基于多源卫星遥感数据的气溶胶反演及基本特性研究

发布时间：2018-11-25 10:30

【摘要】：随着我国经济的高速发展,工业生产、汽车尾气、化石燃料燃烧等带来的大气污染日益严重,极端雾霾、极寒、短时强降水、闪电等灾害天气也频频发生,这些现象已严重影响到人们的正常工作与生活。气溶胶是研究气候变化与大气污染的重要参数,也是进行定量遥感必需获得的参数,其参数反演精度、物化性质以及空间分布一直是大气污染和气候变化研究中的热点,也是我们了解最薄弱的部分。因此,借助多源卫星数据,结合地基观测数据,进行气溶胶参数遥感定量反演、大气气溶胶特性及空间分布的精确探测具有重要科学价值。论文主要研究了如下内容:首先,基于边界层气溶胶垂直廓线分布的订正方案,提出了一种全新的基于CALIPSO卫星数据的气溶胶光学厚度(AOD)定量反演算法。利用地面资料对AOD反演结果进行验证,并分析其可能的误差源。结果表明:滤云处理后反演得到的CALIPSO AOD值普遍低于地基观测站的AOD值,而订正后的CALIPSO AOD值与地基观测的AOD值相比,除个别站(龙凤山)外,其他站的拟合斜率和相关系数都有提高,订正效果较好;CALIPSO卫星与地面站的相互位置对验证结果会有影响,其中类型2的拟合结果好于其他类型的拟合结果;不同相对湿度对地基、卫星AOD有一定影响,总而言之,不管是CALIPSO AOD还是订正后的CALIPSO AOD(corrected),拟合斜率和拟合系数都随相对湿度的增加而减小。其次,利用长时间序列的MODIS数据探讨了江西地区AOD空间变化特征,同时利用CALIPSO/CALIOP VFM数据,计算了气溶胶、不同子类型气溶胶和云的垂直概率分布和MPH(最大似然高度)。发现,江西地区AOD水平分布由南向北逐渐递增。其中,九江、南昌地区达到最高。垂直方向上,气溶胶主要累积在1~3.5km,云和气溶胶的混合状态出现概率大于分离状态。在2~4km垂直高度上,春季污染沙尘出现的概率最高,冬季次之,夏季与秋季相当。而烟尘气溶胶夏季出现的概率最高,春、冬季相当,秋季次之。最后,针对北京地区一次极端雾霾事件,通过利用地面和卫星观测数据,结合气象条件,采用后向轨迹模式来分析雾霾形成过程和气溶胶光学特性,并探讨可能的污染来源。结果表明:高的相对湿度、差的扩散条件(低风速和稳定的大气层)和在高排放与气溶胶吸湿增长条件下气溶胶二次转化是这次重霾事件发生的主要原因。雾霾期间,AOD500nm日均值为1.15,水气(CWV)的日均值为0.42cm。Angstrom指数,细粒子模态的日均值分别为1.19和0.81,这表明细小的气溶胶颗粒在大气中含量较多,成为霾期间大气消光的主要贡献者。卫星观测显示,雾霾期间北京地区有一层明显的气溶胶层集中在地面上层2km内,气溶胶类型主要由混合污染气溶胶组成。
[Abstract]:With the rapid economic development of our country, the air pollution brought by industrial production, automobile exhaust gas and fossil fuel combustion is becoming increasingly serious. Extreme haze, extreme cold, short term heavy precipitation, lightning and other disaster weather also occur frequently. These phenomena have seriously affected people's normal work and life. Aerosol is not only an important parameter to study climate change and atmospheric pollution, but also a necessary parameter for quantitative remote sensing. The inversion accuracy, physicochemical properties and spatial distribution of aerosol are always hot spots in the research of atmospheric pollution and climate change. It is also the weakest part we know. Therefore, with the help of multi-source satellite data and ground-based observation data, the remote sensing quantitative inversion of aerosol parameters and the accurate detection of atmospheric aerosol characteristics and spatial distribution are of great scientific value. The main contents of this paper are as follows: firstly, a new (AOD) quantitative inversion algorithm of aerosol optical thickness based on CALIPSO satellite data is proposed based on the revised scheme of boundary layer aerosol vertical profile distribution. The AOD inversion results are verified by ground data, and the possible error sources are analyzed. The results show that the CALIPSO AOD obtained by cloud filtering inversion is generally lower than the AOD value of the ground-based observation station, and the revised CALIPSO AOD value is compared with the AOD value of the ground-based observation, except for the individual stations (Longfeng Mountain). The slope of fitting and correlation coefficient of other stations are improved, and the effect of correction is better. The mutual position of CALIPSO satellite and earth station will affect the verification results, and the fitting results of type 2 are better than those of other types. The satellite AOD is affected by different relative humidity. In a word, both the CALIPSO AOD and the revised CALIPSO AOD (corrected), fitting slope and fitting coefficient decrease with the increase of relative humidity. Secondly, using the MODIS data of long time series, the spatial variation characteristics of AOD in Jiangxi are discussed, and the vertical probability distribution and MPH (maximum likelihood height) of aerosols, different subtypes of aerosols and clouds are calculated by using CALIPSO/CALIOP VFM data. It is found that the horizontal distribution of AOD in Jiangxi is increasing gradually from south to north. Among them, Jiujiang, Nanchang area reached the highest. In the vertical direction, the aerosol mainly accumulates at 1 ~ 3.5km, and the probability of the mixed state of cloud and aerosol is greater than that of the separation state. In the vertical height of 2~4km, the probability of contaminated dust in spring is the highest, followed by winter, and the same in summer and autumn. The probability of soot aerosol appeared in summer was the highest, spring and winter were the same, and autumn was the second. Finally, in the light of an extreme haze event in Beijing, the formation process of haze and the optical properties of aerosol are analyzed by using ground and satellite observation data and meteorological conditions, and the possible sources of pollution are discussed. The results show that high relative humidity, poor diffusion conditions (low wind speed and stable atmosphere) and secondary aerosol conversion are the main causes of this heavy haze event. During haze, the daily average value of AOD500nm is 1.15, the daily average value of water (CWV) is 0.42cm.Angstrom exponent, the daily average value of fine particle mode is 1.19 and 0.81, respectively, which indicates that the fine aerosol particles are more in the atmosphere. To become a major contributor to atmospheric extinction during haze. Satellite observations show that there is an obvious aerosol layer in Beijing during haze, which is mainly composed of mixed polluted aerosols (2km).
【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X513;X87

【参考文献】