基于生态系统服务供需的城市PM2.5污染风险时间变化特征与影响因素分析OA
Temporal Dynamics and Influencing Factors of Urban PM2.5 Pollution Risk Based on Ecosystem Service Supply and Demand
随着中国城市化和工业化进程的加快,城市 PM2.5 污染风险日益突出,PM2.5 污染风险的时空格局变化特征和影响机制成为学界关注的热点.以北京市五环内区域为例,构建了基于生态系统服务供需的PM2.5污染风险表征方法,定量评估了2008-2021 年该区域的PM2.5 污染风险,使用基于k-means的时间聚类方法分析了PM2.5 污染风险的年际和年内时间变化特征,应用 XGBoost 机器学习方法解析了影响 PM2.5 去除服务供需和污染风险的气象因素.结果表明,北京市五环内区域的PM2.5 污染风险在 2008-2021 年呈现出先加重后减轻的变化特征;不同年份的高风险和低风险聚类在年内的时间分布基本稳定,高、低风险类交叉的时期大致为每年的春季、秋季和冬季,低风险类和无风险为主的时期大致为夏季.PM2.5 去除服务供给的最主要气象影响因素为比湿度和风速,服务需求的最主要气象影响因素为地表短波辐射.降水、比湿度、地表短波辐射在2012年和2016年均为PM2.5污染风险的主要影响因素;地表长波辐射在夏季和2021年为污染风险的主要影响因素;风速和温度在季节尺度的影响虽较弱,但与污染风险有显著的负相关性.该研究可为理解城市PM2.5污染风险时间格局特征提供新视角,并为定量分析污染风险的气象影响因素提供新方法.
With the accelerated urbanization and industrialization in China,the risk of PM2.5 pollution in cities has become increasingly prominent.The spatiotemporal pattern characteristics and influencing mechanisms of PM2.5 pollution risk have become a focus of academic research.Taking the area within the Fifth Ring Road of Beijing(AWFRRB)as a case study,this study constructed a PM2.5 pollution risk characterization method based on ecosystem service supply and demand,quantitatively assessed the PM2.5 pollution risk in this area from 2008 to 2021.Moreover,this study analyzed the interannual and intra-annual temporal variation characteristics of PM2.5 pollution risk using the temporal clustering method based on k-means,and applied the XGBoost machine learning method to analyze the meteorological factors influencing the supply and demand of PM2.5 removal services and the pollution risk.The results show that the PM2.5 pollution risk in the AWFRRB exhibited a changing pattern of first increasing and then decreasing from 2008 to 2021.The risk levels within a year in the four evaluated years were generally stable.The period exhibiting an intersection of high-risk and low-risk clusters generally prevailed from autumn through winter to spring,with a consistent period of low-risk cluster and non-risk occurring during the summer.The primary meteorological influencing factors of PM2.5 removal service supply were specific humidity and wind speed,while the primary influencing factor of service demand was surface shortwave radiation.Precipitation,specific humidity,and surface shortwave radiation were the main influencing factors of PM2.5 pollution risk in 2012 and 2016.Long-wave radiation was the main influencing factor of pollution risk in summer and 2021.Although the influence of wind speed and temperature was relatively weak on a season scale,they had a significant negative correlation with the pollution risk.This study provides a new perspective for understanding the temporal pattern characteristics of urban PM2.5 pollution risk and offers a new method for quantitatively analyzing the meteorological influencing factors of pollution risk.
慕浩枫;宋喆禄;高镇;侯鹰;陈卫平
郑州大学河南先进技术研究院,河南 郑州 450003||中国科学院生态环境研究中心/区域与城市生态安全全国重点实验室,北京 100085中国科学院生态环境研究中心/区域与城市生态安全全国重点实验室,北京 100085||中国科学院大学,北京 100049中国科学院生态环境研究中心/区域与城市生态安全全国重点实验室,北京 100085||中国科学院大学,北京 100049||上海交通大学环境科学与工程学院,上海 200240中国科学院生态环境研究中心/区域与城市生态安全全国重点实验室,北京 100085||中国科学院大学,北京 100049中国科学院生态环境研究中心/区域与城市生态安全全国重点实验室,北京 100085||中国科学院大学,北京 100049
资源环境
生态系统服务供需PM2.5污染风险影响因素时间序列分析XGBoost模型
ecosystem service demand and supplyPM2.5 pollution riskinfluencing factorstemporal sequence analysisXGBoost model
《生态环境学报》 2026 (2)
256-266,11
国家自然科学基金项目(42471313)
评论