, University of Minnesota, St. Paul, MN, USA, 4Department of Civil and Environmental Engineering, Pennsylvania State University, University Park, PA, USA  \nAbstract We present a knowledge‐guided machine learning framework for operational hydrologic forecasting at the catchment scale. Our approach, a Factorized Hierarchical Neural Network (FHNN), has two main components: inverse and forward models. The inverse model uses observed precipitation, temperature, and streamflow data to generate a representation of the current underlying catchment state. The forward model predicts streamflow using the learned catchment state. The FHNN architecture is designed to model multi‐scale processes and capture their interactions, a critical ability for flood modeling. FHNN also improves forecasts based on real‐time data through an inference‐based data integration approach using inverse modeling. FHNN's data integration approach improves forecasts in response to observed data more efficiently than data assimilation methods that require computationally intensive optimization. We compare the FHNN to a leading deep learning alternative (autoregressive LSTM) on the large‐sample CAMELS‐US data set, and operational flood forecast data from the US National Weather Service (NWS) . Official NWS flood forecasts are generated by expert human forecasters using a physics‐based model, in a human‐in‐the‐loop process. Thus, we assess the flood forecast ability of FHNN by directly comparing its performance against these NWS expert‐derived forecasts. The human forecaster creates a more accurate forecast within the first 12–18 hr of a forecast's issuance, but FHNN has significantly better predictions thereafter. This research lays the groundwork for leveraging the predictive performance of AI‐based models with the expertise in forecasting agencies to produce better river forecasts.  \nPlain Language Summary Recent advances in Machine Learning (ML) have made strides in improving the prediction of river levels, and often outperform physics‐based models that make a forecast based on the physics of how riv","cbCaidjaeFlFr9Ew","https://ap.wps.com/l/cbCaidjaeFlFr9Ew","pdf",3903594,1,18,"English","en",105,"# Abstract\n# Introduction\n## Operational forecasting and real-time data integration\n# Model framework\n## Factorized Hierarchical Neural Network (FHNN)\n# Experiments and evaluation\n## Comparison against expert NWS forecasts\n## Comparison with autoregressive LSTM","[{\"question\":\"What is the core idea of the knowledge-guided ML approach in this work?\",\"answer\":\"It uses a Factorized Hierarchical Neural Network whose architecture is informed by hydrologic science, combining inverse modeling for catchment state estimation with forward modeling for streamflow prediction.\"},{\"question\":\"How does FHNN use real-time observations during forecasting?\",\"answer\":\"FHNN performs inference-based data integration, updating forecasts in response to observed precipitation, temperature, and streamflow without requiring computationally intensive optimization typical of some data assimilation methods.\"},{\"question\":\"How does FHNN perform compared with expert human forecasts and an autoregressive LSTM?\",\"answer\":\"FHNN generally outperforms the autoregressive LSTM, especially in arid catchments, and it yields significantly better predictions than NWS expert-derived forecasts after roughly 12–18 hours, while experts are more accurate in the first 12–18 hours.\"}]","Knowledge-Guided Machine Learning for Operational Flood Forecasting | PDF",1785811416,45,{"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},"knowledge-guided-machine-learning-for-operational-flood-forecasting","",{"@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/knowledge-guided-machine-learning-for-operational-flood-forecasting/122579/",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":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the core idea of the knowledge-guided ML approach in this work?","Question",{"text":75,"@type":76},"It uses a Factorized Hierarchical Neural Network whose architecture is informed by hydrologic science, combining inverse modeling for catchment state estimation with forward modeling for streamflow prediction.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does FHNN use real-time observations during forecasting?",{"text":80,"@type":76},"FHNN performs inference-based data integration, updating forecasts in response to observed precipitation, temperature, and streamflow without requiring computationally intensive optimization typical of some data assimilation methods.",{"name":82,"@type":73,"acceptedAnswer":83},"How does FHNN perform compared with expert human forecasts and an autoregressive LSTM?",{"text":84,"@type":76},"FHNN generally outperforms the autoregressive LSTM, especially in arid catchments, and it yields significantly better predictions than NWS expert-derived forecasts after roughly 12–18 hours, while experts are more accurate in the first 12–18 hours.","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"]