基于随机森林的高分辨率PM<sub>2.5</sub>浓度时空变化模拟——以中原城市群核心区为例

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (3069 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要为探究城市尺度或更小区域内近地面PM_2.5浓度模拟方法,以空气监测站实测的近地面PM_2.5浓度为因变量、MODIS L1B卫星影像中计算得出的大气顶部反射率( TOA)数据和直接提取的图像特征值(Image)为2组主要自变量,融合气象特征、地形特征和时空特征等辅助因子,构建RF(PM_2.5~TOA)和RF(PM_2.5~Image)2个随机森林模型,并最终选用RF(PM_2.5~Image)对中原城市群核心区2020年250m空间分辨率下的近地面PM_2.5浓度进行评估.结果表明:2个模型交叉验证的决定系数(R²)均为0.93,均方根误差(RMSE)分别为9.23, 8.28μg/m³,模型性能良好;当2个模型达到相近拟合度时,RF(PM_2.5~Image)的预测偏差更低,且空间分辨率高于250m的特征重要性占比高达44.2%,能够更准确地描述250m空间分辨率的近地面PM_2.5浓度变化;中原城市群核心区2020年年均PM_2.5浓度为53.80μg/m³,污染现象严重,不同单元内发生0~8次污染过程不等,整体上呈现由西南西北山区向东部平原递增的空间分异特征和冬高夏低、春秋过渡的时间变化规律.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	卢鋆镆
	曾穗平
	曾坚
	王森
	宋苑震

关键词 ： PM_2.5, 随机森林, MODIS L1B, 250m, 中原城市群核心区

Abstract：In order to explore the estimation method of near-surface PM_2.5 concentration in urban scale or smaller area, two random forest models, RF (PM_2.5~TOA) and RF (PM_2.5~Image), were constructed by taking the PM_2.5 concentration measured by the air monitoring station as the dependent variable, taking the top of atmosphere reflectance data (TOA) calculated from MODIS L1B satellite images and the eigenvalues directly extracted from the images (Image) as two groups of main independent variables, and integrating the auxiliary factors of meteorological features, topographic features and temporal and spatial features. Finally, RF (PM_2.5~Image) was selected to estimate the monthly mean concentration of near-surface PM_2.5 in the core area of Central Plains Urban Agglomeration under the spatial resolution of 250m in 2020. The results showed that: the coefficient of determination (R²) of the two models were 0.93, and the root mean square prediction errors ( RMSE) were 9.23, 8.28μg/m³, respectively. When the two models reached a similar degree of R², a lower prediction deviation was found in RF (PM_2.5~Image), and the importance of features with spatial resolution in it higher than 250m accounts for up to 44.2%, which can more accurately describe the change of near-surface PM_2.5 concentration with a spatial resolution of 250m. The average annual concentration of PM_2.5 in the core area of the Central Plains urban agglomeration in 2020 was 53.80μg/m³, and the pollution phenomenon was serious, with 0~8 pollution processes occurring in different units. On the whole, there was a spatial differentiation characteristic of increasing from southwest and northwest mountainous areas to eastern plain and a time variation pattern of high in winter, low in summer and transition in spring and autumn.

Key words： PM_2.5 random forest MODIS L1B 250m Central Plains Urban Agglomeration Core Area

收稿日期: 2022-11-03

PACS:

X513

基金资助:国家自然科学基金资助项目(52078320)

通讯作者: 曾穗平,副教授,176116126@qq.com E-mail: 176116126@qq.com

作者简介: 卢鋆镆(1997-),女,河北保定人,天津大学硕士,主要从事大气污染治理与城乡规划相关研究.发表论文1篇.lujunmo@tju.edu.cn

引用本文:

卢鋆镆, 曾穗平, 曾坚, 王森, 宋苑震. 基于随机森林的高分辨率PM_2.5浓度时空变化模拟——以中原城市群核心区为例[J]. 中国环境科学, 2023, 43(7): 3299-3311. LU Jun-mo, ZENG Sui-ping, ZENG Jian, WANG Sen, SONG Yuan-zhen. High resolution simulation of temporal and spatial variation of PM_2.5 concentration based on random forest——A case study of Central Plains Urban Agglomeration Core Area. CHINA ENVIRONMENTAL SCIENCECE, 2023, 43(7): 3299-3311.

链接本文:

http://www.zghjkx.com.cn/CN/ 或 http://www.zghjkx.com.cn/CN/Y2023/V43/I7/3299

[1]	Gupta P, Christopher S A, Wang J, et al. Satellite remote sensing of particulate matter and air quality assessment over global cities [J]. Atmospheric Environment, 2006,40(30):5880-5892.
[2]	李忠宾,王楠,张自力,等.中国地区MODIS气溶胶光学厚度产品综合验证及分析 [J]. 中国环境科学, 2020,40(10):4190-4204. Li Z B, Wang N, Zhang Z L, et al. Validation and analyzation of MODIS aerosol optical depth products over China [J]. China Environmental Science. 2020,40(10):4190-4204.
[3]	李婷苑,谭浩波,王春林,等.卫星遥感AOD反演地面细颗粒物浓度方法与效果 [J]. 中国环境科学, 2020,40(1):13-23. Li T Y, Tan H B, Wang C L, et al. The method and the corresponding effect of ground fine partical concentration retrieved by satellite remote sensing AOD [J]. China Environmental Science. 2020,40(1): 13-23.
[4]	Engel-Cox J A, Holloman C H, Coutant B W, 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.
[5]	Wang Z, Chen L, Tao J, 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.
[6]	贾松林,苏林,陶金花,等.卫星遥感监测近地表细颗粒物多元回归方法研究 [J]. 中国环境科学, 2014,34(3):565-573. Jia S L, Su L, Tao J H, et al. A study of multiple regression method for estimating concentration of fine particulate matter using satellite remote sensing [J]. China Environmental Science, 2014,34(3):565-573.
[7]	周爽,王春林,孙睿,等.基于LME/BME的珠江三角洲PM2.5星地融合技术研究 [J]. 中国环境科学, 2019,39(5):1869-1878. Zhou S, Wang C L, Sun R, et al. Fusion of satellite data and ground observed PM2.5 in Pearl River Delta region with linear mixed effect and Bayesian maximum entropy method [J]. China Environmental Science, 2019,39(5):1869-1878.
[8]	宋春杰,魏强,范丽行,等.基于PM2.5站点监测数据的京津冀AOD补值研究 [J]. 中国环境科学, 2022,42(7):3000-3012. Song C J, Wei Q, Fan L X, et al. Filling the missing data of AOD using the situ PM2.5 monitoring measurements in the Beijing-Tianjin-Hebei region [J]. China Environmental Science, 2022,42(7):3000-3012.
[9]	付宏臣,孙艳玲,王斌,等.基于AOD数据和GWR模型估算京津冀地区PM2.5浓度 [J]. 中国环境科学, 2019,39(11):4530-4537. Fu H C, Sun Y L, Wang B, et al. Estimation of PM2.5 concentration in Beijing-Tianjin-Hebei region based on AOD data and GWR model [J]. China Environmental Science, 2019,39(11):4530-4537.
[10]	Bai Y, Wu L, Qin K, et al. A geographically and temporally weighted regression model for ground-level PM2.5 estimation from satellite-derived 500m resolution AOD [J]. Remote Sensing, 2016,8(3):262.
[11]	He 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.
[12]	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.
[13]	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.
[14]	刘基伟,闵素芹,金梦迪.基于分布式感知深度神经网络的高分辨率PM2.5值估算 [J]. 地理学报, 2021,76(1):191-205. Liu J W, Min S Q, Jin M D. High resolution PM2.5 estimation based on the distributed perception deep neural network model [J]. Acta Geographica Sinica, 2021,76(1):191-205.
[15]	杨晓辉,宋春杰,范丽行,等.京津冀地区高分辨率PM2.5浓度时空变化模拟与分析 [J]. 环境科学, 2021,42(9):4083-4094. Yang X H, Song C J, Fan L X, et al. High-resolution estimation of spatio-temporal variation in PM2.5 concentrations in the Beijing-Tianjin-Hebei Region [J]. Environmental Science, 2021,42(9):4083-4094.
[16]	范丽行,杨晓辉,宋春杰,等.基于时空混合效应模型的京津冀PM2.5浓度变化模拟 [J]. 环境科学, 2022,43(5):2262-2273. Fan L X, Yang X H, Song C J, et al. Modeling of PM2.5 concentrations in the Beijing-Tianjin-Hebei Region using a spacetime linear mixed effects model [J]. Environmental Science, 2022,43(5):2262-2273.
[17]	李卓建,赵尚民,郭鹏程.太原市城区PM2.5浓度时空分布特征研究 [J]. 环境污染与防治, 2021,43(3):353-358. Li Z J, Zhao S M, Guo P C. Study on the spatial and temporal characteristics of PM2.5 concentration in urban Taiyuan [J]. Environmental Pollution & Control, 2021,43(3):353-358.
[18]	周琪,于洋,刘苗苗,等.基于机器学习和非参数估计的城市PM2.5风险评估及特征识别 [J]. 中国环境科学, 2022,42(8):3554-3560. Zhou Q, Yu Y, Liu M M, et al. Risk assessment and feature identification of PM2.5 pollution based on machine learning and nonparametric estimation [J]. China Environmental Science, 2022, 42(8):3554-3560.
[19]	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. Yang L J, Zhang J X, Lin M S. Research on methods of remotely sensed PM2.5 concentrations estimation in four provinces and one city along the East Coast of China [J]. Remote Sensing Technology and Application, 2021,36(6):1408-1415.
[20]	李建新,刘小生,刘静,等.基于MRMR-HK-SVM模型的PM2.5浓度预测 [J]. 中国环境科学, 2019,39(6):2304-2310. Li J X, Liu X S, Liu J, et al. Prediction of PM2.5 concentration based on MRMR-HK-SVM model [J]. China Environmental Science, 2019, 39(6):2304-2310.
[21]	Park Y, Kong B, Heo J, et al. Estimating PM2.5 concentration of the conterminous United States via interpretable convolutional neural networks [J]. Environmental Pollution, 2020,256:112295.
[22]	Wang B, Yuan Q, Yang Q, et al. Estimate hourly PM2.5 concentrations from Himawari-8TOA reflectance directly using geo-intelligent long short-term memory network [J]. Environmental Pollution, DOI:10. 1016/j.envpol.2020.116327.
[23]	梁泽,王玥瑶,岳远紊,等.耦合遗传算法与RBF神经网络的PM2.5浓度预测模型 [J]. 中国环境科学, 2020,40(2):523-529. Liang Z, Wang Y Y, Yue Y W, et al. A coupling model of genetic algorithm and RBF neural network for theprediction of PM2.5 concentration [J]. China Environmental Science, 2020,40(2):523-529.
[24]	耿冰,孙义博,曾巧林,等.基于深度学习方法的大气细颗粒物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 atmospheric fine particulate matter pollution based on deep learning method [J]. China Environmental Science, 2021,41(8):3502-3510.
[25]	康新礼,张文豪,刘原萍,等.基于随机森林的京津冀地区PM2.5遥感反演及变化分析 [J]. 遥感技术与应用, 2022,37(2):424-435. Kang X L, Zhang W H, Liu Y P, et al. PM2.5 remote sensing retrieval and change analysis in Beijing-Tianjin-Hebei Region based on random forest model [J]. Remote Sensing Technology and Application, 2022,37(2):424-435.
[26]	吴宇宏,杜宁,王莉,等.基于iLME+Geoi-RF模型的四川省PM2.5浓度估算 [J]. 环境科学, 2021,42(12):5602-5615. Wu Y H, Du N, Wang L, et al. Estimation of PM2.5 concentration in Sichuan based on improved linear mixed effect model and geo-intelligent random forest [J]. Environmental Science, 2021,42(12): 5602-5615.
[27]	汤宇磊,杨复沫,詹宇.四川盆地PM2.5与PM10高分辨率时空分布及关联分析 [J]. 中国环境科学, 2019,39(12):4950-4958. Tang Y L, Yang F M, Zhan Y. High resolution spatiotemporal distribution and correlation analysis of PM2.5 and PM10 concentrations in the Sichuan Basin [J]. China Environmental Science, 2019,39(12): 4950-4958.
[28]	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, DOI:10.1016/j.jenvman, 2020.111061.
[29]	Yan X, Zang Z, Luo N, et al. New interpretable deep learning model to monitor real-time PM2.5 concentrations from satellite data [J]. Environment International, 2020,144:106060.
[30]	Liu J J, Weng F Z, Li Z Q. Satellite-based PM2.5 estimation directly from reflectance at the top of the atmosphere using a machine learning algorithm [J]. Atmospheric Environment, DOI:10.1016/j.atmosenv, 2019.04.002.
[31]	Zheng T S, Bergin M H, Hu S J, et al. Estimating ground-level PM2.5 using micro-satellite images by a convolutional neural network and random forest approach [J]. Atmospheric Environment, 2020,230: 117451.
[32]	Imani M. Particulate matter (PM2.5 and PM10) generation map using MODIS Level-1satellite images and deep neural network [J]. Journal of Environmental Management, DOI:10.1016/j.jenvman.2020.111888.
[33]	Mao F, Hong J, Min Q, et al. Estimating hourly full-coverage PM2.5 over China based on TOA reflectance data from the Fengyun-4A satellite [J]. Environmental Pollution, 2021,270:116119.
[34]	Wei J, Li Z, Cribb M, et al. Improved 1km resolution PM2.5 estimates across China using enhanced space-time extremely randomized trees [J]. Atmospheric Chemistry and Physics, 2020,20(6):3273-3289.
[35]	Zhao C, Wang Q, Ban J, et al. Estimating the daily PM2.5 concentration in the Beijing-Tianjin-Hebei region using a random forest model with a 0.01°×0.01° spatial resolution [J]. Environment International, 2020, 134:105297.
[36]	Yao F, Wu J, Li W, et al. Estimating daily PM2.5 concentrations in Beijing using 750-M VIIRS IP AOD retrievals and a nested spatiotemporal statistical model [J]. Remote Sensing, 2019,11(7):841.
[37]	Zhang T, Zhu Z, Gong W, et al. Estimation of ultrahigh resolution PM2.5 concentrations in urban areas using 160m Gaofen-1AOD retrievals [J]. Remote Sensing of Environment, 2018,216:91-104.
[38]	Xie Y, Wang Y, Bilal M, et al. Mapping daily PM2.5 at 500m resolution over Beijing with improved hazy day performance [J]. Science of The Total Environment, 2019,659:410-418.
[39]	严莹婷,陆小曼,王嘉佳,等.基于GF-4卫星的长三角城市群PM2.5遥感反演 [J]. 中国环境科学, 2022,42(3):1005-1012. Yan Y T, Lu X M, Wang J J, et al. Remote Estimation of PM2.5 Based on Gaofen-4Satellite Data in the Yangtze River Delta Urban Agglomeration [J]. China Environmental Science, 2022,42(3):1005-1012.
[40]	Kaufman Y J, Tanré D, Remer L A, et al. Operational remote sensing of tropospheric aerosol over land from EOS moderate resolution imaging spectroradiometer [J]. Journal of Geophysical Research: Atmospheres, 1997,102(D14):17051-17067.
[41]	周广强,谢英,吴剑斌,等.基于WRF-Chem模式的华东区域PM2.5预报及偏差原因 [J]. 中国环境科学, 2016,36(8):2251-2259. Zhou G Q, Xie Y, Wu J B, et al. Based on WRF-Chem model, the PM2.5 forecast and deviation reason in East China region [J]. China Environmental Science, 2016,36(8):2251-2259.
[42]	徐发昭,李净,褚馨德,等.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 F Z, Li J, Chu X D, et al. Research on daily PM2.5 simulation based on MODIS and multiple machine learning algorithms [J]. China Environmental Science, 2022,42(6):2523-2529.
[43]	Hu X, Belle J H, 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.
[44]	杨立娟.基于两层随机森林模型估算中国东部沿海地区的PM2.5浓度 [J]. 国土资源遥感, 2020,32(4):137-144. Yang L J. Estimating PM2.5 concentrations in eastern coastal area of China using a two-stage random forest model [J]. Renmote Sensing for Land and Resources, 2020,32(4):137-144.
[45]	Jiang G, Wang W. Error estimation based on variance analysis of k-fold cross-validation [J]. Pattern Recognition, 2017,69:94-106.
[46]	GB 3095-2012 环境空气质量标准 [S]. GB 3095-2012 Environmental air quality standards [S].
[47]	AQG2021 全球空气质量准则 [S]. AQG2021 Global air quality guidelines [S].
[48]	Guo B, Zhang D, Pei L, et al. Estimating PM2.5 concentrations via random forest method using satellite, auxiliary, and ground-level station dataset at multiple temporal scales across China in 2017 [J]. Science of The Total Environment, 2021,778:146288.
[49]	谈小生,葛成辉.太阳角的计算方法及其在遥感中的应用 [J]. 国土资源遥感, 1995,(2):48-57. Tan X S, Ge C H. A method for calculating solar angles and its application on remote sensing [J]. Remote Sensing for Land and Resources, 1995,(2):48-57.
[50]	Yan X, Zang Z, Jiang Y, et al. A spatial-temporal interpretable deep learning model for improving interpretability and predictive accuracy of satellite-based PM2.5 [J]. Environmental Pollution, 2021,273: 116459.