Modeling Cloud Reflectance Fields using Conditional Generative Adversarial Networks

Thumbnail Image

Files

2002.07579.pdf (2.62 MB)

Links to Files

https://arxiv.org/abs/2002.07579

Permanent Link

http://hdl.handle.net/11603/17650

Collections

UMBC Joint Center for Earth Systems Technology (JCET)
UMBC Faculty Collection

Author/Creator

Author/Creator ORCID

Date

2020-02-10

Type of Work

conference papers and proceedings preprints
Text

Department

Program

Citation of Original Publication

Schmidt, Victor; Alghali, Mustafa; Sankaran, Kris; Bengio, Yoshua; Yuan, Tianle; Modeling Cloud Reflectance Fields using Conditional Generative Adversarial Networks; Atmospheric and Oceanic Physics (2020); https://arxiv.org/abs/2002.07579

Rights

This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)

Subjects

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.