Ξ -ML: a dimensional analysis-based machine learning parameterization of optical turbulence in the atmospheric surface layer
M. Pierzyna (TU Delft - Atmospheric Remote Sensing)
R Saathof (TU Delft - Space Systems Egineering)
Sukanta Basu (State University of New York at Albany)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Turbulent fluctuations of the atmospheric refraction index, so-called optical turbulence, can significantly distort propagating laser beams. Therefore, modeling the strength of these fluctuations (πΆ2π) is highly relevant for the successful development and deployment of future free-space optical communication links. In this Letter, we propose a physics-informed machine learning (ML) methodology, Ξ -ML, based on dimensional analysis and gradient boosting to estimate πΆ2π. Through a systematic feature importance analysis, we identify the normalized variance of potential temperature as the dominating feature for predicting πΆ2π. For statistical robustness, we train an ensemble of models which yields high performance on the out-of-sample data of R2β=β0.958βΒ±β0.001.