大气气溶胶光学厚度反演及其在雾霾检测中的应用

发布时间：2018-03-15 10:36

本文选题：气溶胶　切入点：AOT　出处：《杭州电子科技大学》2015年硕士论文　论文类型：学位论文

【摘要】：伴随着工业化和城市化的迅猛发展，大气中的颗粒物含量增多，我国广大地区遭遇严重的大气污染，尤其是大范围的雾霾天气在全国范围内频发。雾霾天气是一种重要的城市气象灾害，其对交通运输、工农业生产以及人体健康都会带来多重不利的影响。本文以杭州地区为例，基于MODIS数据、AERONET观测站数据、国家气象局以及杭州气象站提供的地面气象数据，对气溶胶的光学厚度（AOT）反演和雾霾检测方法进行了研究，所做的主要研究工作及取得的成果如下： 1、分析了大气AOT遥感反演的基本原理，建立了反演的数学模型，在研究6S大气传输方程的基础上提出了一种自定义气溶胶类型的确定方法，分别使用此自定义的气溶胶类型和6S方程中默认的城市型气溶胶类型对杭州地区的大气AOT进行了反演，利用太阳光度计的观测值作为依据对反演结果进行了检验，检验结果表明反演值与观测值的相对误差均在20%以内，采用自定义气溶胶类型时的反演值相对误差比采用默认气溶胶类型时的反演值相对误差要低3%以上；反演结果说明杭州AOT值整体略偏大，考虑到气溶胶吸湿性增大的特点，该地区气候湿润，与浙江气象站提供的杭州天气状况较为吻合。 2、研究大气雾霾检测的基本原理和等级划分标准，明确了雾霾检测的两个关键指标能见度和相对湿度，使用AOT数据对能见度建立了估算模型，根据计算结果对模型的表达式进行了修正，对模型的计算精度进行检验得出结论：除去个别点外，能见度的估算值与真实值的相对误差最大不超过12%。以MOD05大气数据和MOD07温度数据作为数据源建立了相对湿度的遥感估算模型，重点提出了大气可降水量取值的两种加权方法，对模型进行检验，结果显示真实值与估算值的相对误差最小2.1%。基于能见度和相对湿度两个指标对2012年杭州地区的雾霾天进行了检测，使用2012年杭州气象站提供的全年雾霾数据作为标准值对检测的结果进行了检验，结果表明对清洁天和雾天的检测准确度要高于霾天的检测准确度，对轻微、轻度霾天的检测准确度要高于中度、重度霾天的检测准确度，，对春、夏季的检测准确度要高于秋、冬季的检测准确度，文章最后对雾霾检测中可能出现的影响因子和检测误差进行了分析。
[Abstract]:With the rapid development of industrialization and urbanization, the content of particulate matter in the atmosphere increases, and the vast areas of our country suffer from serious air pollution. Haze weather is an important urban meteorological disaster, which has many adverse effects on transportation, industry and agriculture production and human health. This paper takes Hangzhou as an example. Based on the MODIS data and the surface meteorological data provided by the National Meteorological Administration and Hangzhou Meteorological Station, the methods of aerosol optical thickness inversion and haze detection are studied. The main research work and the results obtained are as follows:. 1. The basic principle of atmospheric AOT remote sensing inversion is analyzed, and the mathematical model of inversion is established. Based on the study of 6s atmospheric transport equation, a method for determining the type of aerosol is proposed. The atmospheric AOT in Hangzhou area is inversed by using the self-defined aerosol type and the default urban aerosol type in the 6S equation, respectively. The inversion results are verified by using the observed values of the solar photometer as the basis. The test results show that the relative error between the inversion value and the observed value is within 20%, and the relative error of the inversion value using the self-defined aerosol type is more than 3% lower than that with the default aerosol type. The inversion results show that the AOT value of Hangzhou is slightly larger as a whole, considering the characteristics of the increase of aerosol moisture absorption, the climate in this area is humid, which is in good agreement with the weather condition of Hangzhou provided by Zhejiang Meteorological Station. 2. The basic principle and classification standard of atmospheric haze detection are studied, and the visibility and relative humidity of two key indexes of haze detection are defined. The visibility estimation model is established by using AOT data. According to the result of calculation, the expression of the model is modified, and the accuracy of the model is tested. The conclusion is: except for a few points, The relative error between the estimated value of visibility and the real value is not more than 12. Using MOD05 atmospheric data and MOD07 temperature data as data sources, a remote sensing estimation model of relative humidity is established, and two weighting methods for determining atmospheric precipitable water are put forward. The results show that the relative error between the real value and the estimated value is the least 2.1. Based on the two indexes of visibility and relative humidity, the haze days in Hangzhou area on 2012 are tested. The annual haze data provided by Hangzhou Meteorological Station on 2012 were used as standard values to test the results. The results showed that the detection accuracy of clean and smog days was higher than that of haze days, and it was slight. The detection accuracy of mild haze days is higher than that of moderate and severe haze days. The detection accuracy in spring and summer is higher than that in autumn and winter. Finally, the influence factors and detection errors in haze detection are analyzed.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X513;X87

【参考文献】