基于观测、模拟和同化数据的PM2.5污染回顾分析

发布时间：2018-02-10 22:23

  本文关键词： 最优插值 资料同化 PM.　出处：《气候与环境研究》2016年06期 　论文类型：期刊论文

【摘要】：基于观测数据空间插值、数值模拟以及最优插值同化方法构建了京津冀地区PM2.5(空气动力学当量直径小于等于2.5μm的颗粒物,即细颗粒物)空间插值数据、模拟数据和同化数据,并首次比较分析了三种数据在PM2.5污染回顾分析上的应用潜力和优缺点。针对2014年2月19~28日京津冀地区PM2.5污染过程的分析发现:(1)观测空间插值数据难以完整表征PM2.5污染的时空演变特征,在没有观测覆盖区域误差较大,容易出现虚假的高低值中心;(2)模拟数据具有较高时空分辨率,对PM2.5污染时空演变特征描述更加细致,但在这次污染过程中仍存在较大不确定性,其均方根误差大于100μg/m3;(3)同化数据不仅能对PM2.5空间分布特征进行细致描述,其数据精度在独立验证站点也显著高于模拟数据,其均方根误差比模拟数据低约50%,与站点观测数据的相关系数也比模拟数据高0.2以上。基于PM2.5同化数据,对这次京津冀PM2.5污染过程的时空演变特征进行了详细回顾分析,发现这次污染过程存在自京津冀南部PM2.5污染累积并向北输送发展的生成特点,消亡过程为风向转换下自北向南清除,造成京津冀南部城市先污染后清除,北部城市后污染先清除,并且有慢累积、快清除的特征。从发展演变过程中污染所占空间面积来看,25日PM2.5污染范围最大,覆盖模式第三区域60.5%面积。
[Abstract]:Based on the spatial interpolation of observation data, numerical simulation and optimal interpolation assimilation method, the spatial interpolation data, simulated data and assimilation data of PM2.5 (particle with aerodynamics equivalent diameter less than 2.5 渭 m) in Beijing-Tianjin-Hebei region are constructed. The application potential, advantages and disadvantages of three kinds of data in PM2.5 pollution retrospective analysis were compared and analyzed for the first time. Based on the analysis of PM2.5 pollution process in Beijing-Tianjin-Hebei area from February 2014 to 28th, it was found that the spatial interpolation data of the three kinds of data could not fully represent PM2.5. The temporal and spatial characteristics of pollution, The simulated data have higher spatial and temporal resolution and more detailed description of the spatio-temporal evolution characteristics of PM2.5 pollution, but there is still a great deal of uncertainty in the process of this pollution. The root mean square error (RMS) is more than 100 渭 g / m ~ (3)) assimilation data can not only describe the spatial distribution characteristics of PM2.5 in detail, but also have a higher accuracy than simulation data at independent verification stations. The root mean square error is about 50% lower than the simulated data, and the correlation coefficient with the observed data is more than 0.2 higher than that of the simulated data. Based on the PM2.5 assimilation data, the temporal and spatial evolution characteristics of the PM2.5 pollution process in Beijing-Tianjin-Hebei are reviewed and analyzed in detail. It is found that the pollution process is characterized by accumulation of PM2.5 pollution from the southern part of Beijing-Tianjin-Hebei Province and its northward transport and development. The process of extinction is the removal from north to south under the change of wind direction, resulting in the first pollution and then the clean-up of the cities in the southern part of Beijing-Tianjin-Hebei. In 25th, the pollution area of PM2.5 was the largest, covering 60.5% area in the third region of the model.
【作者单位】： 中国科学院大气物理研究所大气边界层物理与大气化学国家重点试验室;中国科学院大学;中国环境监测总站;
【基金】：中国科学院战略性先导科技专项XDB05030200 国家自然科学基金项目41575128、41305111~~
【分类号】：X513
，

本文编号：1501581