京津石三市空气污染指数(API)的时空分布特征及其与气象要素的关系

Relationship between Temporal-Spatial Distribution Pattern of Air Pollution Index and Meteorological Elements in Beijing, Tianjin and Shijiazhuang

依据2001年-2010年的空气污染指数(API)日报数据和相应时段的地面气象要素数据,从季节平均和月平均、空气污染等级和空气污染过程3个角度出发,分别研究北京、天津和石家庄3市的空气污染指数(API)与气象要素之间的相关关系,统计分析结果表明:①北京、天津、石家庄3市的空气污染状况在2001年-2010年期间逐步改善,夏季空气质量最好;北京春季受沙尘的影响、天津和石家庄冬季受取暖的影响导致对应季节的空气质量最差;②气象要素对空气污染的影响存在较大时空差异,北京、天津和石家庄3市的月均尺度的API与气温(平均、最高和最低)、最高气压、水汽压和最大风速4类气象要素存在一致的相关关系,大部分相关系数的绝对值天津>石家庄>北京;季均值尺度下,北京、天津和石家庄3市的API与气象要素的关系差异性较大,但3市的API与水汽压存在一致的相关关系;和季节与月份均值处理分析的相关性相比,基于空气污染不同级别分析的相关系数明显偏低,可能与较大的样本量和气象要素对空气污染影响机制的复杂性有关;基于空气污染过程的相关系数的绝对值介于均值和级别分析结果的中间,但是3个城市气象要素对空气污染的影响相关系数仍显不足;③根据不同风向上污染日数的分布统计分析,北京、天津和石家庄分别主要受西南和南西南、北西北和北以及东南和东东南方向上极大风速和最大风速的影响。为了使研究结果更具机理性和普适性,更好的服务于空气污染未来预报,仍需进一步开展包括上述多种方式的联合与深入研究。

Based on the daily data of air pollution index （API） and the ground meteorological elements data of corresponding period from 2001 to 2010, this paper has researched the correlations between API and meteorological elements in Beijing, Tianjin and Shijiazhuang respectively from three perspectives of seasonally and monthly average, different air pollution levels and several typical kinds of air pollution process. The conclusions are as follows： 1） During the period of 2001 and 2010, the air pollution conditions in Beijing, Tianjin and Shijiazhuang have been improved gradually, and the air quality in summer is the best among four seasons. The air quality of Beijing is worst in spring due to dust weather, and that of Tianjin and Shijiazhuang is worst in winter due to heating; 2） The impacts of meteorological elements on air pollution vary greatly with time and space. The monthly average API has consistent correlation with air temperature （mean, highest and lowest）, the highest air pressure, average vapor pressure and the maximum wind speed. Most of absolute values of the correlation coefficients in Shijiazhuang are bigger than those in Beijing but smaller than those in Tianjin. Under seasonally average size, the differences of correlation coefficient between API and other meteorological elements among the three cities are considerably big, but API and vapor pressure have consistent correlation. Compared with the correlations of seasonally and monthly average disposal cases, the correlation coefficient based on various air pollution levels is obviously lower, which is probably related to larger sample size and the complicated mechanism. The absolute value of the correlation coefficient based on specified and continuous air pollution process is between the mean value and the result of grade case analysis. However, none of the three correlation coefficients has reached sufficient accuracy; 3） According to an analysis of the distribution of pollution days for different wind directions, Beijing, Tianjin and Shijiazhuang are mainly influenced by extreme wind speed and maximum wind speed from southwest and south-southwest, north-northwest and north, as well as southeast and east-southeast respectively. To make the research conclusions more mechanismic and universal as well as to provide better services for air pollution forecast in future, we need to do more integrated and intensive study.