自动与人工观测霾日、雾日序列连续性分析  

Analysis of Continuity of Fog,Light Fog and Haze Climate Series by Automatic Identification and Manual Observation

作　　者：任芝花[1] 余予[1] 韩瑞[1] 冯明农[1] REN Zhihua;YU Yu;HAN Rui;FENG Mingnong(National Meteorological Information Centre, Beijing 100081, China)

机构地区：[1]国家气象信息中心,北京100081

出　　处：《高原气象》2018年第3期863-871,共9页Plateau Meteorology

基　　金：国家自然科学基金重大研究计划(91744209);中国气象局预报预测核心业务发展专项(CMAHX20160703);中国气象局气象关键技术集成与应用重点项目(CMAGJ2015Z16);公益性行业(气象)科研专项(GYHY201106038)

摘　　要：由原始观测数据统计显示,2014年全国雾霾(包括雾、轻雾和霾)日数偏高,尤其霾日数是2000-2013年均值的4.2倍。依据2014年雾霾现象保留人工观测或采用自动观测,将全国2 400余个国家站分为人工站和自动站,分别就两类站2014年雾霾现象日数与其历史序列进行了对比分析。结果表明,人工站2014年平均雾霾日数与其2000年以来平均状况及2013年平均雾霾日数接近;而自动站2014年平均雾霾日数明显偏高,成为2000年以来之最,且明显高于2013年统计结果。通过对比分析自动与人工观测方法表明,雾霾现象自动观测采用瞬间观测记录,是造成2014年全国雾霾日数异常偏高的主要原因。基于霾现象持续性特征,针对自动观测霾数据,研究确定了1天至少6个连续时次"现在天气现象"有霾记录,"连续天气现象"方记霾现象的订正方法。统计结果表明,订正前全国2014年平均霾日数为59天,而订正后下降为31天,基本与2013年持平。进一步分析表明,2000年以来我国霾发生频率呈增加趋势,尤其2013年和2014年,霾发生频率成为2000年以来之最。Based on the fog, light fog and haze daily data from more than 2 400 national level surface stations, comparisons were made between the statistics of 2014 and their historical series. Results showed that the national mean fog, light fog and haze days were 20, 154 and 59 respectively and were all highest values from the year of 2000. Especially, haze day of 2014 was 4. 2 times of the means from 2000 to 2013, which may not match the actual conditions. Further analysis were presented by dividing the stations into two parts, as about 963 stations were carried out automatic identifications of the vision obstruction weather phenomenon from the beginning of 2014, while the other stations still maintained manual observations. The mean fog, light fog and haze days of manual stations in 2014 were close to their averages of recent decade and the statistics of 2013. As for the auto- matic stations, the mean values were rocketed to the highest starting from the year of 2000 and were extremely higher than the statistics of 2013. Through the analysis of the differences between automatic and manual observa- tion methods and their data, the unusual higher values in 2014 were mainly ascribed to the weather phenomenon identification employed by automatic stations, as instantaneous occurrences of fog, light fog and haze were im- properly written into the daily records. Based on the characteristics of haze persistence, correction method was studied on haze identification at automatic stations by using present weather phenomenon data. When at least six consecutive present weather phenomenon occurred in one day, haze was then written into the consecutive weather phenomenon record, i. e. the daily record and that day was consequently identified as a haze day. This method was verified by comparing the corrected data with the air quality conditions between 2013 and 2014 in 74 cities released by the Ministry of Environmental Protection, which shown a good agreement. The national mean haze day number of 2014 dropped down to about 31 days fr

关 键 词：雾霾现象 自动观测 人工观测 连续性 

分 类 号：P413[天文地球—大气科学及气象学]