Abstract：In order to further understand the spatial distribution of PM2.5 on the ground, based on the PM2.5 measured data in 2019, MCD19A2 aerosol optical depth product of the Moderate Resolution Imaging Spectroradiometer (MODIS) at the L3 level, taking Shandong Province as the study area, and fully considering the factors including population, terrain, and weather. The daily PM2.5 in 2019 was simulated by using the four machine learning algorithms of Random Forest (RF), Support Vector Regression (SVR), Back Propagation Neural Network (BPNN), and Deep Neural Networks (DNN). The result shows the RMSE and MAE values of the RF are 12.67 and 6.62, respectively, which are better than BPNN, SVR and DNN models. RF is most suitable for the daily PM2.5 simulation in Shandong Province.
徐发昭, 李净, 褚馨德, 满元伟. 基于MODIS数据与多机器学习法的日PM2.5模拟研究[J]. 中国环境科学, 2022, 42(6): 2523-2529.
XU Fa-zhao, LI Jing, CHU Xin-de, MAN Yuan-wei. Simulation of daily PM2.5 based on MODIS data and multi-machine learning method. CHINA ENVIRONMENTAL SCIENCECE, 2022, 42(6): 2523-2529.
Zhao C, Wang Y, Su Z, et al. Respiratory exposure to PM2.5 soluble extractdisrupts mucosal barrier function and promotes the development of experimental asthma[J]. Science of The Total Environment, 2020,730:139145.
[2]
曾贤刚,阮芳芳.中国PM2.5污染健康效应的国民经济影响分析[J].中国环境科学, 2020,40(7):3228-3238.Zeng X G, Ruan F F. Analysis on the national economic impact of PM2.5 induced health effects in China[J]. China Environmental Science, 2020,40(7):3228-3238.
[3]
Man Y, Huang Y, Shen H, et al. Effects of urban form on haze pollution in China:Spatial regression analysis based on PM2.5 remote sensing data[J]. Applied Geography, 2018,98:215-223.
[4]
Guan Y, Xiao Y, Wang Y, et al. Assessing the health impacts attributable to PM2.5 and ozone pollution in 338 Chinese cities from 2015 to 2020[J]. Environmental Pollution, 2021,287(9):117623.
[5]
Zhang Y, Li Z. Remote sensing of atmospheric fine particulate matter (PM2.5) mass concentration near the ground from satellite observation[J]. Remote Sensing of Environment, 2015,160:252-262.
[6]
梅杨,党丽娜,杨勇,等.基于时空克里格的PM2.5时空预测及分析[J].环境科学与技术, 2016,39(7):157-163.Mei Y, Dang L N, Yang Y, et al. Spatiotemporal prediction and analysis of PM2.5 based on spatiotemporal kriging[J]. Environmental Science and Technology, 2016,39(7):157-163.
[7]
李杰,翟亮,桑会勇,等.PM2.5浓度插值中不同空间插值方法对比[J].测绘科学, 2016,41(4):50-54+101.Li J, Zhai L, Sang H Y, et al. Comparison of different spatial interpolation methods for PM2.5[J]. Science of Surveying and Mapping, 2016,41(4):50-54+101.
[8]
Wang J, Christopher S A. Intercomparison between satellite-derived aerosol optical thickness and PM2.5 mass:Implications for air quality studies[J]. Geophysical Research Letters, 2003,30(21):2095.
[9]
Geng G, Zhang Q, Martin R V, et al. Estimating long-term PM2.5 concentrations in China using satellite-based aerosol optical depth and a chemical transport model[J]. Remote Sensing of Environment, 2015,166:262-270.
[10]
Liu Y, Rokjin, et al. Mapping annual mean ground-level PM2.5 concentrations using multiangle imaging spectroradiometer aerosol optical thickness over the contiguous United States[J]. Journal of Geophysical Research:Atmospheres, 2004,109(D22).
[11]
Li Z Q, et al. Remote sensing of atmospheric particulate mass of dry PM2.5 near the ground:Method validation using ground-based measurements[J]. Remote Sensing of Environment, 2016,173:59-68.
[12]
Lin C, Li Y, Yuan Z, et al. Using satellite remote sensing data to estimate the high-resolution distribution of ground-level PM2.5[J]. Remote Sensing of Environment, 2015,156:117-128.
[13]
景悦,孙艳玲,徐昊,等.基于混合效应模型的京津冀地区PM2.5日浓度估算[J].中国环境科学, 2018,38(8):2890-2897.Jing Y, Sun Y L, Xu H, et al. Daily estimation of PM2.5 concentrations based on mixed effects model in Beijing-Tianjin-Hebei region[J]. China Environmental Science, 2018,38(8):2890-2897.
[14]
孙成,王卫,刘方田,等.基于线性混合效应模型的河北省PM2.5浓度时空变化模型研究[J].环境科学研究, 2019,32(9):1500-1509.Sun C, Wang W, Liu F T, et al. Spatial-Temporal Simulation of PM2.5 Concentration in Hebei Province based on Linear Mixed Effects Model[J]. Research of Environmental Sciences, 2019,32(9):1500-1509.
[15]
冷湘梓,王勤耕,钱新,等.基于非线性数学方法的PM2.5中重金属浓度模拟[J].中国环境科学, 2017,37(3):821-828.Leng X Z, Wang Q G, Qian X, et al. Simulation of heavy metal concentrations in PM2.5 based on nonlinear mathematics methods[J]. China Environmental Science, 2017,37(3):821-828.
[16]
He Q, Huang B. Satellite-based high-resolution PM2.5 estimation over the Beijing-Tianjin-Hebei region of China using an improved geographically and temporally weighted regression model[J]. Environmental Pollution, 2018,236(MAY):1027-1037.
[17]
Guo Y X, Tang Q H, Gong D Y, et al. Estimating ground-level PM2.5 concentrations in Beijing using a satellite-based geographically and temporally weighted regression model[J]. Remote Sensing of Environment, 2017,198:140-149.
[18]
Liu N, Zou B, Li S, et al. Prediction of PM2.5 concentrations at unsampled points using multiscale geographically and temporally weighted regression[J]. Environmental Pollution, 2021,284:117116.
[19]
Guo B, Wang X, Pei L, et al. Identifying the spatiotemporal dynamic of PM2.5 concentrations at multiple scales using geographically and temporally weighted regression model across China during 2015~2018[J]. Science of The Total Environment, 2020,751:141765.
[20]
陈辉,厉青,张玉环,等.基于地理加权模型的我国冬季PM2.5遥感估算方法研究[J].环境科学学报, 2016,36(6):2142-2151.Chen H, Li Q, Zhang Y H, et al. Estimations of PM2.5 concentrations based on the method of geographically weighted regression[J]. Acta Scientiae Circumstantiate, 2016,36(6):2142-2151.
[21]
夏晓圣,陈菁菁,王佳佳,等.基于随机森林模型的中国PM2.5浓度影响因素分析[J].环境科学, 2020,41(5):2057-2065.Xia X S, Chen J J, Wang J J, et al. PM2.5 Concentration influencing factors in China based on the random forest model[J]. Environmental Science, 2020,41(5):2057-2065.
[22]
刘林钰,张永军,李彦胜,等.基于深度学习的华东地区PM2.5浓度遥感反演[J].环境科学, 2020,41(4):1513-1519.Liu L Y, Zhang Y J, Li Y S, et al. PM2.5 inversion using remote sensing data in Eastern China based on deep learning[J]. Environmental Science, 2020,41(4):1513-1519.
[23]
耿冰,孙义博,曾巧林,等.基于深度学习方法的PM2.5精细化时空估算模型[J].中国环境科学, 2021,41(8):3502-3510.Geng B, Sun Y B, Zeng Q L, et al. Refined spatiotemporal estimation model of PM2.5 based on deep learning method[J]. China Environmental Science, 2021,41(8):3502-3510.
[24]
康俊锋,黄烈星,张春艳,等.多机器学习模型下逐小时PM2.5预测及对比分析[J].中国环境科学, 2020,40(5):1895-1905.Kang J F, Huang L X, Zhang C Y, et al. Hourly PM2.5 prediction and its comparative analysis under multi-machine learning model[J]. China Environmental Science, 2020,40(5):1895-1905.
[25]
Li J, Garshick E, Hart J E, et al. Estimation of ambient PM2.5 in Iraq and Kuwait from 2001 to 2018 using machine learning and remote sensing[J]. Environment International, 2021,151(151):106445.
[26]
Mg A, Yc A, As A, et al. Estimating daily high-resolution PM2.5 concentrations over Texas:Machine learning approach[J]. Atmospheric Environment, 2021,247:118209.
[27]
Chen W, Ran H, Cao X, et al. Estimating PM2.5 with high-resolution 1-km AOD data and an improved machine learning model over Shenzhen, China[J]. Science of the Total Environment, 2020,746:141093.
[28]
陈优良,李亚倩.长三角PM2.5和O3变化特征及与气象要素的关系[J].长江流域资源与环境, 2021,30(2):382-396.Chen Y L, Li Y Q. Characteristics of O3 and PM2.5 and its relationship with meteorological factors in Yangtze River Delta[J]. Resourcesand Environmentin the Yangtze Basin, 2021,30(2):382-396.
[29]
史旭荣,逯世泽,易爱华,等.全国2018~2019年秋冬季气象条件变化对PM2.5影响研究[J].中国环境科学, 2020,40(7):2785-2793.Shi X R, Lu S Z, Yi A H, et al. Impact of meteorological conditions on PM2.5 in autumn and winter of 2018~2019 in key areas[J]. China Environmental Science, 2020,40(7):2785-2793.
[30]
潘晨,康志明.2001~2019年气象条件对江苏省PM2.5分布的影响[J].环境科学, 2022,43(2):649-662.Pan C, Kang Z M. Impact of meteorological conditions on PM2.5 in Jiangsu Province from 2001 to 2019[J]. Environmental Science, 2022,43(2):649-662.
[31]
臧振峰,张凤英,李永华,等.我国主要粮产区PM2.5、PM10时空分布特征及影响因素——以河南省为例[J].自然资源学报, 2021,36(5):1163-1175.Zang Z F, Zhang F Y, Li Y H, et al. Spatiotemporal distribution and affecting factors of PM2.5 and PM10 in major grain producing areas in China:A case study of Henan province[J]. Journal of Natural Resources, 2021,36(5):1163-1175.
[32]
Breiman. Random forests[J]. Mach Learn, 2001,2001,45(1)(-):5-32.
[33]
DE Rumelhart, Hinton G E, Williams R J. Learning representations by backpropagating errors[J]. Nature, 1986,323(6088):533-536.
[34]
Cortes C. Support-vector networks[J]. Machine Learning, 1995,20(3):273-297.
[35]
Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2014,18(7):1527-1554.
[36]
Donkelaar A V, Martin R V, Li C, et al. Regional estimates of chemical composition of fine particulate matter using a combined geoscience-statistical method with information from satellites, models, and monitors[J]. Environmental Science and Technology, 2019,53(5):2595-2611.
