基于MODIS影像的济南市气溶胶反演研究与系统开发

发布时间：2018-04-04 01:21

本文选题：气溶胶光学厚度　切入点：改进暗像元法　出处：《山东农业大学》2017年硕士论文

【摘要】：气溶胶含量是关系空气质量的重要指标,对于全球能量平衡、陆地生态系统的良性循环有着重要的作用。大气气溶胶光学厚度(Aerosol Optical Thickness,AOT)是表征大气浑浊度的重要物理量,是气溶胶监测最重要的参数之一。本研究借助MODIS遥感影像与太阳光度计CE318实测数据,利用改进暗像元法、Flaash大气校正法与6S辐射传输模型,构建了自定义的济南市气溶胶模型。基于济南市气溶胶模型,制作精准AOT查找表,反演得到济南地区AOT;建立了济南地区AOT与PM2.5含量的数学模型;通过与检验数据对比分析,验证了模型的有效性;结合IDL语言,流程化反演过程,开发了气溶胶反演系统。主要研究内容及结论如下:(1)结合改进暗像元法与Flaash大气校正法,计算MODIS影像地表反射率。应用改进暗像元法计算研究区暗地表反射率,Flaash大气校正法计算研究区亮地表反射率,两者结合实现研究区MODIS影像地表反射率计算。比较AOT反演结果,改进暗像元法与Flaash大气校正结合法的反演结果较单一改进暗像元法反演结果的连续性更好,空值区域更少。(2)建立了济南市气溶胶模型与PM2.5反演模型,实现济南市AOT以及PM2.5反演。根据济南市的大气条件以及区域特征,假定气溶胶中四种组分的比重,通过太阳光度计观测数据与MODIS影像相结合,构建了自定义的济南市气溶胶模型。本文计算得到济南市气溶胶模型为沙尘性气溶胶比重40%,水溶性气溶胶比重56%,海洋性气溶胶比重1%,煤烟性气溶胶比重3%。分别基于自定义气溶胶模型、大陆型气溶胶模型、城市型气溶胶模型制作AOT查找表进行AOT的反演,将反演结果与MOD04气溶胶产品以及太阳光度计实测数据对比分析,结果表明:自定义气溶胶模型的AOT反演精度明显高于大陆型和城市型气溶胶模型的反演精度,自定义气溶胶模型对济南市AOT的精确反演具有巨大潜力。建立了济南市AOT与PM2.5含量的数学模型xy(10)(28)008.0049.0,2R为0.424。验证PM2.5反演模型的反演结果与检验数据的相关性,相关系数为0.767。(3)开发气溶胶反演系统,流程化反演过程。IDL编写AOT、PM2.5反演程序,完成气溶胶反演系统的开发,提高济南市AOT与PM2.5的反演效率。
[Abstract]:Aerosol content is an important index related to air quality and plays an important role in global energy balance and the benign cycle of terrestrial ecosystem.Atmospheric aerosol optical thickness (AOT) is one of the most important parameters of aerosol monitoring. It is an important physical quantity to characterize atmospheric turbidity.In this study, a self-defined aerosol model in Jinan was constructed by using the MODIS remote sensing image and the CE318 measured data of the solar photometer, using the improved dark pixel method, Flaash atmospheric correction method and the 6s radiation transfer model.Based on Jinan aerosol model, the accurate AOT lookup table is made to retrieve the content of AOT and PM2.5 in Jinan area. The mathematical model of AOT and PM2.5 content in Jinan area is established. The validity of the model is verified by comparing with the test data.The aerosol inversion system is developed based on the flow inversion process.The main contents and conclusions are as follows: (1) combined with the improved dark pixel method and Flaash atmospheric correction method, the surface reflectivity of MODIS image is calculated.The improved dark pixel method is used to calculate the dark surface reflectance of the study area and the Flaash atmospheric correction method is used to calculate the bright surface reflectance of the study area. The calculation of the surface reflectance of the MODIS image in the study area is realized by the combination of the two methods.Compared with the AOT inversion results, the inversion results of the improved dark pixel method and the Flaash atmospheric correction method are more continuous than those of the single improved dark pixel method, and the empty value region is less. The aerosol model and the PM2.5 inversion model of Jinan City are established.The inversion of AOT and PM2.5 in Jinan is realized.According to the atmospheric conditions and regional characteristics of Jinan, a self-defined aerosol model of Jinan was constructed by combining the observation data of solar photometer with the MODIS image, assuming the specific gravity of the four components in the aerosol.In this paper, the aerosol model of Jinan is calculated as dust aerosol weight 40, water soluble aerosol specific gravity 56, oceanic aerosol weight 1, soot aerosol specific gravity 3.Based on the customized aerosol model, the continental aerosol model and the urban aerosol model, the AOT lookup table is made for the inversion of AOT. The inversion results are compared with the measured data of MOD04 aerosol products and solar photometers.The results show that the AOT inversion accuracy of the self-defined aerosol model is obviously higher than that of the continental aerosol model and the urban aerosol model. The self-defined aerosol model has great potential for accurate inversion of AOT in Jinan.A mathematical model for the content of AOT and PM2.5 in Jinan was established.To verify the correlation between the inversion results of the PM2.5 inversion model and the test data, the correlation coefficient is 0.767.03) the aerosol inversion system is developed.The inversion efficiency of AOT and PM2.5 in Jinan is improved.
【学位授予单位】：山东农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X87;X513

【参考文献】