Machine learning emulation of gravity wave drag in numerical weather forecasting

TL;DR

This paper explores how machine learning emulators can improve the accuracy and speed of gravity wave drag parameterisation in weather forecasting, showing promising results for operational and GPU-based implementations.

Contribution

It demonstrates that machine learning emulators can produce more accurate gravity wave drag forecasts and run significantly faster on GPU hardware compared to traditional schemes.

Findings

Emulators are stable and accurate up to seasonal timescales.

More complex networks yield more accurate emulators.

On GPU hardware, emulators are ten times faster than CPU-based schemes.

Abstract

We assess the value of machine learning as an accelerator for the parameterisation schemes of operational weather forecasting systems, specifically the parameterisation of non-orographic gravity wave drag. Emulators of this scheme can be trained to produce stable and accurate results up to seasonal forecasting timescales. Generally, more complex networks produce more accurate emulators. By training on an increased complexity version of the existing parameterisation scheme we build emulators that produce more accurate forecasts. {For medium range forecasting we find evidence our emulators are more accurate} than the version of the parametrisation scheme that is used for operational predictions. Using the current operational CPU hardware our emulators have a similar computational cost to the existing scheme, but are heavily limited by data movement. On GPU hardware our emulators perform ten times faster than the existing scheme on a CPU.