High spatial resolution estimating of exposure to particulate matter 2.5 (PM2.5) is currently very limited in China. This study uses the newly released nationwide, hourly PM2.5 concentrations to create a nationwide, geographically weighted regression (GWR) model to estimate ground-level PM2.5 concentrations in China. A3 km resolution aerosol optical depth (AOD) product from MODIS is used as the primary predictor. Fire emissions detected by MODIS fire count were considered in the model development process. Additionally, meteorological features were used as covariates in the model to improve the estimation of ground-level PM2.5 concentrations. The model performed well and explained 81% of the daily PM2.5 concentration variations in model predictions, and the cross validations R² is 0.79. The cross-validated root mean squared error (RMSE) of the model was 18.6 μg/m³.Annual PM2.5 concentrations retrieved by the MODIS 3 km AOD product indicated that most of the residential community areas exceeded the new annual Chinese PM2.5 National Standard level 2. Estimated high-resolution national-scale daily PM2.5 maps are useful to identify severe air pollution episodes and determine health risk assessments. These results suggest that this approach is useful for estimating large-scale ground-level PM2.5 distributions, especially for regions without PM monitoring sites.

Emission inventories are essential for modelling studies and pollution control, but traditional emission inventories are usually updated after a few years based on the statistics of “bottom-up” approach from the energy consumption in provinces, cities, and counties. The latest emission inventories of multi-resolution emission inventory in China (MEIC) was compiled from the statistics for the year 2016 (MEIC_2016). However, the real emissions have varied yearly, due to national pollution control policies and accidental special events, such as the coronavirus disease (COVID-19) pandemic. In this study, a four-dimensional variational assimilation (4DVAR) system based on the “top-down” approach was developed to optimise sulfur dioxide (SO₂) emissions by assimilating the data of SO₂ concentrations from surface observational stations. The 4DVAR system was then applied to obtain the SO₂ emissions during the early period of COVID-19 pandemic (from 17 January to 7 February 2020), and the same period in 2019 over China. The results showed that the average MEIC_2016, 2019, and 2020 emissions were 42.2×10⁶, 40.1×10⁶, and 36.4×10⁶ kg d⁻¹. The emissions in 2020 decreased by 9.2 % in relation to the COVID-19 lockdown compared with those in 2019. For central China, where the lockdown measures were quite strict, the mean 2020 emission decreased by 21.0 % compared with 2019 emissions. Three forecast experiments were conducted using the emissions of MEIC_2016, 2019, and 2020 to demonstrate the effects of optimised emissions. The root mean square error (RMSE) in the experiments using 2019 and 2020 emissions decreased by 28.1 % and 50.7 %, and the correlation coefficient increased by 89.5 % and 205.9 % compared with the experiment using MEIC_2016. For central China, the average RMSE in the experiments with 2019 and 2020 emissions decreased by 48.8 % and 77.0 %, and the average correlation coefficient increased by 44.3 % and 238.7 %, compared with the experiment using MEIC_2016 emissions. The results demonstrated that the 4DVAR system effectively optimised emissions to describe the actual changes in SO₂ emissions related to the COVID lockdown, and it can thus be used to improve the accuracy of forecasts.

Precipitation nowcasting refers to the prediction of small-scale precipitation events at minute and kilometer scales within the upcoming 0 to 2 hours, which significantly impacts both human activities and daily life. However, prevailing deep learning models have primarily focused on a single radar echo data source, limiting their ability to effectively capture intricate and rapidly evolving precipitation patterns. Thus, meteorological satellite is considered to supplement radar echo data. To achieve a comprehensive integration of multi-source data with enhanced details, a two-stage fusion satellite and radar GAN-based prediction network (named FsrGAN) is proposed. In the first stage, we design a satellite-radar fusion prediction network known as FsrNet. This network employs an encoder-fusion-decoder architecture, where a novel spatial-channel attention (SCA) is proposed to enhance the filtering and fusion of multi-source and multi-scale features. In the second stage, we introduce a GAN-based network (FusionGAN) that also mines the complementary information of satellite images to sharpen the first-stage predicted radar maps with more details. Experiments are conducted on meteorological dataset in the Yangtze River Delta (YRD) region. The test results exhibit the notably superior performance of our model in terms of image quality and precipitation forecasting metrics in comparison to traditional optical flow-based methods and some well-known deep learning methods (ConvLSTM, ConvGRU, TrajGRU and PredRNN++). More importantly, our fusion model using satellite and radar data demonstrates the ability to predict convective initiation.

The authors developed a three-dimensional variational (3-DVAR) aerosol extinction coefficient (AEC) and aerosol mass concentration (AMC) data assimilation (DA) system for aerosol variables in the Weather Research and Forecasting–Chemistry (WRF–Chem) model with the WRF–Chem using the Model for Simulating Aerosol Interactions and Chemistry (MOSAIC) scheme. They establish an AEC observation operator and its corresponding adjoint based on the Interagency Monitoring of Protected Visual Environments (IMPROVE) equation and investigate the use of lidar AEC and surface AMC DA to forecast mass concentration (MC) profiles of PM_2.5 (particulate matter with an aerodynamic diameter of less than 2.5 µm) across China. Two sets of data were assimilated: AEC profiles captured by five conventional Mie scattering lidars (positioned in Beijing, Shijiazhuang, Taiyuan, Xuzhou, and Wuhu) and PM_2.5 and PM₁₀ MC data obtained from over 1500 ground environmental monitoring stations across China. Three DA experiments (i.e., a PM_2.5 (PM₁₀) DA experiment, a lidar AEC DA experiment, and a simultaneous PM_2.5 (PM₁₀) and lidar AEC DA experiment) with a 12 h assimilation period and a 24 h forecast period were conducted. The PM_2.5 (PM₁₀) DA reduced the root mean square error (RMSE) of the surface PM_2.5 MC in the initial field of the model by 38.6 µg m⁻³ (64.8 %). When lidar AEC data were assimilated, this reduction was 10.5 µg m⁻³ (17.6 %), and a 38.4 µg m⁻³ (64.4 %) reduction occurred when the two data sets were assimilated simultaneously, although only five lidars were available within the simulation region (approximately 2.33 million km² in size). The RMSEs of the forecasted surface PM_2.5 MC 24 h after the DA period in the three DA experiments were reduced by 6.1 µg m⁻³ (11.8 %), 1.5 µg m⁻³ (2.9 %), and 6.5 µg m⁻³ (12.6 %), respectively, indicating that the assimilation and hence the optimization of the initial field have a positive effect on the PM_2.5 MC forecast performance over a period of 24 h after the DA period.