研究报告
张海霞,程先富,陈冉慧.安徽省PM2.5时空分布特征及关键影响因素识别研究[J].环境科学学报,2018,38(3):1080-1089
安徽省PM2.5时空分布特征及关键影响因素识别研究
- Analysis on the spatial-temporal distribution characteristics and key influencing factors of PM2.5 in Anhui Province
- 基金项目：国家自然科学基金项目（No.41271516）；安徽师范大学研究生科研创新项目（No.2017cxsj059）
- 作者
- 单位
- 张海霞
- 1. 安徽师范大学地理与旅游学院, 芜湖 241002;2. 安徽自然灾害过程-防控研究省级实验室, 芜湖 241002
- 程先富
- 1. 安徽师范大学地理与旅游学院, 芜湖 241002;2. 安徽自然灾害过程-防控研究省级实验室, 芜湖 241002
- 陈冉慧
- 1. 安徽师范大学地理与旅游学院, 芜湖 241002;2. 安徽自然灾害过程-防控研究省级实验室, 芜湖 241002
- 摘要：基于2015年安徽省67个空气质量监测子站的PM2.5浓度数据，分析PM2.5的时空分布特征；运用BP神经网络改进DEMATEL模型，探讨影响PM2.5浓度的关键因素及因子间的关联性.结果表明：①2015年安徽省PM2.5平均浓度为52.03 μg·m-3，总体呈现秋冬高、春夏低的季节性规律；PM2.5浓度日变化总体呈双峰分布，冬季PM2.5浓度昼夜变化剧烈，全年、春季和秋季变化趋势大致相同，夏季相对平缓；②安徽省PM2.5浓度整体上由东向西、由中部向南北两侧呈递减趋势，浓度值由高到低依次为：江淮丘陵、长江中下游平原、淮北平原和皖南山区；③指标体系中，人口城镇化率、年平均气温、单位GDP电耗、工业废气治理设施数等4个指标因子属于强驱动因素，对PM2.5浓度降低起着根本性推动作用；④年降水总量、房屋施工面积、O3浓度等3个指标因子属于强特征因素，是降低PM2.5浓度最直接的因素.结论表明，运用BP-DEMATEL模型能有效识别关键影响因素，有助于为PM2.5综合治理提供参考.
- Abstract：Based on the monitoring data of PM2.5 from 67 air quality monitoring sub-stations in Anhui Province in 2015, the spatial-temporal distribution characteristics of PM2.5 were discussed. The traditional decision-making trial and evaluation laboratory (DEMATEL) model was improved by back propagation (BP) neural network, and then BP-DEMATEL model was applied to investigating the key influencing factors of PM2.5 concentration and the association among factors. The results showed that:① The annual average concentration of PM2.5 in Anhui is 52.03 μg·m-3 in 2015. PM2.5 concentration goes with the seasonal characteristic, which level is high in autumn & winter and low in spring & summer in general. The daily variation of PM2.5 concentration has bimodal curves in cities of Anhui. The diurnal variation of PM2.5 is severe in winter, while it relatively remains stable in summer. The trend is similar among the whole year, spring and autumn. ② The concentration of PM2.5 in Anhui takes a generally decreasing trend from east to west, from the middle to both sides. The annual average concentration of PM2.5 range from high to low in order of the Jianghuai Hill, the middle-lower Yangtze River Plain, the Huaibei Plain and the mountainous areas in southern Anhui. ③ In the index system, the population urbanization rate, annual average temperature, unit GDP power consumption, and numbers of treatment facilities for industrial waste gas play a fundamental role in reducing PM2.5 concentration. ④ The annual precipitation, building construction area and O3 concentration belong to strong characteristic factors, which could most directly reduce the concentration of PM2.5. The BP-DEMATEL model will be helpful for the comprehensive management of PM2.5.
摘要点击次数： 1222 全文下载次数： 1557
