Machine Learning Emulation of 3D Cloud Radiative Effects

Abstract

The treatment of cloud structure in numerical weather and climate models is often greatly simplified to make them computationally affordable. Here we propose to correct the European Centre for Medium‐Range Weather Forecasts 1D radiation scheme ecRad for 3D cloud effects using computationally cheap neural networks. 3D cloud effects are learned as the difference between ecRad's fast 1D Tripleclouds solver that neglects them and its 3D SPARTACUS (SPeedy Algorithm for Radiative TrAnsfer through CloUd Sides) solver that includes them but is about five times more computationally expensive. With typical errors between 20% and 30% of the 3D signal, neural networks improve Tripleclouds' accuracy for about 1% increase in runtime. Thus, rather than emulating the whole of SPARTACUS, we keep Tripleclouds unchanged for cloud‐free parts of the atmosphere and 3D‐correct it elsewhere. The focus on the comparably small 3D correction instead of the entire signal allows us to improve predictions significantly if we assume a similar signal‐to‐noise ratio for both.

A posteriori learning for quasi-geostrophic turbulence parametrization

“…Calibrating physics-based parametrization schemes against observation with ML and emulators is for instance becoming common practice (Ollinaho et al, 2013;Schneider, Lan, et al, 2017;Couvreux et al, 2021). Emulation approaches based on ML have also been proposed as a strat-egy for accelerating or regularizing existing schemes (Ukkonen et al, 2020;Meyer et al, 2021). ML also provides new means to design new subgrid parametrization schemes from high fidelity simulations.…”

Compound Parameterization to Improve the Accuracy of Radiation Emulator in a Numerical Weather Prediction Model

“…Recent advances in machine‐learning techniques have provided new opportunities to significantly accelerate the computation speed of numerical weather prediction (NWP) models. Among the various fields of numerical models, radiation physics for longwave (LW) and shortwave (SW) accounts for the most significant computational burden in models and has the oldest history of developing machine‐learning emulators in radiative transfer modeling (Chevallier et al., 1998; Liu et al., 2020; Meyer et al., 2021; Ukkonen et al., 2020; Veerman et al., 2021), data assimilation (Chevallier et al., 2000), and radiation parameterization in climate simulation models (Belochitski et al., 2011; Krasnopolsky et al., 2005, 2010; Pal et al., 2019) and NWP models (Roh & Song, 2020; Song & Roh, 2021). In previous studies, acceleration of two orders was achieved by replacing the radiation parameterization with a neural network (NN) emulator for numerical models.…”

Benefits of Stochastic Weight Averaging in Developing Neural Network Radiation Scheme for Numerical Weather Prediction

“…Note that the results of Liu et al (2020) should be interpreted differently because the measurements described were obtained under different parallelization conditions. Meanwhile, Meyer et al (2022) showed that using an emulator to add 3D cloud radiative effects was less than 1% more expensive than the 1D scheme; this was a significant decrease in computational cost because the 3D scheme was usually five-times as expensive than the 1D scheme. These results demonstrate the effectiveness of emulating cloud processes in terms of computational cost.…”