Neural network retrieval of optical thickness of horizontally inhomogeneous cloudiness from nadir multispectral radiance data

Tatiana V. Russkova; Alexei V. Skorohodov; Ilya V. Tkachev

doi:10.1117/12.2690709

17 October 2023 Neural network retrieval of optical thickness of horizontally inhomogeneous cloudiness from nadir multispectral radiance data

Tatiana V. Russkova, Alexei V. Skorohodov, Ilya V. Tkachev

Author Affiliations +

Proceedings Volume 12780, 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics; 1278031 (2023) https://doi.org/10.1117/12.2690709
Event: XXIX International Symposium "Atmospheric and Ocean Optics, Atmospheric Physics", 2023, Moscow, Russian Federation

Abstract

The currently accepted algorithm for retrieving the optical parameters of clouds from satellite-based radiance measurement data implies the application of independent pixel approximation and a model of uniform cloudiness. This single-pixel approach cannot capture the 3D effects manifest themselves across multiple satellite pixels. At the same time, three-dimensional radiative transfer effects are the major source of retrieval errors in the passive remote sensing of clouds. The paper describes an alternative approach that extends the capabilities of traditional methods to obtain the estimates of cloud parameters. It is based on the building artificial neural network trained with the simulation data of solar radiative transfer in the cloudy atmosphere. The possibility of retrieving the optical thickness of broken horizontally inhomogeneous warm clouds is demonstrated. High values of the correlation coefficient between the benchmark and retrieved values of the optical depth were achieved at a fixed effective radius of cloud drops. The dependence of the correlation coefficient on the hyperparameters of the neural network, the volume and structure of the training dataset is studied.

Citation Download Citation

Tatiana V. Russkova, Alexei V. Skorohodov, and Ilya V. Tkachev "Neural network retrieval of optical thickness of horizontally inhomogeneous cloudiness from nadir multispectral radiance data", Proc. SPIE 12780, 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics, 1278031 (17 October 2023); https://doi.org/10.1117/12.2690709

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available