Estimation of ultrahigh resolution PM2.5 concentrations in urban areas using 160 m Gaofen-1 AOD retrievals

Zhang, Tianhao1; Zhu, Zhongmin1,2; Gong, Wei1,3; Zhu, Zerun1; Sun, Kun4; Wang, Lunche5; Huang, Yusi1; Mao, Feiyue1,3,6; Shen, Huanfeng1,3,7; Li, Zhiwei7; Xu, Kai1

2018-10

10.1016/j.rse.2018.06.030

REMOTE SENSING OF ENVIRONMENT   影响因子和分区

0034-4257

1879-0704

Satellite-derived aerosol optical depth (AOD) has been widely used to estimate ground-level PM2.5 concentrations due to its spatially continuous observation. However, the coarse spatial resolutions (e.g., 3 km, 6 km, or 10 km) of the primary satellite AOD products have weakness to capture the characteristics of urban-scale PM2.5 patterns. Moreover, high-resolution (e.g., 1 km) PM2.5 estimations are still unable to be related to the urban landscape or to small geographical units, which is crucial for analyzing the urban pollution structure. In this study, the daily PM2.5 concentrations were estimated using the new AOD data with a 160 m spatial resolution retrieved by the Gaofen-1 (GF) wide field of view (WFV) along with the nested linear mixed effects model and ancillary variables from the Weather Research & Forecasting (WRF) meteorological simulation data. The experiment was conducted in Wuhan, Beijing, and Shanghai, which suffers from severe atmospheric fine particle pollution in recent years. The proposed model performed well for both GF and Moderate Resolution Imaging Spectroradiometer (MODIS), with slight over-fitting and little spatial autocorrelation. Regarding to the GF PM2.5 estimation, model fitting yielded R-2 values of 0.96, 0.91 and 0.95 and mean prediction error (MPE) of 10.13, 11.89 and 7.34 mu g/m(3) for Wuhan, Beijing and Shanghai, respectively. The site-based cross validation achieved R-2 values of 0.92, 0.88 and 0.89, and MPE of 13.69, 16.76 and 12.59 mu g/m(3) for Wuhan, Beijing and Shanghai, respectively. The day-of-years based cross validation resulted in R-2 of 0.54, 0.58 and 0.50, and MPE of 30.46, 27.12 and 31.58 mu g/m(3) for Wuhan, Beijing and Shanghai, respectively, indicating that it was practicable to estimate the GF PM2.5 in the days without enough AOD-PM2.5 matchups. The ultrahigh resolution PM2.5 estimations offer substantial advantages for providing finer spatially resolved PM2.5 trends. Additionally, it offers new approaches to locate main PM2.5 emission sources, evaluate the local PM2.5 contribution proportion, and quantify the daily PM2.5 emission level via remote sensing techniques. Along with the joint observations via other high-resolution satellites, the temporal resolution of GF PM(2.5)will be further improved. In all, this study not only provides possibilities for further applications in the precise analysis of urban inner PM2.5 pollution patterns but also establishes a foundation for constructing a high-resolution satellite air monitoring network in China.

Aerosol optical depth Ultrahigh spatial resolution Gaofen-1 WFV Nested LME model Urban-scale PM2.5

SCI ; EI

英语

通讯

China Postdoctoral Science Foundation[2015M572198]

Environmental Sciences & Ecology ; Remote Sensing ; Imaging Science & Photographic Technology

Environmental Sciences ; Remote Sensing ; Imaging Science & Photographic Technology

WOS:000445990100007

ELSEVIER SCIENCE INC

GEOSCIENCES

Web of Science

1.Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan 430079, Hubei, Peoples R China
2.Wuchang Shouyi Univ, Coll Informat Sci & Engn, Wuhan 430064, Hubei, Peoples R China
3.Collaborat Innovat Ctr Geospatial Technol, Wuhan 430079, Hubei, Peoples R China
4.Southern Univ Sci & Technol, Sch Environm Sci & Engn, Shenzhen 518055, Peoples R China
5.China Univ Geosci, Sch Earth Sci, Lab Crit Zone Evolut, Wuhan 430074, Hubei, Peoples R China
6.Wuhan Univ, Sch Remote Sensing & Informat Engn, Wuhan 430079, Hubei, Peoples R China
7.Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Hubei, Peoples R China

Zhang, Tianhao,Zhu, Zhongmin,Gong, Wei,et al. Estimation of ultrahigh resolution PM2.5 concentrations in urban areas using 160 m Gaofen-1 AOD retrievals[J]. REMOTE SENSING OF ENVIRONMENT,2018,216:91-104.

Zhang, Tianhao.,Zhu, Zhongmin.,Gong, Wei.,Zhu, Zerun.,Sun, Kun.,...&Xu, Kai.(2018).Estimation of ultrahigh resolution PM2.5 concentrations in urban areas using 160 m Gaofen-1 AOD retrievals.REMOTE SENSING OF ENVIRONMENT,216,91-104.

Zhang, Tianhao,et al."Estimation of ultrahigh resolution PM2.5 concentrations in urban areas using 160 m Gaofen-1 AOD retrievals".REMOTE SENSING OF ENVIRONMENT 216(2018):91-104.