Π-ML: a dimensional analysis-based machine learning parameterization of optical turbulence in the atmospheric surface layer

Turbulent fluctuations of the atmospheric refraction index, so-called optical turbulence, can significantly distort propagating laser beams. Therefore, modeling the strength of these fluctuations (C_n² ) 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, ΠML, based on dimensional analysis and gradient boosting to estimate C_n². Through a systematic feature importance analysis, we identify the normalized variance of potential temperature as the dominating feature for predicting C_n². For statistical robustness, we train an ensemble of models which yields high performance on the out-of-sample data of R² = 0.958 ± 0.001.