Conditional generation of cloud fields

TL;DR

This paper presents a generative adversarial network that produces realistic cloud fields conditioned on meteorological data, aiming to improve climate model representations of clouds despite current limitations in detail accuracy.

Contribution

The study introduces a GAN-based method for generating cloud fields conditioned on reanalysis data, advancing the use of deep learning in climate science.

Findings

Successfully generates realistic large-scale cloud patterns

Captures key features of cloud fields conditioned on data

Identifies need for further refinement for finer details

Abstract

Processes related to cloud physics constitute the largest remaining scientific uncertainty in climate models and projections. This uncertainty stems from the coarse nature of current climate models and relatedly the lack of understanding of detailed physics. We train a generative adversarial network to generate realistic cloud fields conditioned on meterological reanalysis data for both climate model outputs as well as satellite imagery. While our network is able to generate realistic cloud fields, especially their large-scale patterns, more work is needed to refine its accuracy to resolve finer textural details of cloud masses to improve its predictions.