Machine Learning Emulation of 3D Cloud Radiative Effects

TL;DR

This paper introduces a neural network-based correction method for 3D cloud radiative effects in weather models, significantly improving accuracy with minimal additional computational cost.

Contribution

It presents a novel neural network approach to efficiently emulate 3D cloud effects, enhancing existing 1D radiation schemes in weather models.

Findings

Neural networks reduce errors by 20-30% of the 3D signal.

The correction increases computational cost by only about 1%.

The method improves model accuracy without replacing the entire radiation scheme.

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.