基于MODIS数据的南昌市气溶胶光学厚度反演研究

发布时间：2018-06-22 06:29

本文选题：MODIS + 暗像元算法　；参考：《东华理工大学》2015年硕士论文

【摘要】：大气气溶胶是液态或者固态微粒在空气中形成的微悬浮物,它在地球大气辐射收支平衡和全球气候变化扮演着非常重要的角色,其最重要的参数之一是光学厚度,精确探测气溶胶光学厚度对于研究气候变化具有十分重要的意义,卫星遥感反演气溶胶光学厚度具有传统的地基探测不具有的优势。本文选取目前技术成熟、应用广泛的暗像元算法,利用MODIS遥感数据中的MOD021KM与MOD03角度数据,采用6S辐射传输模型建立查找表,选取2014年3月至9月间南昌市以晴天为主的MODIS遥感影像。利用ENVI软件处理该时间段内质量相对较高的一系列遥感影像数据,提取处理后的Band 1红波段(0.66um)、Band 3蓝波段(0.47um)与Band 7近红外波段(2.1um)的数据对大气气溶胶光学厚度(AOD)进行反演。根据本文所定义的查找表,预先预测本文所反演的气溶胶结果与PM10浓度值呈负相关。查找省站处的AOD值并与南昌市9个地面监测站获取的PM10浓度平均值对比,发现AOD值与PM10浓度结果类似呈正态分布与事先预期的负关系相悖。当AOD值在1.2时,PM10浓度达到最大值,AOD值在0-1.2区间,AOD随着与PM10浓度的增大而增大,与实际不符;AOD值在1.2-2区间,AOD随着与PM10浓度的增大而减小,与事实较为符合。可以得出利用MODIS遥感数据反演AOD不能直接应用于预报空气污染等级,但可以在宏观上指示空气污染的分布区域,为大气污染的预报与治理提供一定的参考作用。
[Abstract]:Atmospheric aerosols are microsuspended particles formed in the air by liquid or solid particles. They play a very important role in the balance of the earth's atmospheric radiation budget and global climate change, and one of its most important parameters is optical thickness. The accurate measurement of aerosol optical thickness is very important for the study of climate change. Satellite remote sensing inversion of aerosol optical thickness has the advantage of traditional ground detection. In this paper, we select the mature technology and widely used dark pixel algorithm, make use of MOD021KM and MOD03 angle data in MODIS remote sensing data, and use 6S radiative transfer model to establish the look-up table. From March to September 2014, MODIS remote sensing images of Nanchang with sunny weather were selected. A series of remote sensing image data of relatively high quality in this time period are processed by ENVI software, and the processed data of 0.66um band (0.66um), band3 blue band (0.47um) and near infrared band (2.1um) are extracted to retrieve the atmospheric aerosol optical thickness (AOD). According to the look-up table defined in this paper, it is predicted that the aerosol inversion in this paper is negatively correlated with PM10 concentration. The AOD values at provincial stations and compared with the average values of PM10 concentrations obtained from nine ground monitoring stations in Nanchang City were compared. It was found that the AOD values and PM10 concentration results were similar to those of normal distribution and the negative relationship between AOD values and the predicted values was contrary. When the AOD value is 1.2, the AOD value of PM10 increases with the increase of the concentration of PM10 in the range of 0-1.2, and decreases with the increase of the concentration of PM10 in the range of 1.2-2, which is more consistent with the fact that the AOD increases with the increase of the concentration of PM10. It can be concluded that AOD inversion using MODIS remote sensing data can not be directly used to predict air pollution level, but it can indicate the distribution of air pollution on the macro level, and provide a certain reference for air pollution prediction and control.
【学位授予单位】：东华理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X513;X87

【相似文献】