Machine Learning, Physics, and Generative AI for Hydrologic and River Modeling

Moreno, Hernan

H102
Machine Learning, Physics, and Generative AI for Hydrologic and River Modeling

Session ID#: 249442

Session Description:

The integration of machine learning (ML), physics-informed AI, and generative artificial intelligence (GenAI) is rapidly advancing hydrology and river system modeling. These emerging approaches enhance predictive capabilities, improve model interpretability, and support data-driven scientific discovery. This session invites contributions on: (1) Data-driven and hybrid AI/ML models; (2) Physics-informed and learnable physical models; (3) Generative AI for data enhancement, downscaling, and uncertainty quantification; (4) AI-integrated Earth system modeling; (5) Scalable AI/ML approaches for watershed to global scales; (6) AI-assisted discovery of hydrologic patterns and relationships; and (7) Trustworthy, explainable, and responsible AI. We seek work that pushes the frontiers of AI for hydrologic science and water resource management.

Index Terms:

1805 Computational hydrology [HYDROLOGY]
1856 River channels [HYDROLOGY]
1942 Machine learning [INFORMATICS]
1954 Natural language processing [INFORMATICS]

Primary Convener: Hernan A Moreno, University of Texas at El Paso, Earth, Environmental and Resource Sciences, El Paso, TX, United States and Machine Learning in Hydrology

Conveners: Chaopeng Shen, Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, United States, Praveen Kumar, University of Illinois at Urbana-Champaign, Department of Civil and Environmental Engineering, Urbana, United States, Laura Alvarez, University of Texas at El Paso, Department of Earth, Environmental and Resource Sciences, El Paso, United States and Machine Learning in Hydrology

Student/Early Career Convener: Leila Constanza Hernandez Rodriguez, Lawrence Berkeley National Laboratory, Berkeley, United States and Machine Learning in Hydrology

Autonomous Scientific Reasoning for Hydrological Model Configuration: An AI Agent Approach to Iterative SUMMA Optimization (1850765)

Darri Eythorsson and Martyn P Clark, University of Calgary, Department of Civil Engineering, Schulich School of Engineering, Calgary, AB, Canada

Soil Moisture as an Alternative Approach Towards Better Water Management in Scarce Data Regions (1864395)

Khaoula Ait Naceur¹, El Mahdi El Khalki¹, Luca Brocca², Abdessamad Hadri³, Oumar Jaffar¹, Mohammed Elmehdi Saidi⁴, Vincent Simonneaux⁵, Simon Gascoin⁶, Mariame Rachdane¹ and Abdelghani Chehbouni⁷, (1)International Water Research Institute (IWRI), Mohammed VI Polytechnic University (UM6P), Ben Guerir, Morocco, (2)CNR-IRPI National Research Council, Perugia, Italy, (3)Mohammed VI Polytechnic University (UM6P), International Water research Institute (IWRI), Benguerir, Morocco, (4)L3G Laboratory, Department of Earth Sciences, Faculty of Sciences and Techniques, Cadi Ayyad University, Marrakech, Morocco, (5)Centre d’Etudes Spatiales de la Biosphère (CESBIO), Université de Toulouse, CNES/CNRS/INRAE/IRD/UPS,, Toulouse, France, (6)Centre d'Etudes Spatiales de la Biosphere, Toulouse Cedex 9, France, (7)Mohammed VI Polytechnic University, Center for Remote Sensing Applications, Benguerir, Morocco

KAN-Matrix: Visualizing Nonlinear Pairwise and Multivariate Contributions for Physical Insight (1865067)

Luis De la Fuente, University of Texas at El Paso, El Paso, United States, Hernan A Moreno, University of Texas at El Paso, Earth, Environmental and Resource Sciences, El Paso, TX, United States, Laura Alvarez, University of Texas at El Paso, Department of Earth, Environmental and Resource Sciences, El Paso, United States and Hoshin Gupta, Hydrology and Atmospheric Sciences, The University of Arizona, Tucson, United States

Scoring Rule-Based LSTM Models for Flow Forecasting (1865106)

Mohammad Farmani, University of Arizona, Hydrology and Atmospheric Sciences, Tucson, AZ, United States, Dr. Arman Oliazadeh, PhD, University of Georgia, Department of Geography, Athens, United States, Habtamu Tamiru, University of Alabama, Civil, Construction and Environmental Engineering, Tuscaloosa, United States, Fatemeh Yavari, The City University of New York (CCNY), Civil Engineering, New York, United States, Amobichukwu Amanambu, University of Alabama, Department of Geography and the Environment, Tuscaloosa, United States and Jonathan M Frame, University of Alabama, Department of Geological Sciences, Tuscaloosa, AL, United States

Sub-Pixel Soil Moisture in Arid Ecosystems: From Data-Driven Models to Interpretable Equations (1866236)

Stephanie N Marquez and Hernan A Moreno, University of Texas at El Paso, Earth, Environmental and Resource Sciences, El Paso, TX, United States

Introducing Knowledge Distillation for Deep Learning-Based Hydrological Prediction (1875924)

Mohammad Sina Jahangir, Cornell University, Biological and Environmental Engineering, Ithaca, United States, John Quilty, University of Waterloo, Civil and Environmental Engineering, Waterloo, ON, Canada, Scott Steinschneider, Cornell University, Ithaca, United States and Jan F Adamowski, McGill University, Department of Bioresource Engineering, Montreal, QC, Canada

A hybrid Approach for River Flood Forecasting by Combining a Hydrodynamic Flow Simulation Model and Deep Learning Networks (1876789)

Kyung Soo Jun and Jae Gyeong Lee, Graduate School of Water Resources, Sungkyunkwan University, Suwon, South Korea

Physical-informed Graph Learning for Networked Water Systems Flow Monitoring (1883749)

Xudong Fan and Linyue Luo, University at Buffalo, Department of Civil, Structural, and Environmental Engineering, Buffalo, United States

A Comparative Study of Physics-Informed and Data-Driven Neural Networks for Compound Flood Simulation at River-Ocean Interfaces: A Case Study of Hurricane Irene (1887074)

Dongyu Feng¹, Zeli Tan¹, Honghui Du², Donghui Xu¹, Cheng-Wei Yu³ and Qizhi He⁴, (1)Pacific Northwest National Laboratory, Richland, WA, United States, (2)University of Minnesota, Department of Civil, Environmental, and Geo- Engineering, Minneapolis, United States, (3)National Taiwan University, Taipei, Taiwan, (4)University of Minnesota, Department of Civil, Environmental, and Geo- Engineerin, Minneapolis, United States

Modeling Groundwater Levels by Latent Space Gaussian Process Regression with Physics-Informed Constraints from InSAR Data (1888429)

Anshuman Pradhan, C3 AI Inc., Data Science, Redwood City, United States and Zhen Liu, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, United States

Hybrid Deep Learning Models for Streamflow Prediction and Hydrologic Understanding in Small Watersheds: A Case Study of the Rivanna River, Virginia (1923131)

Mr. Aashutosh Aryal, University of Virginia, Environmental Institute, Charlottesville, United States and Venkataraman (Venkat) Lakshmi, University of Virginia, Civil and Environmental Engineering, Charlottesville, United States

A Hybrid SWAT–Machine Learning Framework for Streamflow and Sediment Simulation in the Big Sandy River Basin, Kentucky (1927926)

Dipesh Oli and Buddhi Gyawali, Kentucky State University, Frankfort, United States

Machine Learning-Based Streamflow Simulation for Large-Scale River Networks: Enhancing Data-Driven Channel Routing with Physics-Based Data Augmentation (1928124)

Xing Zheng, Abhishek Basu, Kiran Chinnayakanahalli, Keith A Sawicz and Hemant Chowdhary, Verisk Analytics, Research, Boston, United States

Accurate and Efficient Hourly Streamflow Downscaling or Prediction with Diffusion Model (1930875)

Wencong Yang¹, Haoyu Ji² and Chaopeng Shen², (1)Pennsylvania State University University Park, State College, PA, UNITED STATES, (2)Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, United States

Estimation of Saturated Hydraulic Conductivity from Environmental Covariates: Machine Learning Approach (1932419)

Berhanu Sinshaw¹, Saswata Nandi¹, Teamrat A Ghezzehei², Joshua H Viers¹ and Mohammad Safeeq¹, (1)Civil and Environmental Engineering, University of California, Merced, United States, (2)Life and Environmental Sciences, University of California, Merced, United States

Spatiotemporal Super-Resolution of Urban Flood Predictions using Multimodal Deep Learning (1940166)

Ms. Hyeonjin Choi, Integrated Master's and Ph.D. Program, Kumoh National Institute of Technology, Civil Engineering, Gumi, Korea, Republic of (South), Hyuna Woo, Kumoh National Institute of Technology, Gumi-si, South Korea, Dongsop Rhee, Korea Institute of Civil Engineering and Building Technology, Goyang, South Korea, Hyungon Ryu, NVIDIA, Seoul, South Korea and Seong Jin Noh, Kumoh National Institute of Technology, Civil Engineering, Gumi, South Korea

Enhancing Parameter Calibration in Land Surface Models Using a Multi-Task Surrogate Model within a Differentiable Parameter Learning Framework (1940418)

Wenpeng Xie¹, Hongmei Li² and Kei Yoshimura², (1)The University of Tokyo, Institute of Industrial Science, Tokyo, Japan, (2)The University of Tokyo, Kashiwa, Japan

Leveraging Large-Scale Meteorological Geospatial Information to Predict Snow Water Equivalent with Machine Learning (1947219)

Savalan Naser Neisary, The University of Alabama, Department of Civil, Construction, and Environmental Engineering, Tuscaloosa, United States and Jonathan M Frame, University of Alabama, Department of Geological Sciences, Tuscaloosa, AL, United States

Indirect Parameter Optimization of Hybrid Flood prediction Models Using SHAP-Based Explainable AI (1953366)

Faridoddin Darroudi, Sharif University of Technology, Tehran, Iran and Mohammad Danesh-Yazdi, Sharif University of Technology, Department of Civil Engineering, Tehran, Iran

A Modular Physics-Inspired Hybrid Machine Learning HBV Model with Dynamic Parameter Selection for Streamflow Prediction (1963066)

Rahmathulla Madathi Parambath and Jayaprakash Rajasekharan, Norwegian University of Science and Technology, Trondheim, Norway

Forecasting Hydrological Extremes in the Fraser River Basin using a Hybrid LSTM-GRU Deep Learning Model (1971049)

Muhammed Ucun, University of Virginia, Charlottesville, VA, United States, Sophia Bakar, RTI International, Research Triangle Park, United States and Venkataraman (Venkat) Lakshmi, University of Virginia, Civil and Environmental Engineering, Charlottesville, United States

Long Short-Term Memory Model to Forecast River Ice Breakup Throughout Alaska USA (1973603)

Jitendra Kumar, Oak Ridge National Laboratory, Environmental Sciences Division, Oak Ridge, United States, Russell Limber, Oak Ridge National Laboratory, Oak Ridge, TN, United States, Forrest Hoffman, Oak Ridge National Laboratory, Computational Sciences & Engineering Division, Oak Ridge, United States and Jonathon P Schwenk, Los Alamos National Laboratory, Los Alamos, United States

Interpretable Equations for Reservoir Temperature Prediction Using Symbolic Machine Learning (1974043)

Isabela Suaza Sierra, University of Texas at El Paso, El Paso, United States and Hernan A Moreno, The University of Texas at El Paso, El Paso, United States

The power of many: machine learning ensembles to improve hydrologic predictions (Invited) (1985791)

Charuleka Varadharajan, Lawrence Berkeley National Laboratory, Earth and Environmental Sciences Area, Berkeley, United States, Jared Willard, University of Minnesota, Minneapolis, United States, Yuan-Heng Wang, The University of Arizona, Hydrology and Atmospheric Science, Tucson, United States, Fabio Ciulla, Lawrence Berkeley National Laboratory, Earth and Environmental Sciences Area, Berkeley, CA, United States, Pu Ren, Lawrence Berkeley National Laboratory, Berkeley, CA, United States and Michael Mahoney, University of California Berkeley, Berkeley, United States

See more of: Hydrology

<< Previous Session Proposal | Next Session Proposal >>

H102 Machine Learning, Physics, and Generative AI for Hydrologic and River Modeling

Session Description:

Index Terms:

H102
Machine Learning, Physics, and Generative AI for Hydrologic and River Modeling