This study utilized Ensemble Kalman Filter (EnKF) and Nested Air Quality Prediction Simulation System (NAQPMS) to assimilate over 1000 surface air quality monitoring points from China National Environmental Monitoring Center (CNEMC), and established a high-resolution China Air Quality Reanalysis Dataset (CAQRA) for six years. Provided high spatial resolution (15 kilometers) of six conventional air pollutants (i.e. PM2.5, PM10, SO2, NO2, CO and O3) in China from 2013 to 2018 × A surface field with a resolution of 15 kilometers and high temporal resolution. This dataset will be the first high-resolution air quality reanalysis dataset in China that can simultaneously provide surface concentrations of six conventional air pollutants. It is of great value for multiple studies such as air pollution health impact assessment, research on changes in air quality in China, and providing training data for statistical or artificial intelligence (AI) based forecasting.
Ground observation data on PM2.5, PM10, sulfur dioxide, nitrogen dioxide, carbon monoxide, and ozone concentrations obtained from the China Environmental Monitoring Station
The dataset is generated by ChemDAS, which uses the NAQPMS model as the prediction model, and LETKF uses a post-processing model to assimilate observation data. The background error covariance is calculated by set simulation, taking into account the uncertainty of major air pollutant emissions. In addition, an inflation technique was employed to dynamically inflate the background error to prevent underestimation of the true background error covariance.
Five fold cross validation (CV) method was used to evaluate the quality of CAQRA. The cross validation results indicate that CAQRA performs well in reproducing the size and variability of surface air pollutants in China, and also reflects the interannual changes in air quality in China
By comparing with the Copernican Atmospheric Monitoring Service Reanalysis (CAMSRA) developed by the European Centre for Medium Range Weather Forecasts (ECWMF) based on satellite product assimilation, we found that due to the assimilation of ground observation data, CAQRA has higher accuracy in representing surface gaseous air pollutants in China. The finer horizontal resolution of CAQRA also makes it more suitable for regional scale air quality research
Further validation was conducted between the PM2.5 reanalysis dataset and the independent dataset of the US State Department's China Air Quality Monitoring Program, and the accuracy of PM2.5 reanalysis was compared with satellite estimated PM2.5 concentrations. The results indicate that there is good consistency between the reanalysis data of PM2.5 and independent observation data (R2=0.74-0.86, RMSE=16.8-33.6) μ G/m3), its accuracy is higher than most satellite estimates.
This work is licensed under
CC BY 4.0 (Creative Commons Attribution 4.0 International License).
https://cstr.cn/31253.11.sciencedb.00092
10.11922/sciencedb.00092
