Machine Learning Emulation of Gravity Wave Drag in Numerical Weather Forecasting. (8th July 2021)

• Record Type: Journal Article
• Authors: Chantry, Matthew; Hatfield, Sam; Dueben, Peter; Polichtchouk, Inna; Palmer, Tim
• Is Part Of: Journal of advances in modeling earth systems. Volume 13:Number 7(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-07-08
• Subjects: machine learning -- numerical weather prediction; Geological modeling -- Periodicals; Climatology -- Periodicals; Geochemical modeling -- Periodicals; 551.5011
• DOI: 10.1029/2021MS002477
• Languages: English
• ISSNs: 1942-2466

Abstract

Abstract: We assess the value of machine learning as an accelerator for the parameterization schemes of operational weather forecasting systems, specifically the parameterization of nonorographic gravity wave drag. Emulators of this scheme can be trained to produce stable and accurate results up to seasonal forecasting timescales. Generally, networks that are more complex produce emulators that are more accurate. By training on an increased complexity version of the existing parameterization scheme, we build emulators that produce more accurate forecasts. For medium range forecasting, we have found evidence that our emulators are more accurate than the version of the parametrization 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 10 times faster than the existing scheme on a CPU. Plain Language Summary: The ability of computers to construct models from data (machine learning) has had significant impacts on many areas of science. Here, we use this ability to construct a model of an element of a numerical weather forecasting system. This element captures one physical process in the model, a part of the model that describes the propagation of large‐scale waves through the atmosphere, but the long‐term aim would be to make many models each capturing a process. The goal is that the computer‐generated model will perform the task more efficiently than the existing model. Testing is then carried out to ensure that our computer model performs as accurately as the existing model. This is a challenging step, as learning is carried out over short time periods (seconds), but forecasts need to be accurate over years. Our computer‐generated models produce accurate forecasts on all tested timescales. On current computers, they are not faster, but will be if weather forecasting centers invest in computers with graphics processing units. Key Points: Nonorographic gravity wave drag parametrization can be accurately emulated with a neural network These emulators produce accurate and stable forecasts over long timescales Neural networks can reduce the cost of an increased complexity scheme