A Six-year long High-resolution Air Quality Reanalysis Dataset over China from 2013 to 2018

Xiao Tang Xiao Tang,Zifa Wang Zifa Wang,Miaomiao Lu Miaomiao Lu,Yuepeng Pan Yuepeng Pan,Haitao Zheng Haitao Zheng,Jianjun Li Jianjun Li,Lin Wu Lin Wu,Fei Li Fei Li,Wei Wang Wei Wang,Tao Song Tao Song,Duohong Chen Duohong Chen,Lei Kong Lei Kong,Huansheng Chen Huansheng Chen,Guanglin Jia Guanglin Jia,Qizhong Wu Qizhong Wu,Qian Wang Qian Wang,Lili Zhu Lili Zhu,Gregory R Carmichael Gregory R Carmichael,Bing Liu Bing Liu,Huangjian Wu Huangjian Wu,Jiang Zhu Jiang Zhu

doi:10.11922/sciencedb.00053

Xiao Tang Xiao Tang, Zifa Wang Zifa Wang + Show 19 more

https://doi.org/10.11922/sciencedb.00053

Copy DOI

Export

Save

Cite

Publication Date: Jul 19, 2022

Citations: 8

Abstract
Full-Text
Similar Papers

Abstract

Listen

Main content and developer: The Chinese Air Quality Reanalysis dataset was produced by the Institute of Atmospheric Physics, Chinese Academy of Sciences (IAP/CAS), collaborated with China National Environmental Monitoring Centre (CNEMC) and other research institutes. It provides the surface gridded fields of six conventional air pollutants (i.e. PM2.5，PM10，SO2，NO2，CO and O3) and the simulated surface fields of wind speed (u, v), pressure (psfc), relative humidity (RH) and temperature (temp) by WRF model. The spatial and temporal resolutions are respectively 15km and 1 hour. Currently, the time period of the dataset is from 2013 to 2019, and will be updated irregularly.Data assimilation method: The dataset was produced by the chemical data assimilation system (ChemDAS) developed by IAP, CAS, which assimilates over 1000 surface air quality monitoring sites from CNEMC based on the ensemble Kalman filter (EnKF) and the Nested Air Quality Prediction Modeling system (NAQPMS). This method broke through the problems of instability, insufficient adjustment and negative assimilation effect in atmospheric chemistry data assimilation and develops the multi-air pollutant collaborative assimilation including the monitoring data automatic quality control methods, adaptive mode error estimation and other advanced algorithms. It has been published in Earth System Science Data, where detailed descriptions and validation of this dataset are available (https://doi.org/10.5194/essd-13-529-2021).Data accuracy: The dataset was evaluated by cross-validation and independent data validation. 2013-2018: The calculated root of mean square error (RMSE) at assimilation (validation) sites for hourly concentrations were estimated to be 15.2 (21.3) μg/m3 for PM2.5, 28.0 (39.3) μg/m3 for PM10, 16.9 (24.9) μg/m3 for SO2, 12.7 (16.4) μg/m3 for NO2, 0.38 (0.54) mg/m3 for CO and 17.5 (21.9) μg/m3 for O3. 2019: The calculated root of mean square error (RMSE) at assimilation (validation) sites for hourly concentrations were estimated to be 10.2 (13.3) μg/m3 for PM2.5, 19.1 (24.5) μg/m3 for PM10, 6.1 (7.7) μg/m3 for SO2, 10.0(12.4) μg/m3 for NO2, 0.24 (0.30) mg/m3 for CO and 14.0 (17.2) μg/m3 for O3.Dataset’s versions : The first version of this datasets (V1) is from 2013 to 2018 including 72 zip files and each zip file contains one month of reanalysis data. The second version of this datasets (V2) is from 2013 to 2018, splitting by days and in all 2191 zip files. The third version of this datasets (V3) was extended to 2019 based on the same algorithm and validation as the V1 and V2, including seven folders in a year. Each folder contains the reanalysis data compression files in days. The description on the content of each data file is available in README.txt.

Full Text