Joint Session 66 Machine Learning for Subgrid Parameterization in Weather and Climate Models

Thursday, 16 January 2020: 10:30 AM-12:00 PM
156BC (Boston Convention and Exhibition Center)

Cochairs:
Ryan A. Lagerquist, CIMMS, Meteorology, Norman, OK; Christiane Jablonowski, University of Michigan, Climate and Space Science and Enigneering, Ann Arbor, MI and Carlos F. Gaitan, Arable Labs, Inc., Machine Learning and Artificial Intelligence, Princeton, NJ

Description:

This session will feature presentations on using machine learning models for parameterizing subgrid processes in numerical weather and climate models.


Papers:
10:30 AM
J66.1
Building a Hierarchy of Hybrid, Neural-network Parametrizations of Convection
Tom Beucler, UCI, Irvine, CA; Columbia University, New York, CA; and P. Gentine, M. S. Pritchard, S. Rasp, and V. Eyring

10:45 AM
J66.2
Data-driven super-parameterization using deep learning: Experimentations with a multi-scale Lorenz 96 model
Pedram Hassanzadeh, Rice University, 6100 Main St., Houston, TX; and A. Chattopadhyay, A. Subel, and K. Palem

11:00 AM
J66.3
Machine Learning Parameterization of the Surface Layer: Integration with WRF
David John Gagne II, NCAR, Boulder, CO; and T. C. McCandless, B. Kosovic, A. DeCastro, R. D. Loft, S. E. Haupt, and B. Yang

11:15 AM
J66.4
Data-driven approaches for simulating rainfall in climate models
R. Saravanan, Texas A&M Univ., College Station, TX; and J. Yang, M. Jun, C. Schumacher, J. Wang, and R. K. W. Wang

11:30 AM
J66.5
Towards sub-grid scale parameterizations using a super-resolution generative adversarial network
Eden Au, University of Edinburgh, Edinburgh, United Kingdom; and K. Kashinath, A. Albert, M. Prabhat, and S. F. B. Tett

11:45 AM
J66.6
Utilizing Machine Learning to Replace Physical Parameterization Schemes: How do Different Techniques Compare?
Garrett Limon, University of Michigan, Ann Arbor, MI; and C. Jablonowski

Abstract Submissions Closed



See more of: 19th Conference on Artificial Intelligence for Environmental Science
<< Previous Session | Next Session >>