基于Landsat TM/ETM+/OLI遥感大数据分析南京市1984—2015年大气能见度变化趋势

对南京市1984—2015年Landsat 4／5／7／8卫星TM／ETM＋／OLI传感器获取的遥感数据，利用ENVI遥感软件的FLAASH大气校正模块，进行了区域大气能见度（ VIS）遥感反演。结果表明，时间跨度达30余年的Landsat卫星遥感数据影像序列反演的VIS呈明显的下降趋势，20世纪80年代数值较高，“差”能见度（＜10 km）的观测率不到6％，21世纪以来VIS下降明显，“差”能见度的观测率为20％～25％。与2010—2015年南京市PM10、PM2．5监测数据进行了对比，在城市空气清洁及污染较轻时，星地监测结果有较好的一致性，但中到重污染天气时FLAASH算法反演VIS偏高，侧重于代表离主城区距离远的偏远乡野山林地区的能见度状况。

Trends of Nanjing Atmospheric Visibilities from 1984 to 2015 Based on Landsat TM/ETM+/OLI Remote Sensing Big Data

Landsat 4/5/7/8 remote sensing data for the Nanjing area spanning the years from 1984 to 2015 were obtained using TM/ETM+/OLI sensors and processed with FLAASH atmospheric correction module of ENVI remote sensing software, from which the remote sensing retrieval of regional atmospheric visibilities ( VIS) were extracted.The results showed that the VIS retrieved from Landsat image series spanning more than 30 years showed evident declining trends.VIS in the 1980s were relatively higher, with poor VIS (<10 km) observed at a 6%frequency.VIS were obviously decreased since the beginning of this century, with poor VIS occurred at a frequency of about 20%~25%based on satellite remote sensing.The VIS data were compared with PM10 and PM2.5 monitoring data in Nanjing from 2010 to 2015, and it was found that these two data sets showed reasonable consistency when air quality was good or only slightly polluted, but in the cases of moderate to heavy air pollution, VIS retrieved by FLAASH were higher than the ground monitoring values.This might indicate VIS tends to represent local visibilities in the rural or forest areas which were far away from urban area.