不同时间尺度大气PM_(2.5)和PM_(10)卫星影像反演  

Air particulate matter PM_(10) and PM_(2.5) satellite imageries inversion under different time scales

作　　者：武鹏飞[1] 赵银军[1] 卢远[1] 吴海燕[1] 金健[1] 

机构地区：[1]广西师范学院地理科学与规划学院,广西南宁530001

出　　处：《广西师范学院学报（自然科学版）》2016年第3期80-88,95,共10页Journal of Guangxi Teachers Education University(Natural Science Edition)

基　　金：广西科学研究与技术开发项目(1598017-13);国家自然科学基金资助项目(41461021);北部湾环境演变与资源利用教育部重点实验室系统基金(2014BGERLXT15)

摘　　要：利用南宁市地面8个监测站与中分辨率成像光谱仪(MODIS)数据反演得到的气溶胶光学厚度值作为数据源,运用回归分析法,选取月、季、年三种时间尺度,分别对PM_(2.5)、PM_(10)浓度与AOD值进行相关性研究。结果表明,PM_(2.5)与AOD相关性好于PM_(10),月尺度PM_(2.5)和PM_(10)与AOD值相关性强,除个别月份外,R2均在0.7以上;季尺度PM_(2.5)和PM_(10)与AOD值相关性,随季节变化显著,但R2均在0.5以上;年尺度PM_(2.5)和PM_(10)与AOD值拟合,采用一元二次模型,R2在0.5以上。上述结果表明AOD在月尺度上与地面站点污染物监测数据PM_(2.5)和PM_(10)的相关性最为显著,故可在月尺度上通过卫星遥感影像反演的AOD推算地面PM_(2.5)和PM_(10)的空间浓度场。By using eight ground atmospheric monitoring stations in Nanning and moderate reso- lution Imaging Spectroradiometer （MODIS） data inversion we get the Aerosol Optical Depth as a data source. Using the regressive analysis, selecting Annual, Seasonal and Monthly different time scales we study the correlation between PM2.5 and PM 10 concentrations and AOD. The results show that the correlation of PM2.5 and AOD is better than that of PM 10 ; in Monthly scale PM2.5 and PM10 and AOD have a stronger relevance. In addition to the individual months, R2 is above 0.7. In Seasonal scale, the correlation between PM2.5 and PMx0 and AOD with the seasonal changes is obvious, and the R2 is above 0.5. The correlation between PM2.5 and PM10 and AOD in Yearly scale is fitted. By using the Unary Quadratic Model, the R2 is above 0.5. The results indicate that the correlation between ground stations monitor data of pollutants PM2.5 and PM10 and AOD is the most significant in Monthly scale. So we can, through the Satellite Imagery inversion AOD in Monthly scale, predict the PM2.5 and PM10 concentration field in ground space.

关 键 词：MODIS 气溶胶光学厚度 回归分析法 时间尺度 相关性 

分 类 号：X513[环境科学与工程—环境工程]