Automated Algorithm For Cloud Parameter Retrieval Using Lidar Data And Its Applied Research

The temporal-spatial distributions and microphysics structure of clouds determinetheir optical characteristics,which has a great significance for the research on globalclimate change, weather forecast, aviation security, weather modification and otherareas. Along with the rapid development of lidar techniques in recent years, lidar hasincreasingly become an effective tool for continuous observation of atmosphericphysical parameter and the spatial distribution of aerosol and cloud.The aim of the present paper is to achieve a complete automated algorithm appliedto accurately determine the cloud-base, cloud-peak and cloud-top heights (CBH, CPH,CTH) in real time. Firstly, several instruments used the observation of clouds,especially for cloud-base height observation are shortly introduced, with analysis ofthe detection principle and the technical problems existed in radiosonde, all sky IRscanning imager and cloud radar respectively. Secondly focus of the developmentsituation of Mie scattering lidar in domestic and abroad, several ceilometers withhigher degree of automation in the international are studied, including the summaryand comparison of main parameters. Then the paper summarizes laser-atmosphereinteraction, the principle of lidar and the characters of lidar profiles, etc.In the main part of the paper, the detinition of cloud parameter and typicalcharacters of lidar backscattering signal from cloud are firstly described, and then itanalyzes several published cloud property retrieval algorithms and discusses theirscope of application and errors. Secondly, according to the the property of lidarprofiles, this paper promotes an improved algorithms based on differentialzero-crossing method, which can differentiate various targets such as cloud, aerosols,or random noise through by introducing various threshold in the data processing.Thirdly, a large body of lidar data that were obtained from June to July,2009inZhuhai have been evaluated, focusing on CBH, the results of the synchronousintercomparison show that the correlation coefficients (CC) of CBH and CTH measured by the mobile Doppler lidar and cloud radar are0.98and0.95, respectively;The lidar data obtained from March to April2011and April to May2013in Beijingare used to analyze cloud vertical structure such as the occurrence, layers, cloud etage,mean CBH, mean CTH, mean thickness, etc. The CC of CBH and CTH measured bylidar and radiosonde are0.86and0.83, respectively. The results agree well in mostcases and the possible causes of difference between measurements were brieflyanalyzed; The continuous dynamic observation data obtained by meteorologic lidarand Vaisala CL31celiometer in Qingdao during September and December2013areused to do a further test for the algorithm, the CC of CBH are0.81,which will rise to0.86by eliminating some differences produced because of cloud changes, and then apreliminary study about the cloud structure over this area is carried out. Finally, thepaper points out the improvements of the improved algorithm and provides an outlookin application development.