Automatic methods to detect the top of atmospheric boundary layer

The main objective of this work is to obtain methods that automatically allow qualitative detections of Atmospheric Boundary Layer heights from LIDAR data. Case studies will be used to describe the more relevant days of a campaign carried out in July of 2012 in Vitória, Espírito Santo, Brazil. The data analysis compares three mathematical algorithms that automatically provide the ABL height: Gradient Method (GM), using the derivative of the Range Corrected Signal (RCS) logarithm, WCT (Wavelet Covariance Transform), and Bulk Richardson's Number, which was used to validate the methods mentioned above. The comparison between the methods has shown that as the presence of clouds and the aerosol sublayer increased, the more sensitive was the refinement needed to choose the right parameters, whereas even Richardson's method had ambiguities in finding a good estimate of the ABL top. © 2013 SPIE.