H101
Machine Learning, Data Analytics and Data Assimilation for Earth System Modeling and Discovery
Session ID#: 252968
Session Description:
Machine Learning (ML), Data Analytics, and Data Assimilation (DA) have rapidly advanced Earth science applications, driven by progress in satellite technologies, Earth observations, and computational infrastructures. This session will be devoted to the methodological development and applications of data-driven and physics-informed ML, deep learning, generative AI, post-hoc explainable AI, Large Language Models, and DA in improving our understanding and enhancing the prediction of different aspects of hydrological and hydrometeorological processes and their interactions across spatial and temporal scales. We encourage studies with a focus on predictability, interpretability, and uncertainties in these approaches while utilizing complex remotely sensed satellite data, in-situ observations, and reanalysis data using high-performance computing and physics-based ML emulators. We also welcome contributions on statistically based model–data fusion methods, including inverse modelling, Markov Chain Monte Carlo (MCMC), and related Bayesian techniques. Contributions on coupled ML and DA algorithms for hydroclimatic extremes are encouraged.
Index Terms:
1817 Extreme events [HYDROLOGY]
1847 Modeling [HYDROLOGY]
1869 Stochastic hydrology [HYDROLOGY]
1873 Uncertainty assessment [HYDROLOGY]
Primary Convener: Peyman Abbaszadeh, Portland State University, Civil and Environmental Engineering, Portland, United States
Conveners: Hamid Moradkhani, The University of Alabama, Center for Complex Hydrosystems Research, Tuscaloosa, United States, Harrie-Jan Hendricks Franssen, Forschungszentrum Jülich GmbH, 52425 Jülich, Germany and Xin Li, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China
Explainable Deep Transfer Learning for Spatiotemporal Runoff Prediction at the Continental Scale (1857040)
Seyed Mohammad Mosavat1, Ehsan Foroumandi1, Parnian Ghaneei2 and Hamid Moradkhani1, (1)The University of Alabama, Center for Complex Hydrosystems Research, Tuscaloosa, United States, (2)University of Alabama, Civil, Construction, and Environmental Engineering, Tuscaloosa, AL, United States
NLP and BERT Deep Learning-based Approaches for Improved Impact-based Forecasting and Disaster Management (1857337)
Akshay Singhal1, Mohammadsepehr Karimiziarani2 and Hamid Moradkhani1, (1)The University of Alabama, Center for Complex Hydrosystems Research, Tuscaloosa, United States, (2)The University of Alabama, Center for Complex Hydrosystems Research, Tuscaloosa, AL, United States
Exploring the Benefits of Assimilating the USGS NGWOS Streamflow Observations into a Distributed Hydrology Model (1868917)
Fatemeh Shirkhanloo1, Arezoo Rafieeinasab2, Adam Stonewall3, Susan Wherry3, Caelan Simeone4 and Peyman Abbaszadeh5, (1)Portland State University, Portland, OR, United States, (2)University Corporation for Atmospheric Research, Boulder, United States, (3)USGS Oregon Water Science Center, Portland, United States, (4)U.S. Geological Survey, Oregon Water Science Center, Portland, United States, (5)Portland State University, Portland, United States
Enhancing Snow Simulation in the E3SM Land Model (ELM) over the Western U.S. using Ensemble-Based Data Assimilation (1870050)
Fatemeh Gholizadeh1, Ali Gholinia2, Hoang Tran3, Yilin Fang4, L. Ruby Leung4 and Peyman Abbaszadeh1, (1)Portland State University, Civil and Environmental Engineering, Portland, United States, (2)Portland State University, Civil and Environmental Engineering, Portland, OR, United States, (3)Pacific Northwest National Laboratory, Richland, United States, (4)Pacific Northwest National Laboratory, Richland, WA, United States
Coupling the E3SM Land Model (ELM) with Evolutionary Data Assimilation to Improve the Representation and Modeling of Land Surface Processes (1874199)
Ali Gholinia1, Fatemeh Gholizadeh2, Hoang Tran3, Yilin Fang4, L. Ruby Leung3 and Peyman Abbaszadeh2, (1)Portland State University, Civil and Environmental Engineering, Portland, OR, United States, (2)Portland State University, Civil and Environmental Engineering, Portland, United States, (3)Pacific Northwest National Laboratory, Richland, United States, (4)Pacific Northwest National Laboratory, Richland, WA, United States
Streamflow-Informed Correction of Radar Rainfall Biases for Improved Flood Forecasting (1894242)
Zhihua Li, University of Iowa, Iowa Flood Center and IIHR-Hydroscience and Engineering; Department of Mathematics, Iowa City, United States, Ruben Molina, University of Iowa, Iowa Flood Center and IIHR-Hydroscience and Engineering, Iowa City, IA, United States, Nicolás Velásquez, Florida Institute of Technology, Mechanical and Civil Engineering Department, Melbourne, FL, United States, Xueyu Zhu, University of Iowa, Department of Mathematics, Iowa City, United States and Witold F Krajewski, University of Iowa, Iowa Flood Center and IIHR-Hydroscience and Engineering, Iowa City, United States
Modeling Depth-to-Water (DTW) Across the Contiguous U.S. Using Environmental, Remote Sensing and Human Drivers (1903686)
Qinyuan Dai1, Karem Abdelmohsen2, James S Famiglietti2 and Tianfang Xu1, (1)Arizona State University, School of Sustainable Engineering and the Built Environment, Tempe, United States, (2)Arizona State University, School of Sustainability, Tempe, United States
Machine-Learning Based Prediction of Water Surface Elevation Using SWOT Satellite Data and USGS Observational Data (1919059)
Anika Islam1, Susan Wherry2, Adam Stonewall2, Caelan Simeone3 and Peyman Abbaszadeh4,5, (1)Portland State University, Portland, OR, United States, (2)USGS Oregon Water Science Center, Portland, United States, (3)U.S. Geological Survey, Oregon Water Science Center, Portland, United States, (4)San Francisco, CA, United States, (5)Portland State University, Civil and Environmental Engineering, Portland, United States
Inverse Machine Learning for 3D Subsurface Characterization and Pumping Estimation in CONUS-Scale Models (1923625)
Louisa Pawusch, Princeton University, Civil and Environmental Engineering, Princeton, NJ, United States, Andrew Bennett, University of Arizona, Department of Hydrology & Atmospheric Sciences, Tucson, AZ, United States, Peter Melchior, Princeton University, Astrophysical Sciences, Princeton, United States, Laura E Condon, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, AZ, United States, Danielle Tijerina-Kreuzer, University of California Berkeley, Energy and Resources Group, Berkeley, United States and Reed M Maxwell, Princeton University, Civil and Environmental Engineering, Princeton, United States
Enhancing Quantile Mapping Bias Correction of Land Reanalysis Products (1944217)
Ali Nazemi, Concordia University, Department of Building, Civil and Environmental Engineering, Montreal, QC, Canada and Amir Zokaei, Concordia University, Building, Civil and Environmental Engineering, Montreal, QC, Canada
Modeling of Storm Surges using Deep Learning and Transfer Learning (1952449)
Meghana Nagaraj, University of Central Florida, Orlando, FL, United States, Alejandra R Enriquez, University of South Florida Tampa, Tampa, United States, Thomas Wahl, Department of Civil, Environmental, and Construction Engineering & National Center for Integrated Coastal Research, University of Central Florida, Orlando, United States and Alejandra Enriquez
Peeking into Complex Hydrologic Dynamics: Inferring Storage–Stormflow Relationships with a Copula-State Space Model Framework (1953562)
Helen Weierbach1, Majid Bayati2, Joseph Janssen1 and Ali Ameli1, (1)University of British Columbia, Department of Earth, Ocean and Atmospheric Sciences, Vancouver, BC, Canada, (2)University of British Columbia, Department of Earth, Ocean and Atmospheric Sciences, Vancouver, Canada
