Data driven parameterization of unresolved gravity wave momentum transport: A 1-D Quasi Biennial Oscillation model testbed
My first in-person seminar since April 2019!
Gravity waves, or buoyancy waves, so named because their restoring force is the action of gravity on a stratified fluid, present a challenge to atmospheric modeling. They play an important role in the global atmospheric circulation by transporting momentum, but their horizontal scales of 10^2 to 10^5 meters cannot be fully resolved even by the highest resolution climate prediction systems. Furthermore, many of their sources, including convection and frontogenesis, are themselves not directly represented. Their impacts must therefore be approximated, or parameterized, based on resolved flow variables. The availability of new observations has raised hopes for a data driven approach to the representation of gravity waves.
In this talk, I’ll present encouraging evidence that a variety of machine learning techniques may be able to successfully capture gravity wave momentum transport, and then turn to the more challenging question of adapting, or calibrating, a data driven parameterization to work with imperfect models that partially captures gravity wave effects. We focus on the ability of parameterizations to capture macroscopic effects in the circulation, the Quasi-Biennial Oscillation (QBO), a 28 month oscillation of jets in the tropical stratosphere that depends critically on gravity wave forcing. I’ll show that a number of machine learning approaches, including neural networks, regression trees, and support vector machines, can successfully emulate an existing, physics based parameterization in a global climate model. Most critically, we find that schemes trained on limited data sets have sufficient ability to learn the physics of the conventional parameterization to successfully emulate out-of-sample conditions when coupled online with the climate model. I’ll then present a simpler, 1-D model of the QBO to address the harder question of calibrating a data driven scheme to work in models with biases in numerics, forcing conditions, and/or partially resolved gravity waves.