基于连续MODIS真彩图的霾监测预警新方法——以浙江省一次严重霾污染过程为例

采用NASA MODIS提供的全球影像浏览服务技术(GIBS)支持,并借助开源地理空间数据抽象库(GDAL)工具,实现了每日2次(上午和下午)用户自定义区域MODIS 250m分辨率真彩色图产品从互联网的近实时自动获取, 并基于该产品对2014年11月9日~14日浙江省一次大范围霾严重污染事件提前2d成功实现了预警.结果表明:连续的MODIS真彩图产品大幅改善并提高了传统可见光遥感监测灰霾的效果;MODIS 250m真彩色图3通道数据值与地面站点颗粒物浓度之间具有较好的相关性,前者与AQI,PM2.5和PM10之间的相关系数(R)分别为0.62,0.65和0.59.基于连续的 MODIS真彩图产品,并结合地面站点数据和天气形式进行综合分析能够及时提供相关预警服务;在业务流程上简单,高效,稳定,尤其对浙江省外源输入性霾识别有着良好的指示作用,具有较好的应用前景.

Monitoring haze events from the MODIS True-Color composite image series: A case study of severe haze pollution event in Zhejiang Province

MODIS True-Color images downloaded from NASA Worldview website had been successfully used in near real-time global air pollution monitoring. In this paper, we introduced how to grab NASA MODIS 250m True-Color image automatically from the website by using GDAL utilities. Using multiple linear regression, we found a significant relationship between the R, G, B values in MODIS True-Color image and the surface-level particle matter concentrations (AQI, PM2.5 and PM10), with a multiple R value of 0.62, 0.65 and 0.59, respectively. By combining MODIS True-Color image series with stational observation and atmospheric circulation analysis, we established a new operational procedure for haze pollution monitoring and early warming. With such a simple technical method, a severe haze pollution event that occurred in Zhejiang Province from November 9 th to 14 th, 2014 was successfully predicted based on the information two days earlier. The results showed that haze monitoring and early warming based on the MODIS True-Color image series was relatively easy and efficient in operational survey, and the new method had great potential in regional haze pollution monitoring and early warming, especially as an indicator for pollution input from other regions.