Abstract: The atmospheric boundary layer height (BLH) is one of the most strategic parameter to create meteorological estimations, including air quality forecasts. Several algorithms have been proposed to automatically estimate BLH from lidar backscatter profiles. This paper describes and compares two machine-learning methods, the K-means unsupervised algorithm and the Support Vector Machine (SVN) supervised algorithm, to derive BLH from lidar backscatter profiles. The MPLNET federated network L1 data was used in order to reproduce same lidar equipment in several locations around the planet. MPLNET data has mixed-layer-top information to be compared with machine-learning root- mean-square-error results (RMSE).

Keywords: micropulse lidar, machine learning, aerosols, clouds, boundary layer