Application Of CALIOP In Characteristics Of Aerosol On Sea And Statistical Forecast Of Sea Fog

The Sea area under the jurisdiction of China is vast,with a large number of aerosol particles at sea,and as a common weather phenomenon that greatly reduces the visibility at sea due to a large number of aerosols,sea fog greatly affects the safety of human activities at sea.It is difficult to obtain the distribution characteristics of marine aerosols over a large space-time scale by conventional marine monitoring methods such as coastal fixed-point observations and ship navigation observations.Therefore,this study uses cloud and aerosol polarization lidar detectors that can achieve high vertical resolution observations to launch an analysis of aerosol characteristics in the three seas of China,and applied the CALIOP data to the related research work on sea fog monitoring and the construction of a statistical forecast model of sea fog.The dust frequency is gradually decreasing from north to south and there is a seasonal difference in the Bohai Sea and the East China Sea(highest in spring,followed by winter,and lowest in summer);there are zonal and seasonal differences in particle morphological characteristics in the three seas: from north to south,the aerosol color ratio(CR)in the sea area is gradually increasing,and the depolarization ratio(DR)is gradually decreasing;the proportion of spherical particles in the three sea areas in the rainy season is abnormally increased,indicating that relative humidity is an important factor affecting the morphology of aerosol particles in sea region of China;The zonal difference is relatively obvious,and it gradually decreases from north to south in each season(except in the East China Sea and South China Sea in spring).From the perspective of Aerosol Optical Depth(AOD),the relatively high value of AOD in the three seas are generally consistent with the climatic statistical characteristics of China 's sea fog;CALIOP Level 1B and vertical feature layer classification data products are jointly applied to two sea fog recognition algorithms,and a double threshold method based on sea surface misjudgment is found.The climate statistical characteristics of China 's sea fog occurrence are more consistent in time and space distribution,and the algorithm is applied to the training sample extraction of sea fog statistical prediction model based on Support Vector Machine(SVM).ERA5 reanalysis data are used to perform spatial-temporal matching and obtains predictive factors through feature selection and then constructs a statistical forecasting model.The forecast model performed well in the training verification and release phases.The recall rate of the three sea area forecast models was above 70%,the ETS score reached 0.55,and the HSS score was also above 0.6,indicating that the sea fog forecast model based on CALIOP data It has certain practical application value.