基于卫星遥感气溶胶产品评估体系的气溶胶时空分布及变化趋势表征

卫星遥感气溶胶虽然空间覆盖度高，但不同产品数据在准确性和适用性上存在显著差异.为了能够科学定量的衡量各个产品的优劣，选择最合适的气溶胶产品，该研究以数学统计为基础，提出了一种卫星遥感气溶胶产品评估体系，并依此确定了全球6个人口密集地区的最优卫星数据集，在此基础上开展了近10年的（2009—2018）气溶胶时空变化分析.结果表明暗像元算法在植被覆盖度高的地区表现最好，深蓝算法在亮地表的沙漠干旱地区更占优势，而暗像元-深蓝融合算法在土地类型复杂且气溶胶来源多变时得以突出.人口密集的6个区域中，除了区域A和区域B，其它地区的大气污染水平整体偏高.其中，A、B、E的气溶胶负荷呈下降变化，区域F基本不变，其余区域上升变化.评估体系的建立为气溶胶遥感产品的选取提供了新的衡量方法，且该文关于气溶胶时空分布及变化的分析可以为区域性气溶胶研究提供一定的参考价值.

Assessment of aerosol spatiotemporal distribution and variations based on satellite remote sensing aerosol product evaluation system

Although the space coverage of satellite remote sensing aerosol is high, there are significant differences in accuracy and applicability among different product data. A satellite remote sensing aerosol product evaluation system was proposed in this study based on mathematical statistics in order to evaluate the quality of each product scientifically and select the most suitable aerosol product. The analysis of spatiotemporal variation of aerosol in the past decade (2009-2018) was carried out over six densely populated regions in the world after the optimal satellite datasets of the regions were determined. The results show that dark target algorithm performs best in areas with high vegetation coverage, deep blue algorithm lies in desert arid areas with bright surface, and dark target-deep blue algorithm is prominent when land types are complex and aerosol sources are variable. Heavy atmospheric pollutions were recorded in the densely populated areas, except for area A and B. The aerosol load of area F is relatively stable. In addition, the aerosol load of areas A, B, and E decreased with time, and the opposite changes were found over areas C and D. The establishment of the evaluation system provides a new measurement method for the selection of aerosol remote sensing products, and the analysis of aerosol spatiotemporal distribution and variation can provide certain reference value for regional aerosol research.