Clustering methods usage for the temperature stratification typing of the atmosphere surface layer

Nikolay A. Baranov

doi:10.1117/12.2690848

17 October 2023 Clustering methods usage for the temperature stratification typing of the atmosphere surface layer

Nikolay A. Baranov

Proceedings Volume 12780, 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics; 1278069 (2023) https://doi.org/10.1117/12.2690848
Event: XXIX International Symposium "Atmospheric and Ocean Optics, Atmospheric Physics", 2023, Moscow, Russian Federation

Abstract

This paper considers the use of various clustering methods to characterize the seasonal features of the temperature stratification of the atmosphere surface layer. As initial data, we used series of observations of temperature profiles in a layer up to 1 km, obtained by a passive microwave profiler MTP-5. Clustering methods with automatic determination of the number of clusters are considered: DBSCAN, Affinity Propagation, MeanShift. It is shown that the DBSCAN method is of little use for identifying typical conditions for continuous series of observations. Affinity Propagation and MeanShift methods give similar results, but require fine tuning of the clustering parameters. The results of the temperature profiles clustering are presented with the identification of cluster centers and their parameters: the average temperature profile gradient and extreme values of the profile curvature.

Citation Download Citation

Nikolay A. Baranov "Clustering methods usage for the temperature stratification typing of the atmosphere surface layer", Proc. SPIE 12780, 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics, 1278069 (17 October 2023); https://doi.org/10.1117/12.2690848

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