Skip to Content

Improving the Quasi-Biennial Oscillation (QBO) through Surrogate - Accelerated Parameter Optimization and Vertical Grid Modification

This project aims to improve the Energy Exascale Earth System Model (E3SM), a model development and simulation project that investigates energy science using code optimized for supercomputers. We use machine learning techniques in our research to improve model biases in E3SM due to the quasi-biennial oscillation (QBO), a regular variation of the winds high above the equator and cause changes in the tropical stratosphere.
Keywords convection, convective parameterization, E3SM, mesoscale convection, MMF, model bias, modeling, parameter optimization, quasi-biennial oscillation (QBO), structure and composition, surrogate models, tropical convection
TYPE Project
Principal Investigator (PI)
Jim Benedict
Lead Institution
Los Alamos National Laboratory
Other Collaborators
Lawrence Livermore National Laboratory Sandia National Laboratory
National Center for Atmospheric Research
For more information
Other research you might be interested in
Image of workflow and project architecture for the AI4ESP project. (Image credit: Lawrence Berkeley National Laboratory 2022). AI4ESP: Artificial Intelligence for Earth System Predictability
This high-resolution ocean simulation uses the Energy Exascale Earth System Model (E3SM), which divides the globe into a grid with intervals of only 15 kilometers. (Image credit: Los Alamos National Laboratory) E3SM: Energy Exascale Earth System Model
Drought in the Colorado River in Rocky Mountain National Park. (Image by Charuleka Varadharajan, Lawrence Berkeley National Laboratory) iNAIADS: iNtegration, Artificial Intelligence Analytical Data Services for Rivers and Watersheds