With the completion of the Beijing Five-year Clean Air Action Plan by the end of 2017, the annual mean PM_2.5 concentration in Beijing dropped dramatically to 58.0 µg m⁻³ in 2017 from 89.5 µg m⁻³ in 2013. However, controversies exist to argue that favourable meteorological conditions in 2017 were the major driver for such a rapid decrease in PM_2.5 concentrations. To comprehensively evaluate this 5-year plan, we employed a Kolmogorov–Zurbenko (KZ) filter and WRF-CMAQ (Weather Research and Forecasting and the Community Multi-scale Air Quality) to quantify the relative contribution of meteorological conditions and the control of anthropogenic emissions to PM_2.5 reduction in Beijing from 2013 to 2017. For these 5 years, the relative contribution of emission reduction to the decrease in PM_2.5 concentrations calculated by KZ filtering and WRF-CMAQ was 80.6 % and 78.6 % respectively. KZ filtering suggested that short-term variations in meteorological and emission conditions contributed majorly to rapid changes in PM_2.5 concentrations in Beijing. WRF-CMAQ revealed that the relative contribution of local and regional emission reduction to the PM_2.5 decrease in Beijing was 53.7 % and 24.9 % respectively. For local emission-reduction measures, the regulation of coal boilers, increasing use of clean fuels for residential use and industrial restructuring contributed 20.1 %, 17.4 % and 10.8 % to PM_2.5 reduction respectively. Both models suggested that the control of anthropogenic emissions accounted for around 80 % of the PM_2.5 reduction in Beijing, indicating that emission reduction was crucial for air quality enhancement in Beijing from 2013 to 2017. Consequently, such a long-term air quality clean plan should be continued in the following years to further reduce PM_2.5 concentrations in Beijing.

Black soil is fertile, abundant with organic matter (OM) and is exceptional for farming. The black soil zone in northeast China is the third-largest black soil zone globally and produces a quarter of China’s commodity grain. However, the soil organic matter (SOM) in this zone is declining, and the quality of cultivated land is falling off rapidly due to overexploitation and unsustainable management practices. To help develop an integrated protection strategy for black soil, this study aimed to identify the primary factors contributing to SOM degradation. The geographic detector, which can detect both linear and nonlinear relationships and the interactions based on spatial heterogeneous patterns, was used to quantitatively analyze the natural and anthropogenic factors affecting SOM concentration in northeast China. In descending order, the nine factors affecting SOM are temperature, gross domestic product (GDP), elevation, population, soil type, precipitation, soil erosion, land use, and geomorphology. The influence of all factors is significant, and the interaction of any two factors enhances their impact. The SOM concentration decreases with increased temperature, population, soil erosion, elevation and terrain undulation. SOM rises with increased precipitation, initially decreases with increasing GDP but then increases, and varies by soil type and land use. Conclusions about detailed impacts are presented in this paper. For example, wind erosion has a more significant effect than water erosion, and irrigated land has a lower SOM content than dry land. Based on the study results, protection measures, including conservation tillage, farmland shelterbelts, cross-slope ridges, terraces, and rainfed farming are recommended. The conversion of high-quality farmland to non-farm uses should be prohibited.

High-quality training samples are essential for accurate land cover classification. Due to the difficulties in collecting a large number of training samples, it is of great significance to collect a high-quality sample dataset with a limited sample size but effective sample distribution. In this paper, we proposed an object-oriented sampling approach by segmenting image blocks expanded from systematically distributed seeds (object-oriented sampling approach) and carried out a rigorous comparison of seven sampling strategies, including random sampling, systematic sampling, stratified sampling (stratified sampling with the strata of land cover classes based on classification product, Latin hypercube sampling, and spatial Latin hypercube sampling), object-oriented sampling, and manual sampling, to explore the impact of training sample distribution on the accuracy of land cover classification when the samples are limited. Five study areas from different climate zones were selected along the China–Mongolia border. Our research identified the proposed object-oriented sampling approach as the first-choice sampling strategy in collecting training samples. This approach improved the diversity and completeness of the training sample set. Stratified sampling with strata defined by the combination of different attributes and stratified sampling with the strata of land cover classes had their limitations, and they performed well in specific situations when we have enough prior knowledge or high-accuracy product. Manual sampling was greatly influenced by the experience of interpreters. All these sampling strategies mentioned above outperformed random sampling and systematic sampling in this study. The results indicate that the sampling strategies of training datasets do have great impacts on the land cover classification accuracies when the sample size is limited. This paper will provide guidance for efficient training sample collection to increase classification accuracies.

Wildfires have a strong negative effect on the environment, ecology and public health. However, the potential degradation of mainstream global fire products leads to large uncertainty in the effective monitoring of wildfires and their influence. To fill this gap, we produced Fengyun-3D (FY-3D) global fire products with a similar spatial and temporal resolution, aiming to serve as an alternative to and continuity for Moderate Resolution Imaging Spectroradiometer (MODIS) global fire products. Firstly, the sensor parameters and major algorithms for noise detection and fire identification in FY-3D products were introduced. For visual-check-based accuracy assessment, five typical regions with a large number of fire spots across the globe, Africa, South America, the Indochinese Peninsula, Siberia and Australia, were selected, and the overall accuracy exceeded 94 %. Meanwhile, the consistence between FY-3D and MODIS fire products was examined. The result suggested that the overall consistence was 84.4 %, with a fluctuation across seasons, surface types and regions. The high accuracy and consistence with MODIS products proved that the FY-3D fire product is an ideal tool for global fire monitoring. Based on field-collected reference data, we further evaluated the suitability of FY-3D fire products in China. The overall accuracy and accuracy without considering omission errors were 79.43 % and 88.50 % higher, respectively, than those of MODIS fire products. Since detailed local geographical conditions were specifically considered, FY-3D products should be preferably employed for fire monitoring in China. The FY-3D fire dataset can be downloaded at http://satellite.nsmc.org.cn/portalsite/default.aspx (NSMC, 2021) or at http://figshare.com (last access: 10 January 2021) with the following identifier DOI: https://doi.org/10.6084/m9.figshare.20102210 (Chen et al., 2022).

Selecting samples with high representativeness is very important in the accuracy assessment of land use and land cover (LULC) classification. By optimizing sampling sites in both feature space and geographical space, this paper presented a stratified even sampling method for accuracy assessment of the GlobeLand30 dataset for Beijing, China. In this method, spatial stratification and area-weighted proportion sampling assignment method were adopted to achieve optimal coverage in feature space; spatial-simulated annealing (SSA) and the minimization of the mean of the shortest distances (MMSD) criterion were used to optimize sampling sites in geographical space. Six sample sets with sizes of 150, 300, 450, 600, 750 and 900 were drawn using the proposed method, spatial even sampling method, stratified random sampling method and simple random sampling method, and their overall accuracy (OA), root-mean-square error (RMSE) and standard deviation (STDEV) values were evaluated. The results suggested that the OA, RMSE and STDEV results of the proposed method were 71.36%-73.91% (Mean, 72.26%), 0.90% and 0.96%, respectively. Compared with the other sampling methods, the average OA of the proposed method was much closer to the true OA and corresponding RMSE and STDEV were much lower than the other three sampling methods, respectively. It can improve the representativeness of both feature and geographical space, and provide a robust operational tool for the validation of LULC datasets.