Identification And Its Applications Of Volcanic Ash Cloud For Satellite Remote Sensing Data

After the volcanic eruption,the volcanic ash cloud is often formed in the atmosphere,which can weaken the solar radiation reaching the ground,and cause significant changes in the local and even global climate system.At the same time,with the rapid development of global economic integration,the international air transport industry is increasingly prosperous,and the threat of volcanic ash cloud on aviation safety is also more and more prominent.Therefore,the volcanic ash cloud monitoring has become an important work both scientifically and practically.Satellite remote sensing technology is a comprehensive ground-observation technology which has developed rapidly since the 1960s.It can quickly and accurately obtain spatial dynamic change information of ground objects,and has been widely used in many areas such as global-change monitoring,natural-disaster monitoring,early warning and loss assessment.Especially in recent years,meteorological satellites and LIDAR have shown great superiority in the field of volcanic ash cloud monitoring,and have been widely valued by the academia and related business departments.They have become important supporting technologies in volcanic ash cloud monitoring studies.Based on the background mentioned above,this paper has carried out a series of exploratory studies on the identification and application of volcanic ash cloud for satellite remote sensing data.The research work is carried out for the early formative stage and the later diffusion stage of the volcanic ash cloud from the horizontal distribution and the vertical profile distribution structure of the volcanic ash cloud.Aiming at the two aspects,based on the analysis of principal component analysis(PCA),independent component analysis(ICA),support vector machine(SVM)and fuzzy C-means(FCM),this paper carries out some exploration of the PC A-IC A-weighted and SVM,the PCA and neighborhood-constrained FCM,and the vertical profile distribution structure detection of volcanic ash cloud.To be specific,the main contents of this paper are as follows:(1)Based on the analysis of the distribution characteristics of the early formative stage of volcanic ash cloud,by combining the statistical characteristics and linear relationship between PCA and ICA and the nonlinear classification of SVM,an identification method of the early formative stage of volcanic ash cloud combining PCA-ICA-weighted and SVM is proposed.The MODIS satellite remote sensing image of the Sangeang Api volcanic ash cloud on May 30,2014 is tried in the identification.This method not only overcomes the limitation of the traditional single method,and realizes the feature complement,but also improves the recognition effect of the volcanic ash cloud in the early formation stage.(2)Based on the analysis of the distribution characteristics of the latter diffusion stage of volcanic ash cloud,by combining the PCA linear transformation and the present situation of FCM fuzzy classification,an identification method of the latter diffusion stage of volcanic ash cloud combining PCA and neighborhood-constrained FCM is proposed.The MODIS satellite remote sensing image of the Puyehue volcanic ash on June 11,2011 is tried in the identification.This method not only overcomes the limitation of the traditional single FCM method,but also realizes the effective identification of the volcanic ash cloud in the latter diffusion stage.(3)Based on the analysis of the current situation that the spaceborne CALIOP LiDAR can obtain the atmospheric suspended aerosol particles and the vertical profile distribution structure of clouds by relying on the difference of the bi-band backscattering signal which is sensitive to polarized light,by combining the vertical distribution structure features of volcanic ash cloud,a detection method of using CALIOP LiDAR is proposed for the vertical profile distribution structure of the volcanic ash cloud,and the vertical profile distribution structure of different development stages of the Eyjafjallajokull volcanic ash cloud from April to May 2010 is discussed respectively.(4)Contraposing the typical Eyjafjallajokull volcanic ash cloud case,an identification method is proposed for the different development stages of the volcanic ash cloud from April to May in the horizontal direction,and detects its vertical profile distribution structure using the spacebome CALIOP LiDAR.Hopefully by attempting to give a horizontal and vertical joint monitoring of the typical volcanic ash cloud cases,some references could be provided for other volcanic ash cloud case studies.