• 关键词：夜间灯光遥感  夜间PM_2.5浓度  NPP/VIIRS  SVM模型  辐射传输

• 基金项目：江苏省环境监测科研基金项目（No.1903）；国家自然科学基金面上项目（No.41871028）；江苏省青蓝工程 （No.R2019Q03）

• 陈惠娟
• 南京信息工程大学，南京 210044

• 徐永明
• 南京信息工程大学，南京 210044

• 莫亚萍
• 南京信息工程大学，南京 210044

• 张悦
• 江苏省环境监测中心，南京 210036

• 杨子毅
• 江苏省淮安环境监测中心，淮安 223001

• 摘要：PM_2.5是大气的重要污染物，掌握其空间分布对于大气污染防控具有重要意义．目前，PM_2.5遥感监测主要围绕卫星反演的日间AOD数据开展，无法反映夜间大气污染的空间格局．以2019年9—12月NPP/VIIRS夜间灯光影像和空气质量站点PM_2.5观测数据对江苏省淮安市夜间PM_2.5浓度进行估算研究．基于辐射传输方程分析夜间灯光辐射与PM_2.5浓度之间的关系，在此基础上综合考虑灯光辐射直接衰减和散射补偿确定了计算夜间PM_2.5浓度的空间自变量，运用多元线性回归模型（MLR）、随机森林（RF）、Cubist、极端梯度提升树（XGBoost）、神经网络（NNet）、支持向量机（SVM）及最近邻法（KNN）算法构建夜间PM_2.5浓度遥感估算模型．结果表明，多元线性归回模型精度明显低于各个机器学习模型，所有模型中SVM模型精度最高，决定系数R²为0.77，平均绝对误差MAE为20.83 μg·m^-3，均方根误差RMSE为32.05 μg·m^-3．基于建立的SVM模型估算了淮安市夜间PM_2.5浓度，并对其空间分布特征进行了分析．本研究探索了利用夜间灯光遥感数据估算夜间PM_2.5浓度的方法，为夜间大气环境监测与管理提供了参考．

• Abstract：PM_2.5 is an important pollutant in the atmosphere， and detailed knowledge of its spatial distribution is essential for air pollution prevention and control. Most previous studies have focused on estimating PM_2.5 concentrations from satellite derived daytime AOD data， which cannot effectively depict nighttime atmospheric environment. This paper aims to estimate the nighttime PM_2.5 concentrations in Huai'an City， Jiangsu Province， using NPP/VIIRS nighttime light data and station observed PM_2.5 data during September-December 2019. The relationship between nighttime light radiation and PM_2.5 concentration was first explored based on radiative transfer equation. Taking into account direct attenuation of light radiation and scattering compensation， the spatial independent variables for nighttime PM_2.5 estimation were determined. Multiple Linear Regression （MLR）， Random Forest （RF）， Cubist， Extreme Gradient Boosting Tree （XGBoost）， Neural Network （NNet）， Support Vector Machine （SVM） and Nearest Neighbor Method （KNN） were used to develop remote sensing models for estimating nighttime PM_2.5 concentrations. The results showed that MLR model had a significantly lower accuracy than machine learning models， and SVM model outperformed other models， with a coefficient of determination （R²） of 0.77， a mean absolute error （MAE） of 20.83 μg·m^-3 and a root mean square error （RMSE） of 32.05 μg·m^-3. Huai'an nighttime PM_2.5 concentration was mapped based on the developed SVM model， and its spatial distribution characteristics were analyzed. This paper explores a method for estimating nighttime PM_2.5 concentrations with nighttime light remote sensing data， and provides references for the monitoring and management of the nighttime atmospheric environment.

• Key words：nighttime light remote sensing  nighttime PM_2.5  NPP/VIIRS  SVM  radiative transfer

• 
  摘要点击次数： 303 全文下载次数： 470