H140
Toward a hybrid era of physical and machine learning model in simulating and predicting extreme events.
Session ID#: 252798
Session Description:
Extreme events often result in severe losses in human society and environment. Accurate simulation and prediction of extreme events have been relied highly on the physical models, whereas the computational cost and parameterization deficits have limited the ability of current physical models. Recently, the rapid advancement in machine learning has offered promising solutions to these challenges, though issues related to lack of physical constraints, representational fidelity, and interpretability remain unresolved. Hybrid models appear to be the answer in the new era, promoting the simulation, prediction, and projection in extreme events and consequent impacts. This session invites researchers from both the physical modeling and machine learning communities and aims to build the bridge between two areas.
Index Terms:
1622 Earth system modeling [GLOBAL CHANGE]
1627 Coupled models of the climate system [GLOBAL CHANGE]
1817 Extreme events [HYDROLOGY]
3355 Regional modeling [ATMOSPHERIC PROCESSES]
Primary Convener: Zeyu Xue, PhD, Pacific Northwest National Laboratory, ASGC, Richland, WA, United States
Conveners: Wenli Zhao, Columbia University, Department of Earth and Environmental Engineering, New York, NY, United States, Shiheng Duan, Lawrence Livermore National Laboratory, Livermore, CA, United States and Ye Liu, Pacific Northwest National Laboratory, Richland, WA, United States
Student/Early Career Convener: Yusuke Hiraga, Tohoku University, Department of Civil and Environmental Engineering, Sendai, Japan
Enhanced Soil Moisture Prediction at Subseasonal-to-Seasonal Scales Using a Deep Learning–Physics Hybrid Approach (1874392)
Miao Zhang, University of California Berkeley, Department of Environmental Science, Policy, and Management, Berkeley, CA, United States; Nanjing University of Information Science and Technology, Nanjing, China, Xing Yuan, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China, Wenli Zhao, Columbia University, Department of Earth and Environmental Engineering, New York, United States and Trevor Keenan, University of California, Berkeley, Department of Environmental Science, Policy, and Management, Berkeley, United States
Deep Learning Atmospheric Models Reliably Simulate Out-of-Sample Temperature Extremes, and Blocking Frequencies (1900921)
Zilu Meng, University of Washington, Atmospheric and Climate Science, Seattle, United States, Gregory J Hakim, University of Washington, Department of Atmospheric and Climate Science, Seattle, United States, Wenchang Yang, Princeton University, Department of Geosciences, Princeton, NJ, United States and Gabriel Andres Vecchi, Princeton University, Department of Geosciences and The High Meadows Environmental Institute, Princeton, NJ, United States
High-Resolution Ensemble Forecasting of Extreme Weather Using a Stretched-Grid AI Weather Model (1929449)
Yanbo Nie1, Agniv Sengupta1, Jorge Baño-Medina2, Daniel Steinhoff1, Patrick Mulrooney1 and Luca Delle Monache1, (1)Center for Western Weather and Water Extremes (CW3E), Scripps Institution of Oceanography, UC San Diego, La Jolla, United States, (2)Institute of Physics of Cantabria, Santander, Spain
Prediction of Peak Breach Discharge of Non-Cohesive Earth Dams Using Gradient Boosting and Ensemble Methods (1952075)
MD Ajwad Anwar1, Mohamed Torkomany2, Ezzat Elalfy3, Austin Downey4, Hanif Chaudhry2 and Jasim Imran5, (1)University of South Carolina Columbia, Civil and Environmental Engineering, Columbia, SC, United States, (2)University of South Carolina Columbia, Columbia, United States, (3)University of South Carolina, Civil and Environmental Engineering, Columbia, United States, (4)University of South Carolina, Mechanical Engineering, Columbia, United States, (5)Univ South Carolina, Department of Civil and Environmental Engineering, Columbia, United States
The AI-surpassing Physics-embedded AI model for Capturing Unseen Extreme Flood Events and Climate Change Impact Assessment (Invited) (1977293)
Yalan Song1, Chaopeng Shen2, Kamlesh Sawadekar3, Haoyu Ji2, Wencong Yang4 and Leo Lonzarich2, (1)Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, PA, United States, (2)Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, United States, (3)Pennsylvania State University Main Campus, Civil and Environmental Engineering, University Park, PA, United States, (4)Pennsylvania State University University Park, State College, PA, UNITED STATES
Global High-Resolution Hydrologic Simulation to Enhance Predictions in Ungauged Regions with Hybrid Differentiable Models (Invited) (1991254)
Dapeng Feng, Stanford University, Earth System Science, Stanford, CA, United States, Yuan Yang, University of California San Diego, Scripps Institution of Oceanography, Center for Western Weather and Water Extremes, La Jolla, United States, Hylke Beck, King Abdullah University of Science and Technology, Physical Science and Engineering Division, Thuwal, Saudi Arabia, Chaopeng Shen, Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, United States and Ming Pan, Center for Western Weather and Water Extremes (CW3E), Scripps Institution of Oceanography, University of California San Diego, La Jolla, United States
