Report on optimization of morphological image processing approach algorithm results on lidars and ceilometers observations

This deliverable provides a report on the optimization of MIPA (Morphological Image Processing Approach) algorithm1 for the retrieval of Atmospheric Boundary Layer Height (ABLH) from both High Power Lidar (HPL) and Low Power Lidar (LPL - ceilometer) data. MIPA has been developed at CNR-IMAA in the framework of ACTRIS and it has been evaluated on several HPL observations showing, in general, quite good performance with respect to the traditional ABLH retrieval techniques. The main novelty of MIPA consists in the application of suitable morphological operators on lidar range-resolved timeseries “viewed” as a single 2D image. The usage of suitable structuring elements shape guarantees full exploitation of lidar data for the retrieval of accurate ABLH estimates. Consequently, the ABLHs retrieved by MIPA are quite robust because they rely not only on the vertical dynamic of single lidar profiles, but also on the correlation among contiguous (in time) lidar observations. Additionally, MIPA works mainly on vertical and time correlations of the input dataset, and, consequently, the retrieved ABLHs do not strongly depend on input absolute intensities but on the corresponding relative variations. This feature opens the possibility to extend the applicability of MIPA to many lidar systems (including ceilometers) with different characteristics in terms of Signal-to-Noise Ratio (SNR), detection and acquisition type.