Modeling Cloud Reflectance Fields using Conditional Generative Adversarial Networks

Abstract

We introduce a conditional Generative Adversarial Network (cGAN) approach to generate cloud reflectance fields (CRFs) conditioned on large scale meteorological variables such as sea surface temperature and relative humidity. We show that our trained model can generate realistic CRFs from the corresponding meteorological observations, which represents a step towards a data-driven framework for stochastic cloud parameterization.

Code is available on Github: https://github.com/krisrs1128/clouds_dist