A new algorithm for planetary boundary layer height calculation based on multilayer recognition

Abstract

Aerosol lidar is widely used in the planetary boundary layer (PBL) height calculation due to its high spatiotemporal resolution. Most of the PBL height (PBLH) algorithms for aerosol lidar are valid for single aerosol layer structure, but overestimate the PBLH under multilayer aerosol/cloud structures. To fill the gap, a new algorithm of PBLH calculation based on multilayer recognition and idealized-profile (MR-IP) is developed. In this algorithm, residual layer and/or suspended aerosol/cloud layer are first recognized based on the signal to noise ratio (SNR) of lidar, the range squared correction signal (RCS) and its gradient (∇RCS). In residual and/or suspended aerosol/cloud layer, positive and negative ∇RCS exist simultaneously, while inside the PBL only a single negative ∇RCS exists. These characteristics are used to discern residual layer and/or suspended aerosol/cloud layers. Aerosol and cloud layers are further distinguished by the ratio of RCS in the objective layer (RCS(rS)) to the mean RCS in the background layer. After multilayer recognition, the PBLH is calculated based on idealized-profile (IP) method. A yearlong (642 samples) comparison of the PBLH calculated by lidar and radiosonde verified the new algorithm in PBLH calculation under complicated structures (R = 0.81). The mean PBLH calculated by the MR-IP method is 635.4 m, consistent with the PBLH defined by radiosonde (665.3 m). Furthermore, the residual layer, suspended aerosol layer and cloud layer can also be discerned by the new algorithm.