PM2.5预测浓度的影响因素分析及防治措施
Analysis of influencing factors on predicted PM2.5 concentrations and pollution control measures
安杨1
作者信息
- 1. 上海电气集团国控环球工程有限公司,太原 030024
- 折叠
摘要
为分析大气环境影响评价中气象场对PM2.5 预测浓度的影响,利用NASA MODIS气溶胶光学厚度(AOD)1km网格遥感数据,并结合地面PM2.5 监测数据及气象数据构建普通最小二乘法(OLS)模型及随机森林回归预测模型,得到太原主城区非采暖季污染日PM2.5 浓度.结果表明:补充气象要素前PM2.5 与AOD拟合R2 三季皆低于 0.2,拟合度较低,在加入气象因素订正后拟合R2 提升至 0.6 以上;随机森林模型拟合优度均高于普通最小二乘法模型.大气环境影响评价中气象场对PM2.5 预测浓度的影响显著.
Abstract
To analyze the influence of meteorological fields on the predicted concentration of PM2.5 in atmospheric environmental impact assessments,this study utilized NASA MODIS aerosol optical depth(AOD)1 km grid remote sensing data,combined with ground PM2.5 monitoring data and meteorological data to construct an ordinary least squares(OLS)model and a random forest regression prediction model,obtaining the PM2.5 concentration in the main urban area of Taiyuan during non-heating seasons.The results show that before supplementary meteorological elements,the PM2.5-AOD fitting R2 was below 0.2 for all three seasons,indicating a low fitting degree.After correcting for meteorological factors,the fitting R2 increased to above 0.6.The fitting accuracy of the random forest model was higher than that of the OLS model.The influence of meteorological fields on the predicted concentration of PM2.5 in atmospheric environmental impact assessments is significant.
关键词
遥感/气溶胶光学厚度(AOD)/PM2.5/随机森林Key words
remote sensing/aerosol optical thickness(AOD)/PM2.5/random forest引用本文复制引用
出版年
2025
NETL
NSTL
万方数据