Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (6): 1408-1415    DOI: 10.11873/j.issn.1004-0323.2021.6.1408
遥感应用     
中国东部沿海四省一市PM2.5浓度遥感估算方法研究
杨立娟(),张建霞,林木生
闽江学院 测绘工程系,福建 福州 350018
Research on Methods of Remotely Sensed PM2.5 Concentrations Estimation in Four Provinces and One City along the East Coast of China
Lijuan Yang(),Jianxia Zhang,Musheng Lin
Department of Surveying and Mapping Engineering of Minjiang University,Fuzhou 350118,China
 全文: PDF(3881 KB)   HTML
摘要:

卫星遥感反演的气溶胶光学深度(AOD)产品已被广泛应用于近地面PM2.5浓度的估算。已有研究表明通过构建AOD和PM2.5之间的高级统计模型—线性混合效应模型(LME)可以有效获取近地面PM2.5浓度的空间分布,但由于引入了大量的气象和土地利用等因子,使得模型对变量的解译能力有所降低。为此,基于MODIS AOD(空间分辨率:3 km),以我国东部长江三角洲—福建—广东(YRD-FJ-GD)为研究区,构建了两种非参数机器学习模型,即支持向量机(SVM)和随机森林(RF)模型,来估算2018年YRD-FJ-GD地区的近地面PM2.5浓度,并将其与线性混合效应模型(LME)的估算结果进行对比。研究发现,3种模型估算的PM2.5浓度与地面实测值之间的R2均高于0.6,其中,RF模型的估算精度最优,模型拟合的R2高达0.91,比SVM模型(R2=0.79)和LME模型(R2=0.64)的估算结果分别提高了13%和30%;且RMSE(~9.07 μg/m3)也远低于LME(~19.09 μg/m3)和SVM模型(~17.29 μg/m3)。此外,由随机森林(RF)模型估算的2018年YRD-FJ-GD地区的PM2.5空间分布显示,长江三角洲(YRD)地区的年均PM2.5浓度最高(>46 μg/m3),其次为广东省(GD),福建地区(FJ)的年均PM2.5浓度最低(<37 μg/m3);4个季节的平均PM2.5浓度则呈现冬季(46.32 μg/m3)>春季(38.80 μg/m3)>秋季(36.15 μg/m3)>夏季(30.16 μg/m3)的分布格局。研究结果表明:与高级统计模型(LME)和机器学习(SVM)相比,随机森林(RF)模型能更好地应用于YRD-FJ-GD地区的PM2.5浓度估算。

关键词: LMESVMRFPM2.5估算YRD?FJ?GD    
Abstract:

The Aerosol Optical Depth (AOD) derived from remote sensing imageries has been widely used in estimating ground-level PM2.5 concentrations in large areas. Previous studies that focused on PM2.5 estimation have reported high predictability of PM2.5 concentrations when using AOD and the advanced statistical model (i.e., Linear Mixed Effects model (LME)). However, the interpretation ability of the LME model was lowered, as it introduced many meteorological and land use variables in the model, and the importance of each variable to PM2.5 concentrations was hard to interpret. Therefore, this study developed two nonparametric machine learning methods, i.e., Support Vector Machine (SVM) and Random Forest (RF), to estimate the ground-level PM2.5 concentrations. The eastern Yangtze River Delta-Fujian-Guangdong (i.e., YRD-FJ-GD) region in China was employed as our study case, and we also compared the predictability of these two models with the LME model. The results showed that the overall R2 between estimated and observed PM2.5 concentrations exceeded 0.6 for three models, where RF received a R2 of 0.9, i.e., 13% and 30% higher than SVM (R2=0.79) and LME (R2=0.64) model, respectively. The RMSE values were 9.07, 17.29 and 19.09 μg/m3 for RF, SVM and LME model, respectively. In addition, the spatial distribution of PM2.5 concentrations estimated from the optimal model (i.e., RF) illustrated high annual PM2.5 in YRD (>46 μg/m3), and GD ranked the second. FJ exhibited a relatively low annual PM2.5 (<37 μg/m3). The seasonal PM2.5 concentrations presented a distribution pattern as winter (6.32 μg/m3) > spring (38.80 μg/m3) > autumn (36.15 μg/m3) > summer (30.16 μg/m3). Our results revealed that the AOD and RF model could be a good proxy for estimating PM2.5 concentrations in YRD-FJ-GD region.

Key words: LME    SVM    RF    PM2.5 estimation    YRD-FJ-GD
收稿日期: 2020-09-15 出版日期: 2022-01-26
ZTFLH:  TP79  
基金资助: 闽江学院优秀引进人才科研启动项目(MJY20001);福建省自然科学基金项目(2021J05204)
作者简介: 杨立娟(1985- ),女,福建三明人,博士,副教授,主要从事遥感技术与应用研究。E?mail:subrinarzhong@aliyun.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
杨立娟
张建霞
林木生

引用本文:

杨立娟,张建霞,林木生. 中国东部沿海四省一市PM2.5浓度遥感估算方法研究[J]. 遥感技术与应用, 2021, 36(6): 1408-1415.

Lijuan Yang,Jianxia Zhang,Musheng Lin. Research on Methods of Remotely Sensed PM2.5 Concentrations Estimation in Four Provinces and One City along the East Coast of China. Remote Sensing Technology and Application, 2021, 36(6): 1408-1415.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.6.1408        http://www.rsta.ac.cn/CN/Y2021/V36/I6/1408

图1  研究区(YRD-FJ-GD)
图2  技术路线框图
图3  模型拟合和验证结果
图4  各变量对PM2.5浓度变异的重要性
模型
R2RMSE(μg/m3)R2RMSE(μg/m3)R2RMSE(μg/m3)R2RMSE(μg/m3)
LME0.6018.820.5013.820.5517.330.6429.01
SVR0.7413.120.819.010.7813.320.8020.05
RF0.868.490.926.110.927.840.9112.53
表1  3个模型在四季的PM2.5估算对比
图5  2018年YRD-FJ-GD地区季均和年均PM2.5浓度空间分布
1 Li Z B, Roy David, Zhang H K, et al. Evaluation of Landsat-8 and Sentinel-2A aerosol optical depth retrievals across Chinese cities and implications for medium spatial resolution urban aerosol monitoring[J]. Remote Sensing,2019,11(2):122. DOI: 10.3390/rs11020122.
doi: 10.3390/rs11020122
2 Lu X M,Wang J J,Yan Y T, et al. Estimating hourly PM2.5 concentrations using Himawari-8 AOD and a DBSCAN-modified deep learning model over the YRDUA, China [J]. Atmospheric Pollution Research, 2021, 12(2): 183-192. DOI:10.1016/j.apr.2020.10.020.
doi: 10.1016/j.apr.2020.10.020
3 Wang Z T, Gao S L, Xie J F, et al. Identification of multiple dysregulated metabolic pathways by GC-MS-based profiling of lung tissue in mice with PM2.5-induced asthma [J]. Chemosphere,2019,220:1-10. DOI:10.1016/j.chemosphere.2019. 06.063.
doi: 10.1016/j.chemosphere.2019. 06.063
4 David Rojas-Rueda,Martine Vrijheid,Oliver Robinson,et al. Environmental burden of childhood disease in Europe [J]. International Journal of Environmental Research and Public Health, 2019, 16(6):1084. DOI: 10.3390/ijerph16061084.
doi: 10.3390/ijerph16061084
5 Guo B, Wang X 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, 2021, 751:141765. DOI:10.1016/j.scitotenv. 2020.141765.
doi: 10.1016/j.scitotenv. 2020.141765
6 Zhang P, Ma W J, Wen F, et al. Estimating PM2.5 concentration using the machine learning GA-SVM method to improve the land use regression model in Shaanxi, China [J]. Ecotoxicology and Environmental Safety, 2021, 225:112772. DOI:10.1016/j.ecoenv.2021.112772.
doi: 10.1016/j.ecoenv.2021.112772
7 Wang Y, Yuan Q Q, Li T W, et al. Full-coverage spatiotemporal mapping of ambient PM2.5 and PM10 over China from Sentinel-5P and assimilated datasets: Considering the precursors and chemical compositions [J]. Science of the Total Environment,2021,793:148535. DOI: 10.1016/j.scitotenv.2021. 148535.
doi: 10.1016/j.scitotenv.2021. 148535
8 Alaa Mhawish , Tirthankar Banerjee , Meytar Sorek-Hamer , et al. Comparison and evaluation of MODIS Multi-angle Implementation of Atmospheric Correction (MAIAC) aerosol product over South Asia [J]. Remote Sensing of Environment, 2019, 224: 12-28. DOI:10.1016/j.rse.2019.01.033.
doi: 10.1016/j.rse.2019.01.033
9 Huang Keyong, Xiao Qingyang, Meng Xia, et al. Predicting monthly high-resolution PM2.5 concentrations with random forest model in the North China Plain [J]. Environmental Pollution,2018,242:675-683. DOI:10.1016/j.envpol.2018.07.016.
doi: 10.1016/j.envpol.2018.07.016
10 Daniel Goldberg, Pawan Gupta, Wang Kai, et al. Using gap-filled MAIAC AOD and WRF-Chem to estimate daily PM2.5 concentrations at 1 km resolution in the Eastern United States [J]. Atmospheric Environment, 2019, 199: 443-452.
11 Chen Chuchih, Wang Yinru, Yeh Hungyi, et al. Estimating monthly PM2.5 concentrations from satellite remote sensing data, meteorological variables, and land use data using ensemble statistical modeling and a random forest approach [J]. Environmental Pollution, 2021, 291:118159. DOI:10.1016/j.envpol.2021.118159.
doi: 10.1016/j.envpol.2021.118159
12 Jill E C, Christopher H, Basil C, et al. Qualitative and quantitative evaluation of MODIS satellite sensor data for regional and urban scale air quality [J]. Atmospheric Environment, 2004, 38(16): 2495-2509.
13 Wang Z F, Chen L F, Tao J H, et al. Satellite-based estimation of regional Particulate Matter (PM) in Beijing using vertical-and-RH correcting method [J]. Remote Sensing of Environment, 2010, 114(1): 50-63.
14 Liu Y, Paciorek Christopher and Koutrakis Petros. Estimating regional spatial and temporal variability of PM2.5 concentrations using satellite data, meteorology, and land use information [J]. Environmental Health Perspectives, 2009, 117(6): 886-892.
15 Lee H J, Liu Y, Brent C, et al. A novel calibration approach of MODIS AOD data to predict PM2.5 concentrations [J]. Atmospheric Chemistry and Physics, 2011, 11(15): 7991-8002.
16 Hu X F, Lance W, Al-Hamdan Mohammad, et al. Estimating ground-level PM2.5 concentrations in the southeastern US using geographically weighted regression [J]. Environmental Research,2013,121:1-10. DOI:10.1016/j.envres.2012. 11.003.
doi: 10.1016/j.envres.2012. 11.003
17 He Q Q, Huang B. Satellite-based mapping of daily high-resolution ground PM2.5 in China via space-time regression modeling [J]. Remote Sensing of Environment,2018,206:72-83. DOI:10.1016/j.rse.2017.12.018.
doi: 10.1016/j.rse.2017.12.018
18 Zhang Ying and Li Zhengqiang. Estimation of PM2.5 from fine-mode aerosol optical depth [J]. Journal of Remote Sensing, 2013,17(4):929-943.
18 张莹,李正强. 利用细模态气溶胶光学厚度估计PM2.5 [J]. 遥感学报, 2013,17(4):929-943.]
19 Drew M, Jessie C, Joseph T B, et al. A hybrid land use regression/AERMOD model for predicting intra-urban variation in PM2.5 [J]. Atmospheric Environment, 2016,131:307-315. DOI: 10.1016/j.atmosenv.2016.01.045.
doi: 10.1016/j.atmosenv.2016.01.045
20 Fang X, Zou B, Liu X P, et al. Satellite-based ground PM2.5 estimation using timely structure adaptive modeling [J]. Remote Sensing of Environment, 2016, 186: 152-163. DOI:10.1016/j.rse.2016.08.027.
doi: 10.1016/j.rse.2016.08.027
21 Yang L J, Xu H Q, Yu S D. Estimating PM2.5 concentrations in Yangtze River Delta region of China using random forest model and the Top-of-Atmosphere reflectance[J]. Journal of Environmental Management,2020,272:111061. DOI:10.1016/ j.jenvman.2020.111061.
doi: 10.1016/ j.jenvman.2020.111061
22 Pawan G, Sundar C. Particulate matter air quality assessment using integrated surface, satellite, and meteorological products: 2. a neural network approach [J]. Journal of Geophysical Research-Atmospheres, 2009, 114:D20205. DOI: 10.1029/2008JD011497.
doi: 10.1029/2008JD011497
23 Vahid M, David S, Mahsa M, et al. Comparing different methods for statistical modeling of particulate matter in Tehran, Iran [J]. Air Quality Atmosphere and Health, 2018, 11(10): 1155-1165.
24 Hu X F, Jessica B, Meng X, et al. Estimating PM2.5 Concentrations in the Conterminous United States Using the random forest approach [J]. Environmental Science & Technology, 2017, 51(12): 6936-6944.
25 Cole B, Roman J, Monir H, et al. Predicting daily urban fine particulate matter concentrations using a random forest model [J]. Environmental Science & Technology, 2018, 52(7): 4173-4179.
26 Iman K, Kazem A S. A random forest-based framework for crop mapping using temporal, spectral, textural and polarimetric observations [J]. International Journal of Remote Sensing, 2019, 40(18): 7221-7251.
27 Chai T F, Draxler R. Root Mean Square Error(RMSE) or Mean Absolute Error(MAE)? -arguments against avoiding RMSE in the literature [J]. Geoscientific Model Development, 2014, 7(3): 1247-1250.
28 Cort W, Kenji M. Advantages of the Mean Absolute Error(MAE) over the Root Mean Square Error(RMSE) in assessing average model performance [J]. Climate Research, 2005, 30(1): 79-82.
[1] 尹燕旻,贾立. 基于Sentinel-2的闪电河流域农作物分类研究[J]. 遥感技术与应用, 2021, 36(2): 400-410.
[2] 杨瑞,祁元,苏阳. 深度学习U-Net方法及其在高分辨卫星影像分类中的应用[J]. 遥感技术与应用, 2020, 35(4): 767-774.
[3] 韩天,潘小多,王旭峰,黄广辉,韦海宁. 遥感资料在WRF-Chem沙尘模拟中的应用[J]. 遥感技术与应用, 2020, 35(4): 808-819.
[4] 李超,李雪梅,田亚林,任瑞. 温度植被干旱指数时空融合模型对比[J]. 遥感技术与应用, 2020, 35(4): 832-844.
[5] 李净,温松楠. 基于3种机器学习法的太阳辐射模拟研究[J]. 遥感技术与应用, 2020, 35(3): 615-622.
[6] 董世元,张文娟,许君一,马建行. ESTARFM相似像元选取方法的改进研究[J]. 遥感技术与应用, 2020, 35(1): 185-193.
[7] 徐凡, 张雪红, 石玉立. 基于激光雷达和航拍影像的城市地物分类研究[J]. 遥感技术与应用, 2019, 34(2): 253-262.
[8] 杨涛,黄法融,李倩,白磊,李兰海. 新疆北部植被生长季NDVI时空变化及其与冬季降雪的关系[J]. 遥感技术与应用, 2017, 32(6): 1132-1140.
[9] 任静,潘小多. 基于WRF模式的多源遥感降水资料数据同化研究[J]. 遥感技术与应用, 2017, 32(4): 593-605.
[10] 姜萍,刘修国,陈启浩,邵芳芳. 利用多尺度SVM-CRF模型的极化SAR图像建筑物提取[J]. 遥感技术与应用, 2017, 32(3): 475-482.
[11] 赵彦博,曹学诚,南卓铜,吴小波. 基于开源GIS的DHSVM模型河网数据自动制备方法应用研究[J]. 遥感技术与应用, 2016, 31(4): 793-800.
[12] 陈阳,范建容,张云,李胜,甘泉,应国伟,曹伟超. 多云雾地区高时空分辨率植被覆盖度构建方法研究[J]. 遥感技术与应用, 2016, 31(3): 518-529.
[13] 林志垒,晏路明. 高光谱影像的BDT-SVM地物分类算法与应用[J]. 遥感技术与应用, 2016, 31(1): 177-185.
[14] 康峻,王力,牛铮,高帅,邬明权 . 基于局部空间自相关分析的时空数据融合[J]. 遥感技术与应用, 2015, 30(6): 1176-1181.
[15] 刁宁辉,孙从容,崔倩,吴奎桥,张为良,郝轶萌. RFSCAT散射计数据地理定位与误差分析[J]. 遥感技术与应用, 2015, 30(3): 518-526.