A52G:
Novel Methods for Combining Physical Simulation, Machine Learning, and Data-Driven Analysis in Climate Studies and Geophysical Sciences I

Submit an Abstract to this Session



Session ID#: 33447

Session Description:
Simulation of physical processes through solution of differential equations, or mechanistic models, serves as a core tool of geophysical studies. In contrast to cause and effect-driven Physical Simulation (PS), Machine Learning (ML) models instantiate pattern recognition techniques and often operate as a black box. Additionally, advances in data collection and data-driven analysis provide new opportunities for insights in the climate sciences and beyond. Novel data-based techniques offer new approaches to uncertainty and error quantification, pattern recognition, and high-dimensional data analysis. Questions of how to combine PS and ML solutions and use data-driven analysis to advance understanding of physical systems and improve predictive models are the focus of this session. Cross-disciplinary presentations that demonstrate applied combinations or specific applications of PS with ML and data-driven analysis are encouraged. Examples may include data assimilation, filtering, use of PS to develop training data sets for ML and approaches to classification of PS outputs.
Primary Convener:  Sean A McKenna, IBM Ireland, Dublin, Ireland
Convener:  Ronni Grapenthin, University of Alaska Fairbanks, Geophysical Institute, Fairbanks, AK, United States
Chairs:  Ronni Grapenthin, University of Alaska Fairbanks, Geophysical Institute, Fairbanks, AK, United States, Sean A McKenna, IBM Ireland, Dublin, Ireland and Zhaohua Wu, Florida State University, Tallahassee, United States
OSPA Liaison:  Ronni Grapenthin, University of Alaska Fairbanks, Geophysical Institute, Fairbanks, AK, United States

Cross-Listed:
  • B - Biogeosciences
  • H - Hydrology
  • NG - Nonlinear Geophysics
  • S - Seismology
Index Terms:

0430 Computational methods and data processing [BIOGEOSCIENCES]
1816 Estimation and forecasting [HYDROLOGY]
3315 Data assimilation [ATMOSPHERIC PROCESSES]
7212 Earthquake ground motions and engineering seismology [SEISMOLOGY]

Abstracts Submitted to this Session:

Learning Physics-based Models in Hydrology under the Framework of Generative Adversarial Networks (295246)
Anuj Karpatne, University of Minnesota Twin Cities, Department of Computer Science and Engineering, Minneapolis, MN, United States and Vipin Kumar, University of Minnesota Twin Cities, Department of Computer Science/Engineering, Minneapolis, United States
Advanced Machine Learning Emulators of Radiative Transfer Models (Invited) (294263)
Gustau Camps-Valls1, Jochem Verrelst2, Luca Martino2 and Jorge Vicent2, (1)Image Processing Laboratory, Universitat de València, Paterna, Spain, (2)University of Valencia, Valencia, Spain
Creating Weather System Ensembles Through Synergistic Process Modeling and Machine Learning (293860)
Derek J Posselt1, Hai Nguyen2, Berlin Chen2, Longtao Wu3, Hui Su4 and Amy J Braverman2, (1)NASA Jet Propulsion Laboratory, Pasadena, United States, (2)Jet Propulsion Laboratory, Pasadena, CA, United States, (3)Jet Propulsion Lab, Pasadena, CA, United States, (4)Jet Propulsion Lab, Pasadena, United States
Enabling large-scale viscoelastic calculations via neural network acceleration (Invited) (233017)
Phoebe Robinson DeVries, T Ben Thompson and Brendan J. Meade, Harvard University, Cambridge, MA, United States
Machine learning model trained on generalized groundwater flow models to estimate regional distribution of groundwater age in the glaciated U.S. (260438)
Jon Jeffrey Starn, USGS Connecticut Water Science Center, East Hartford, CT, United States, Kenneth Belitz, USGS, Earth Systems Processes Division, Boston, United States, Leon Jay Kauffman, USGS New Jersey Water Science Center, West Trenton, United States and Carl Carlson, USGS New England Water Science Center - Massachusetts Office, Northborough, MA, United States
Full-Physics Inverse Learning Machine for Satellite Remote Sensing Retrievals (260629)
Diego G Loyola, German Aerospace Center DLR Oberpfaffenhofen, Oberpfaffenhofen, Germany

See more of: Atmospheric Sciences