承德市PM_(2.5)和PM_(10)浓度分布特征及与气象因子关系的研究

Distribution of PM_(2.5) and PM_(10) in Chengde City and its Relationship with Meteorological Factors

文章利用承德市2016—2022年PM_(2.5)和PM_(10)的质量浓度数据以及同时段气象数据,分析了PM_(2.5)和PM_(10)的年、季、月、日变化特征以及与相关气象因子的关系。结果表明:承德市颗粒物污染有明显的季节分布特征,近7年PM_(10)和PM_(2.5)的超标日数和超标率呈现出波动下降的趋势;PM_(10)和PM_(2.5)浓度峰值普遍出现在4、11月和12月;PM_(10)和PM_(2.5)日变化趋势基本一致,呈现出双峰双谷的变化规律,且PM_(10)的这种日变化规律更为明显。各季节PM_(10)与PM_(2.5)浓度相关性显著,CO与PM_(10)、PM_(2.5)浓度的相关系数大于其他污染物;PM_(2.5)和PM_(10)浓度与日平均气温、日降水量、日平均风速、日最大风速、日照时数、最小能见度呈显著负相关;PM_(2.5)与相对湿度呈显著正相关关系,PM_(10)与相对湿度呈显著正相关关系。秋季降水日PM_(2.5)和PM_(10)的浓度分别较非降水日降低了2.5μg·m^(-3)和18.3μg·m^(-3);相对湿度达到75%左右时,PM_(2.5)浓度值达到峰值;在一定范围内气温每升高1.0℃,PM_(2.5)浓度下降0.37μg·m^(-3)。利用逐步回归的方法建立了承德地区PM_(2.5)和PM_(10)的预报模型,拟合优度分别在0.8和0.7以上,对模型进行检验,结果表明,在一定范围内模型效果较好,具有一定的实用性。

Based on the PM_(2.5) and PM_(10) concentration data and meteorological observation data in Chengde from 2016 to 2022, the annual, seasonal, monthly and daily variation characteristics of PM_(2.5) and PM_(10) and their relationship with meteorological factors were analyzed. The results showed that the particulate matter pollution in Chengde City had obvious seasonal variation characteristics, and the number of pollution days and exceedence probability of PM_(10) and PM_(2.5) showed the double trends of fluctuating and decreasing. The peak concentrations of PM_(10) and PM_(2.5) generally occurred in April, November and December. The daily variation trend of PM_(10) was similar to that of PM_(2.5), showing two peaks and two valleys, but was more obvious. There was a significant correlation between PM_(10) and PM_(2.5) concentration in each season, and the correlation coefficient between CO and PM_(10) and PM_(2.5) concentration was greater than others. The concentrations of PM_(2.5) and PM_(10) were negatively correlated with average daily temperature, daily precipitation, average daily wind speed, maximum daily wind speed, sunshine duration and minimum visibility. Both PM_(2.5) and PM_(10) were positively correlated with relative humidity. The concentrations of PM_(2.5) and PM_(10) decreased by 2.5 μg·m^(-3) and 18.3 μg·m^(-3) respectively on precipitation days compared with non-precipitation days. When the relative humidity was about 75%, the concentration of PM_(2.5) reached its peak. Within a certain range, the PM_(2.5) concentration decreased by 0.37 μg·m^(-3) for every 1 ℃ increase in air temperature. The prediction models of PM_(2.5) and PM_(10) in Chengde region were established by stepwise regression method, and the goodness of fit of two models were above 0.8 and 0.7, respectively. The test results showed that the prediction models had good effect and practicability within a certain range.