Monitoring Dust Storm By A New Generation Of Geostationary Meteorological Satellite

Using geostationary meteorological satellite to identify the dust at night has always been a difficult problem in dust remote sensing monitoring.Due to the lack of infrared channels,such as FY-2,MTSAT-2 and other previous generation geostationary meteorological satellites,it is difficult to monitor dust at night.At present,the new generation of geostationary meteorological satellites FY-4 and Himawari-8 have been launched.Compared with the previous generation of satellites MTSAT-2,the temporal and spatial resolution of the new generation of satellites has been significantly improved.More importantly,the number of infrared channels has increased from four to eight/ten,and the large increase of infrared channels has made it possible for allweather identification of sand and dust.This study makes full use of the infrared data of FY-4 and Himawari-8,and realize the all-weather identification of dust only by infrared channels.Firstly,this research analyzes the physical principle of dust monitoring.including dust's radiation characteristics,the influence of satellite spectral channel design on dust identification,and then sensitivity analysis is carried out.MODTRAN is used to simulate the influence of dust concentration and surface temperature on dust identification parameters.In addition to the analysis of sand and dust,this paper compares the spectral differences between dust and cloud/clear surface based on satellite data.Based on these analyses,the following conclusions can be drawn: the design of satellite spectral channels will affect the identification of dust.If the 12-micron channel of the satellite instrument is too wide to cover the wavelength with strong water vapor absorption,the parameter BT11-BT12 will be insensitive to dust.Himawari-8 has such a problem,so the two parameters can complement each other to a certain extent by combining BT11-BT12 and BT10-BT11 to identify sand and dust.In addition,when analyzing the spectral characteristics,it is found that there are some cloud areas and dust areas with similar brightness temperature difference.Therefore,before identifying dust,it is necessary to filter the clouds in order to avoid misjudging the cloud area as dust area.For the new generation of geostationary satellites,the increase of water vapor channels has more advantages in distinguishing dust from clouds.On the basis of the above-mentioned principle analysis,a set of all-weather sand and dust identification algorithm is constructed.The construction process of the algorithm is described in detail,including the selection of identification parameters,the definition of daytime and night,the determination of sand and dust identification threshold,etc.Finally,three typical sand-dust events are selected to verify the sand-dust identification results obtained by this method.In the analysis of ground verification,it is found that the data of night ground meteorological stations are not enough to verify the results of satellite identification,so a set of ground verification data is constructed jointly with ground meteorological stations and PM10.Then the identification results are compared and checked.Firstly,combining with a Sand-Dust case,the results of this paper are compared with FY-2 sand-dust products.It is found that the results of this paper match the foundation verification data better and the identification accuracy is better.In addition,combined with three typical events,the results of this paper are compared with the ground validation data.The results show that both the satellite identification results and the ground observation data have good consistency in day and night.This shows the feasibility of allweather dust monitoring by geostationary meteorological satellite on the one hand,and the universality of the algorithm on the other hand.