基于卫星高光谱遥感的2007年—2017年新疆地区大气NO_(2)时空变化趋势分析

Analysis of Spatial and Temporal Change Trend of Xinjiang NO_(2) in2007—2017 Based on Satellite Hyperspectral Remote Sensing

自2013年国务院颁布《大气污染防治行动计划》等大气综合治理政策以来,我国东部地区大气颗粒物等重要污染物得到了有效控制。伴随着我国能源政策的变化,西北地区在加大能源开发力度的同时,大气污染亦呈现加剧趋势,但在之前的研究中未得到重视。卫星遥感的监测手段较地面观测具有不受地域限制,观测时间长,观测污染物种类多等优势。其中,星载紫外-可见高光谱仪OMI自2005年在轨运行,在大气污染的时空变化趋势探测、排放源估计以及模式的同化和验证等科学应用中得到广泛应用。中国科学技术大学的卫星对流层NO_(2)柱浓度产品,通过对OMI原始测量光谱的二次标定和气体反演算法的关键优化,在与地基观测的结果对比验证中呈现良好的相关性,适合用于我国高气溶胶背景下的大气污染分析。结合中国科学技术大学OMI NO_(2)数据产品,我国新疆地区大气NO_(2)污染的时空分布特征得以表征。在2007年—2017年期间,新疆地区的NO_(2)污染集中分布于北疆,其中“乌鲁木齐—昌吉—石河子”城市群(乌昌石地区)与新疆总体NO_(2)水平月变化相关性很强(相关性系数r=0.942,p-value<0.01)。新疆NO_(2)年际变化存在明显阶段性特征,与相关政策以及能源行业排放变化相符:2007年—2010年变化趋势不明显,2014年较2010年总体年平均浓度增长18.5%,其中乌昌石地区增长41.3%;2017年较2014年总体平均浓度下降26.4%,乌昌石地区下降42.8%。乌昌石地区由于石油化工企业,经济开发区等分布密集,成为NO_(2)污染的聚集区,与乌鲁木齐市,昌吉市NO_(2)变化具有强相关性(r=0.982,p-value<0.01;r=0.951,p-value<0.01)。受采暖时期排放以及特殊气象条件控制,乌昌石地区NO_(2)污染峰值为12月,冬季污染尤其显著;在2007年—2016年采暖期间(每年10月—次年4月初),乌昌石地区NO_(2)水平有显著上升趋势(显著性水平α=0.01),在未来大气治理中需要格外关注。

Since the State Council promulgated the"Atmospheric Pollution Prevention Action Plan"and other comprehensive atmospheric control policies in 2013,the concentration of atmospheric pollutants and other important pollutants in eastern China has been effectively controlled.Along with the changes in China’s energy policy and increasing energy development in the northwest region,air pollution also shows an increasing trend,but it has not been paid attention to in previous studies.Compared with ground observation,the monitoring method of satellite remote sensing has the advantages of not being restricted by area,lasting long observation time and monitoring many types of pollutants simultaneously.The spaceborne ultraviolet-visible hyperspectrometer OMI has been in orbit since 2005 and has been widely used in scientific applications such as the detection of the temporal and spatial changes of atmospheric pollution,the estimation of emission sources,and the assimilation and verification of models.USTC’s tropospheric NO_(2) column concentration product,through the secondary calibration of the OMI original measurement spectrum and the key optimization of the gas inversion algorithm,shows a good correlation in the comparison and verification with the results of ground-based observations.Analysis of air pollution in the background of sol.Combined with USTC’s OMI NO_(2) data product,the spatial and temporal distribution characteristics of atmospheric NO_(2) pollution in Xinjiang,China,can be characterized.From 2007 to 2017,NO_(2) pollution in Xinjiang was concentrated in northern Xinjiang.Among them,the"Urumqi-Changi-Shihezi"urban agglomeration("Wuchangshi"area)had a strong correlation with the monthly change of the overall NO_(2) level in Xinjiang(correlation coefficient R=0.942,p-value<0.01).The inter-annual change of NO_(2) in Xinjiang has obvious phase characteristics,which is consistent with the changes in relevant policies and energy industry emissions:the changing trend is not obvious from 2007 to 2010,the overall average concentration in 2014 increased by 18.5%compared with 2010,and the"Wuchangshi"area increased by 41.3%,in 2017,the overall average concentration of NO_(2) decreased by 26.4%compared with 2014,and"Wuchangshi"decreased by 42.8%.Due to the dense distribution of petrochemical enterprises and economic development zones in the"Wuchangshi"area,it has become a NO_(2) pollution gathering area,which has a strong correlation with Urumqi and Changi NO_(2) changes(R=0.982,p-value<0.01;R=0.951,p-value<0.01).Controlled by the heating period’s emissions and the special meteorological conditions,the peak of NO_(2) change in the"Wuchangshi"area in December,and the winter pollution is particularly significant.The heating period(from October to early April each year)has a significant upward trend from 2007 to 2016(Significance levelα=0.01),which requires special attention in future atmospheric governance.