Search

Extracting road network maps for high-resolution remote sensing images is a critical and challenging remote sensing topic, with significant importance for traffic navigation, disaster management, autonomous driving, and urban planning. Although deep learning has demonstrated its powerful feature extraction capabilities in image processing tasks, including road extraction, existing road extraction algorithms still have some limitations. In particular, semantic segmentation-based methods often lack supervision of the connection relationships between roads and topological correctness constraints, resulting in fragmented and poorly connected road network graphs. Moreover, graph-based methods lack effective supervision strategies for vertex misdetection issues. To deal with these issues, we propose a deeply supervised vertex network (DSVNet) for road network graph extraction. First, to effectively supervise road vertices, we design a road vertex supervision module that yields improved vertex prediction accuracy. Second, to merge the benefits of segmentation-based methods and graph-based methods, we establish a parallel semantic segmentation branch based on the vertex querying task, thereby achieving enhanced road extraction accuracy. Furthermore, we introduce deformable attention into the model to boost its performance and computational efficiency. We validate the effectiveness of DSVNet on two large-scale public datasets. A large number of experimental results show that our research approach achieves the best road detection performance.

Accurate crop mapping can represent the fundamental data for digital agriculture and ecological security. However, current crop classification methods perform poorly in mountainous areas with small cropland field parcel areas and multiple crops under cultivation. This study proposed a new object-oriented classification method to address this issue, using multi-source data and object features to achieve multi-crop classification in mountainous areas. Firstly, a deep learning method was employed to extract cropland field parcels in mountainous areas. Subsequently, the fusion of multi-source data was carried out based on cropland field parcels, while object features tailored for mountainous crops were designed for crop classification. Comparative analysis indicates that the proposed classification method demonstrates exceptional performance, enabling accurate mapping of various crops in mountainous regions. The F1 score and overall accuracy (OA) of the proposed method are 0.8449 and 0.8502, representing a 10% improvement over the pixel-based random forest classification results. Furthermore, qualitative analysis reveals that the proposed method exhibits higher classification accuracy for smaller plots and more precise delineation of crop boundaries. Finally, meticulous crop mapping of corn, sorghum, rice, and other crops in Xishui County, Guizhou Province, demonstrates the significant potential of the proposed method in crop classification within mountainous scenarios.

Cotton (Gossypium hirsutum L.) is one of the main crops in Uzbekistan, which makes a major contribution to the country’s economy. The cotton industry has played a pivotal role in the economic landscape of Uzbekistan for decades, generating employment opportunities and supporting the livelihoods of countless individuals across the country. Therefore, having precise and up-to-date data on cotton cultivation areas is crucial for overseeing and effectively managing cotton fields. Nonetheless, there is currently no extensive, high-resolution approach that is appropriate for mapping cotton fields on a large scale, and it is necessary to address the issues related to the absence of ground-truth data, inadequate resolution, and timeliness. In this study, we introduced an effective approach for automatically mapping cotton fields on a large scale. A crop-type mapping method based on phenology was conducted to map cotton fields across the country. This research affirms the significance of phenological metrics in enhancing the mapping of cotton fields during the growing season in Uzbekistan. We used an adaptive feature-fusion network for crop classification using single-temporal Sentinel-2 images and automatically generated samples. The map achieved an overall accuracy (OA) of 0.947 and a kappa coefficient (KC) of 0.795. This model can be integrated with additional datasets to predict yield based on the identified crop type, thereby enhancing decision-making processes related to supply chain logistics and seasonal production forecasts. The early boll opening stage, occurring approximately a little more than a month before harvest, yielded the most precise identification of cotton fields.

Northern Slopes of Tianshan Mountain (NSTM) in Xinjiang hold significance as a principal agricultural hub within the region’s arid zone. Accurate crop mapping across vast agricultural expanses is fundamental for intelligent crop monitoring and devising sustainable agricultural strategies. Previous studies on multi-temporal crop classification have predominantly focused on single-point pixel temporal features, often neglecting spatial data. In large-scale crop classification tasks, by using spatial information around the pixel, the contextual relationships of the crop can be obtained to reduce possible noise interference. This research introduces a multi-scale, multi-temporal classification framework centered on ConvGRU (convolutional gated recurrent unit). By leveraging the attention mechanism of the Strip Pooling Module (SPM), a multi-scale spatial feature extraction module has been designed. This module accentuates vital spatial and spectral features, enhancing the clarity of crop edges and reducing misclassifications. The temporal information fusion module integration features from various periods to bolster classification precision. Using Sentinel-2 imagery spanning May to October 2022, datasets for cotton, corn, and winter wheat of the NSTM were generated for the framework’s training and validation. The results demonstrate an impressive 93.03% accuracy for 10 m resolution crop mapping using 15-day interval, 12-band Sentinel-2 data for the three crops. This method outperforms other mainstream methods like Random Forest (RF), Long Short-Term Memory (LSTM), Transformer, and Temporal Convolutional Neural Network (TempCNN), showcasing a kappa coefficient of 0.9062, 7.52% and 2.42% improvement in Overall Accuracy compared to RF and LSTM, respectively, which demonstrate the potential of our model for large-scale crop classification tasks to enable high-resolution crop mapping on the NSTM.

Affected by stagnant and transmission air pollution, Regional pollution day (RPD) frequently occurs in Twain-Hu Basin. In the winter halfaear of 2015–2021 (November to March of following year), the Regional pollution days reached 331 days. A variety of weather systems together lead to regional air pollution. Based on the observation data of PM2.5 concentration and multi-source meteorological elements during the study period, the weather types of Regional pollution days in Twain-Hu Basin can be divided into four types by using the T-model (T-PCA) classification method, namely, uniform pressure field (42 days), front of cold high-pressure bottom (180 days), back of high-pressure (81 days), and low pressure with an inverted trough field (28 days). Under the influence of uniform pressure field and front of cold high-pressure bottom, it shows the meteorological characteristics of static pollution and transmission pollution, respectively. A stable index (STI) is constructed to describe the PM2.5 pollution caused by the stagnant weather, and a transport index (TRI) is constructed to describe the PM2.5 pollution caused by the transport weather. They have a satisfactory evaluation effect on PM2.5 pollution under the influence of uniform pressure field and front of cold high-pressure bottom, respectively. Interestingly, the correlation coefficients between the pollution index (A stable index + transport index) and PM2.5 concentration could increase obviously to 0.48–0.71 when A stable index and transport index are considered comprehensively. Furthermore, the validation in the winter halfaear of 2021/2022 show that the correlation coefficient of most stations is above 0.70. Therefore, satisfactory performance in validation suggests that this index construction method can be well applied to regional pollution assessment in Twain-Hu Basin.

Abstract Persistent mechanoluminescence (ML) with long lifetime is highly required to break the limits of the transient emitting behavior under mechanics stimuli. However, the existing materials with persistent ML are completely trap‐controlled, and a pre‐irradiation is required, which severely hinders the practical applications. In this work, a novel type of ML, self‐charging persistent ML, is created by compositing the Sr3Al2O5Cl2:Dy3+ (SAOCD) powders into flexible polydimethylsiloxane (PDMS) matrix. With no need for any pre‐irradiation, the as‐fabricated SAOCD/PDMS elastomer could exhibit intense and persistent ML under mechanics stimuli directly, which greatly facilitates its applications in mechanics lighting, displaying, imaging, and visualization. By investigating the matrix effects as well as the thermoluminescence, cathodoluminescence, and triboelectricity properties, the interfacial triboelectrification‐induced electron bombardment processes are demonstrated to be responsible for the self‐charged energy in SAOCD under mechanics stimuli. Based on the unique self‐charging processes, the SAOCD/PDMS further exhibits mechanics storage and visualized reading activities, which brings novel ideas and approaches to deal with the mechanics‐related problems in the fields of mechanical engineering, bioengineering, and artificial intelligence.

Abstract The less improvement of ambient visibility suspects the government’s efforts on alleviating PM2.5 pollution. The COVID-19 lockdown reduced PM2.5 and increased visibility in Wuhan. Compared to pre-lockdown period, the PM2.5 concentration decreased by 39.0 μg m−3, dominated by NH4NO3 mass reduction (24.8 μg m−3) during lockdown period. The PM2.5 threshold corresponding to visibility of 10 km (PTV10) varied in 54–175 μg m−3 and an hourly PM2.5 of 54 μg m−3 was recommended to prevent haze occurrence. The lockdown measures elevated PTV10 by 9–58 μg m−3 as the decreases in PM2.5 mass scattering efficiency and optical hygroscopicity. The visibility increased by 107%, resulted from NH4NO3 extinction reduction. The NH4NO3 mass reduction weakened its mutual promotion with aerosol water and increased PM2.5 deliquescence humidity. Controlling TNO3 (HNO3 + NO3 −) was more effective to reduce PM2.5 and improve visibility than NH x (NH3 + NH4 +) unless the NH x reduction exceeded 11.7–17.5 μg m−3.

Crop-type mapping is the foundation of grain security and digital agricultural management. Accuracy, efficiency and large-scale scene consistency are required to perform crop classification from remote sensing images. Many current remote-sensing crop extraction methods based on deep learning cannot account for adaptation effects in large-scale, complex scenes. Therefore, this study proposes a novel adaptive feature-fusion network for crop classification using single-temporal Sentinel-2 images. The selective patch module implemented in the network can adaptively integrate the features of different patch sizes to assess complex scenes better. TabNet was used simultaneously to extract spectral information from the center pixels of the patches. Multitask learning was used to supervise the extraction process to improve the weight of the spectral characteristics while mitigating the negative impact of a small sample size. In the network, superpixel optimization was applied to post-process the classification results to improve the crop edges. By conducting the crop classification of peanut, rice, and corn based on Sentinel-2 images in 2022 in Henan Province, China, the novel method proposed in this paper was more accurate, indicated by an F1 score of 96.53%, than other mainstream methods. This indicates our model’s potential for application in crop classification in large scenes.

The effectiveness of landslide disaster prevention depends largely on the quality of early identification of potential hazards, and how to comprehensively, deeply, and accurately identify such hazards has become a major difficulty in landslide disaster management. Existing deep learning methods for potential landslide hazard identification often use fixed-size window modeling and ignore the different window sizes required by landslides of different scales. To address this problem, we propose an adaptive identification method for potential landslide hazards based on multisource data. Taking Yongping County, China, as the study area, we create a multisource factor dataset based on the landslide disaster background in terms of topography, geology, human activities, hydrology, and vegetation as the sample for the identification model after processing. Moreover, we combine differential interferometric synthetic aperture radar (D-InSAR) and multitemporal InSAR (MT-InSAR) to process the surface deformation of the study area, and we measure the deformation richness based on the average of the pixel deformation difference within the current window of a pixel point in the image. Therefore, convolutional neural networks (CNNs) with different window sizes are adaptively selected. The results show that the precision of adaptive identification of potential landslide hazards in the study area is 85.30%, the recall is 83.03%, and the F1 score is 84.15%. The recognition rate for potential hazards reaches 80%, which is better than the fixed-window modeling result and proves the effectiveness of the proposed method. This method can help to improve intelligent identification systems for potential landslide hazards, and also contribute to the identification of other potential geological hazards, such as mudslides and collapses.

The regional transport of air pollutants has been identified as a critical factor that affects air quality over downwind receptor areas; however, the regional transport patterns for air pollution episodes remain poorly understood. In this study, we identified two distinct patterns of regional transport: rapid transit transport (RTT) and stationary accumulation transport (SAT), both of which induced severe PM2.5 pollution episodes in central China during 2015–2020. The differences and similarities between the two regional transport patterns of air pollution were characterized by their influencing meteorological factors. Similar meteorological conditions trigger the RTT and SAT, with a high-PM2.5 concentration air mass from the upstream regions transported to the Twain-Hu Basin (THB) by the strong northerly winds with a southward advance of cold air mass. The average rate of increase in the PM2.5 concentration of the RTT (12.5 μg/m3/h) is more than that of the SAT (5.7 μg/m3/h). However, meteorological conditions evolved differently in RTT and SAT. For RTT with the fast passage of cold air, the THB is located behind the strong cold front during the later stage, and these events rapidly ended owing to the northerly winds and wet scavenging of precipitation. For SAT with the slow cold air accumulation, the THB is mainly occupied by a weak high-pressure system with low wind speeds, strong descending airflow, and a low atmospheric boundary layer height, trapping the air pollutants accumulated in the THB and resulting in persistent heavy pollution incidents. This study provides new insights into the meteorological mechanism underlying the formation of severe air pollution episodes over a receptor region owing to the regional transport patterns of air pollutants.

Abstract Atmospheric turbulence and the planetary boundary layer (PBL) can significantly modulate the fine particulate matter (PM2.5) concentration near the ground surface. However, the mechanism by which they affect the variation in PM2.5 is poorly understood. Using surface‐based meteorological and PM2.5 observations, turbulence measurements, and high‐resolution soundings, this study analyzed the properties of turbulence and PBL stratification during a severe regional pollution transport event in Xiangyang, a city in the middle reaches of the Yangtze River, from January 3 to January 9, 2019. During the growth stage, PM2.5 was observed to increase continuously under a northerly wind of 2–4 m s−1 that brought a polluted air mass to central China. During the entire pollution event, the turbulent kinetic energy and sensible heat flux tended to be lower at high PM2.5 concentrations. This correlation varied greatly depending on the stage of the event. A negative correlation existed in the initial and maintenance stages, whereas the relationship was positive during the growth stage, when regional transport dominated the variation in PM2.5. There was no obvious inversion layer below 500 m during the entire pollution episode, and a weak inversion layer appeared at higher altitudes in the initial and maintenance stages. Vertically homogeneous northerly winds favored the transboundary transport of pollutants, causing a significant increase in ground‐level PM2.5 concentration. By contrast, the presence of wind shear in the prevailing northerly winds suppressed pollutant transport to the ground, resulting in a low‐value gap in PM2.5 concentration between high altitudes and the ground surface.

The influence of the underlying surface on the atmospheric environment over rivers and lakes is not fully understood. To improve our understanding, this study targeted the Twain-Hu Basin (THB) in central China, with a unique underlying surface comprising a dense “water network” over rivers and lakes. In this study, the Weather Research and Forecasting Model with Chemistry (WRF-Chem) was used to simulate the impact of this dense “water network” on a wintertime heavy PM2.5 pollution event in the THB. On this basis, the regulating effects of density and area of the lake groups, with centralized big lakes (CBLs) and discrete small lakes (DSLs), on PM2.5 concentrations over the underlying surface of the dense “water network” in the THB were clarified, and the relative contributions of thermal factors and water vapor factors in the atmospheric boundary layer to the variation of PM2.5 concentrations were evaluated. The results show that the underlying surface of dense “water networks” in the THB generally decreases the PM2.5 concentrations, but the influences of different lake-group types are not uniform in spatial distribution. The CBLs can reduce the PM2.5 concentrations over the lake and its surroundings by 4.90–17.68% during the day and night. The ability of DSLs in reducing PM2.5 pollution is relatively weak, with the reversed contribution between −5.63% and 1.56%. Thermal factors and water vapor–related factors are the key meteorological drivers affecting the variation of PM2.5 concentrations over the underlying surface of dense “water networks”. The warming and humidification effects of such underlying surfaces contribute positively and negatively to the “purification” of air pollution, respectively. The relative contributions of thermal factors and water vapor–related factors are 52.48% and 43.91% for CBLs and 65.96% and 27.31% for DSLs, respectively. The “purification” effect of the underlying surface with a dense “water network” in the THB on regional air pollution highlights the importance of environmental protection of inland rivers and lakes in regional environmental governance. In further studies on the atmospheric environment, long-term studies are necessary, including fine measurements in terms of meteorology and the environment and more comprehensive simulations under different scenarios.

The lack of long-term observational data on secondary PM2.5 (SPM) has limited our comprehensive understanding of atmospheric environment change. This study develops an SPM estimation method, named Single-Tracer Approximate Envelope Algorithm (STAEA), to assess the long-term changes of SPM under different PM2.5 levels and in all seasons in Wuhan, Central China, over the period of anthropogenic pollutant mitigation in 2015–2020. The results show that: (1) the average proportions of SPM in ambient PM2.5 is 59.61% in a clean air environment, rising significantly to 71.60%, 73.73%, and 75.55%, respectively, in light, moderate, and heavy PM2.5 pollution, indicating the dominant role of SPM in air quality deterioration; (2) there are increasing trends of interannual changes of SPM at the light and moderate pollution levels of 1.95 and 3.11 μg·m−3·a−1 with extending SPM proportions in PM2.5 pollution, raising a challenge for further improvement in ambient air quality with mitigating light and moderate PM2.5 pollution; (3) the high SPM contributions ranging from 55.63% to 68.65% on a seasonal average and the large amplitude of seasonal SPM changes could dominate the seasonality of air quality; (4) the wintertime SPM contribution present a consistent increasing trend compared with the declining trends in spring, summer, and autumn, suggesting underlying mechanisms of SPM change for further deciphering the evolution of the atmospheric environment. Our results highlight the effects of air pollutant mitigation on long-term variations in SPM and its contributions with implications for atmospheric environment change.

The impact of structural variations in the atmospheric boundary layer (ABL) during the regional transport of air pollutants on its local pollution changes deserves attention. Based on multi-source ABL detection and numerical simulation of air pollutants over the Twain-Hu Basin (THB) during 4–6 January 2019, the mechanism of the rapid growth of atmospheric pollutant concentrations in Xianning by the synergistic effect of regional transport and ABL evolution is explored, and the main conclusions are obtained as follows. The vertically stratified atmosphere is noticeable at nighttime, and the heavy humidity of near-surface fog within the stable boundary layer (SBL) promoted the generation and cumulative growth of secondary PM2.5 components during the pollution formation stage. The horizontal transport characteristics of atmospheric pollutant concentration peak were observed in the residual layer (RL) of 500–600 m. At the pollution maintenance stage, the convective boundary layer (CBL) developed during the daytime, and northerly wind transported high-concentration pollutants from the north to the THB. Under the combined action of horizontal transport and turbulent mixing, the high-concentration atmospheric pollutants in the mixing layer (ML) from the ground to the 500 m height were mixed uniformly and maintained accumulation growth. The next day, the strong vertical turbulent mixing caused the downward transport of high-concentration pollutants in the RL during nighttime due to the development of the CBL again, resulting in a doubling of near-surface pollutant concentration in a short time. With the development of ABL turbulence, local pollution dissipated rapidly without the continuous input of pollutants from external regions. This study emphasizes the importance of multi-scale processes impact on pollution variation, that is, regional transport of atmospheric pollutants at the CBL development stage for the rapid growth of PM2.5 concentration in the ML.

With the development of deep learning, semantic segmentation technology has gradually become the mainstream technical method in large-scale multi-temporal landcover classification. Large-scale and multi-temporal are the two significant characteristics of Landsat imagery. However, the mainstream single-temporal semantic segmentation network lacks the constraints and assistance of pre-temporal information, resulting in unstable results, poor generalization ability, and inconsistency with the actual situation in the multi-temporal classification results. In this paper, we propose a multi-temporal network that introduces pre-temporal information as prior constrained auxiliary knowledge. We propose an element-wise weighting block module to improve the fine-grainedness of feature optimization. We propose a chained deduced classification strategy to improve multi-temporal classification’s stability and generalization ability. We label the large-scale multi-temporal Landsat landcover classification dataset with an overall classification accuracy of over 90%. Through extensive experiments, compared with the mainstream semantic segmentation methods, our proposed multi-temporal network achieves state-of-the-art performance with good robustness and generalization ability.

Vertical wind shear (VWS) significantly impacts the vertical mixing of air pollutants and leads to changes in near-surface air pollutants. We focused on Changsha (CS) and Jingmen (JM), the upstream and downstream urban sites of a receptor region in central China, to explore the impact of VWS on surface PM2.5 changes using 5-year wintertime observations and simulations from 2016–2020. The surface PM2.5 concentration was lower in CS with higher anthropogenic PM2.5 emissions than in JM, and the correlation between wind speed and PM2.5 was negative for clean conditions and positive for polluted conditions in both two sites. The difference in the correlation pattern of surface PM2.5 and VWS between CS and JM might be due to the different influences of regional PM2.5 transport and boundary layer dynamics. In downstream CS, the weak wind and VWS in the height of 1–2 km stabilized the ABL under polluted conditions, and strong northerly wind accompanied by enhanced VWS above 2 km favored the long-range transport of air pollutants. In upstream JM, local circulation and long-range PM2.5 transport co-determined the positive correlation between VWS and PM2.5 concentrations. Prevailed northerly wind disrupted the local circulation and enhanced the surface PM2.5 concentrations under polluted conditions, which tend to be an indicator of regional transport of air pollutants. The potential contribution source maps calculated from WRF-FLEXPART simulations also confirmed the more significant contribution of regional PM2.5 transport to the PM2.5 pollution in upstream region JM. By comparing the vertical profiles of meteorological parameters for typical transport- and local-type pollution days, the northerly wind prevailed throughout the ABL with stronger wind speed and VWS in transport-type pollution days, favoring the vertical mixing of transported air pollutants, in sharp contrast to the weak wind conditions in local-type pollution days. This study provided the evidence that PM2.5 pollution in the Twain-Hu Basin was affected by long-distance transport with different features at upstream and downstream sites, improving the understanding of the air pollutant source–receptor relationship in air quality changes with regional transport of air pollutants.

In recent years, with the development of deep learning in remotely sensed big data, semantic segmentation has been widely used in large-scale landcover classification. Landsat imagery has the advantages of wide coverage, easy acquisition, and good quality. However, there are two significant challenges for the semantic segmentation of mid-resolution remote sensing images: the insufficient feature extraction capability of deep convolutional neural network (DCNN); low edge contour accuracy. In this paper, we propose a block shuffle module to enhance the feature extraction capability of DCNN, a differentiable superpixel branch to optimize the feature of small objects and the accuracy of edge contours, and a self-boosting method to fuse semantic information and edge contour information to further optimize the fine-grained edge contour. We label three sets of Landsat landcover classification datasets, and achieved an overall accuracy of 86.3%, 83.2%, and 73.4% on the three datasets, respectively. Compared with other mainstream semantic segmentation networks, our proposed block shuffle network achieves state-of-the-art performance, and has good generalization ability.

Microphysics parameterization becomes increasingly important as the model grid spacing increases toward convection-resolving scales. Using observations from a field campaign for Mei-Yu rainfall in China, four bulk cloud microphysics schemes in the Weather Research and Forecasting (WRF) model were evaluated with respect to their ability to simulate precipitation, structure, and cloud microphysical properties over convective and stratiform regimes. These are the Thompson (THOM), Morrison graupel/hail (MOR_G/H), Stony Brook University (SBU_YLIN), and WRF double-moment six-class microphysics graupel/hail (WDM6_G/H). All schemes were able to predict the rain band but underestimated the total precipitation by 23%–35%. This is mainly attributed to the underestimation of stratiform precipitation and overestimation of convective rain. For the vertical distribution of radar reflectivity, many problems remain, such as lower reflectivity values aloft in both convective and stratiform regions and higher reflectivity values at middle level. Each bulk scheme has its advantages and shortcomings for different cloud regimes. Overall, the discrepancies between model output and observations mostly exist in the midlevel to upper level, which results from the inability of the model to accurately represent the particle size distribution, ice processes, and storm dynamics. Further observations from major field campaigns and more detailed evaluation are still necessary.

Based on the PM2.5 and PM10 mass concentration data obtained from 51 national air quality monitoring stations and the corresponding rainfall intensity data in automatic meteorological stations in Hubei Province from 1 January 2015 to 31 December 2017, the impact of rainfall intensity on PM mass concentrations under relatively different humidity conditions was analyzed. The results showed that light rain occurred most frequently in the pollution process, with Xiangyang being affected for up to 587 h. PM concentration would not change drastically under the effect of precipitation. Mean rainfall intensity responsible for wet growth of PM10 and PM2.5 was mainly 1.4 mm/h) than that of PM10 (>1.0 mm/h). Precipitation was more likely to produce a wet removal effect for a greater initial value of PM mass concentration, and on the contrary, a wet growth effect was more likely, with the threshold of PM10 mass concentration being 150 μg/m3 and that of PM2.5 mass concentration being 95 μg/m3. Wet removal played a leading role in lower humidity (∼60%) and greater rainfall intensity, but wet growth played a leading role in higher humidity (∼90%) and lower rainfall intensity. As the precipitation level increased (rainfall ≥1.5 mm·h−1), the wet removal to PM10 mass concentration was enhanced more obviously. The variations of PM2.5 had similar distributions to those of PM10 under the effect of precipitation, but the wet removal effect of precipitation was weakened and the wet growth effect was enhanced.

The extent of PM2.5 pollution has reduced in traditional polluted regions such as the North China Plain (NCP), Yangtze River Delta (YRD), Sichuan Basin (SB), and Pearl River Delta (PRD) over China in recent years. Despite this, the Twain-Hu Basin (THB), which covers the lower flatlands in Hubei and Hunan provinces in central China, was found to be a high PM2.5 pollution region, with annual mean PM2.5 concentrations of 41–63 μg·m−3, which is larger than the values in YRD, SB, and PRD during 2014–2019, and high aerosol optical depth values (>0.8) averaged over 2000–2019 from the MODIS products. Heavy pollution events (HPEs) are frequently observed in the THB, with HPE-averaged concentrations of PM2.5 reaching up to 183–191 μg·m−3, which exceeds their counterparts in YRD, SB, and PRD for 2014–2019, highlighting the THB as a center of heavy PM2.5 pollution in central China. During 2014–2019, approximately 65.2% of the total regional HPEs over the THB were triggered by the regional transport of PM2.5 over Central and Eastern China (CEC). This occurred in view of the co-existing HPEs in the NCP and the THB, with a lag of almost two days in the THB-PM2.5 peak, which is governed by the strong northerlies of the East Asian monsoon (EAM) over CEC. Such PM2.5 transport from upstream source regions in CEC contributes 60.3% of the surface PM2.5 pollution over the THB receptor region. Hence, a key PM2.5 receptor of the THB in regional pollutant transport alters the distribution patterns of PM2.5 pollution over China, which is attributable to the climate change of EAMs. This study indicates a complex relationship between sources and receptors of atmospheric aerosols for air quality applications.

In 2020, when the novel coronavirus disease 2019 (COVID-19) broke out as a global pandemic, cities in Hubei Province first went into lockdown on 23 January and resumed work and production on 20 March. From February to March 2020, human activities in Hubei decreased significantly, with the average particulate matter smaller than 2.5 μm (PM2.5) concentration standing at 40 μg/m3, which is 21% lower than the expected based on a linear fitting trend in thePM2.5 concentration in Hubei. By using the empirical orthogonal function (EOF) method, this paper comparatively analyzes the spatial-temporal variations of Hubei’s PM2.5 concentration anomaly in February and March 2020 and the same periods of 2016–2019. The results show that the daytime peak of the PM2.5 daily variation in Hubei in a low-emission scenario during COVID-19 declined significantly, to which human activities contributed the most. However, during nighttime, the PM2.5 peak became more prominent, and the meteorological conditions had a more noticeable effect on the PM2.5 concentration. In addition, during COVID-19, there was a great drop in PM2.5 pollution accumulated from local sources within the urban circle of Wuhan City, while an increase was seen in central-western Hubei due to the inter-regional pollutant transport. Thus, the high PM2.5 concentration center in the urban circle of Wuhan disappeared, but the pollution transport channel cities in central-western Hubei remained as high-PM2.5-concentration centers.

The Twain-Hu basin (THB), covering the lower plain of Hubei and Hunan provinces in Central China, has experienced severe air pollution in recent years. However, the terrain effects of such sub-basin on air quality over the THB have been incomprehensibly understood. A heavy PM2.5 pollution event occurred over the THB during 4–10 January 2019. By using the observations and WRF-Chem simulations, we investigated the underlying mechanisms of sub-basin effects on the air pollution with several sensitivity experiments. Observationally, air pollution in the western THB urban area with an average PM2.5 concentration of 189.8 μg m−3, which was more serious than the eastern urban area with the average PM2.5 concentration of 106.3 μg m−3, reflecting a different influence of topography on air pollution over the THB. Simulation results revealed that the terrain effect can contribute 12.0% to increasing the PM2.5 concentrations in the western THB, but slightly mitigate the pollution extent in the eastern THB with the contribution of −4.6% to PM2.5 during the heavy pollution episode. In particular, the sub-basin terrain was conducive to the accumulation of PM2.5 by regional transport with the contribution of 39.1 %, and contrarily lowered its local pollution by −57.0% via the enhanced atmospheric boundary layer height and ventilation coefficients. Given a heavy air pollution episode occurring over the THB, such inverse contribution of terrain effects reflected a unique importance of sub-basin topography in regional transport of air pollutants for air pollution in central China.

Recently, surface diffuse solar radiation (Rdif) has been attracting a growing interest in view of its function in improving plant productivity, thus promoting global carbon uptake, and its impacts on solar energy utilization. To date, very few radiation products provide estimates of Rdif, and systematic validation and evaluation are even more scare. In this study, Rdif estimates from Reanalysis Fifth Generation (ERA5) of European Center for Medium-Range Weather Forecasts and satellite-based retrieval (called JiEA) are evaluated over East Asia using ground measurements at 39 stations from World Radiation Data Center (WRDC) and China Meteorological Administration (CMA). The results show that JiEA agrees well with measurements, while ERA5 underestimates Rdif significantly. Both datasets perform better at monthly mean scale than at daily mean and hourly scale. The mean bias error and root-mean-square error of daily mean estimates are −1.21 W/m2 and 20.06 W/m2 for JiEA and −17.18 W/m2 and 32.42 W/m2 for ERA5, respectively. Regardless of over- or underestimation, correlations of estimated time series of ERA5 and JiEA show high similarity. JiEA reveals a slight decreasing trend at regional scale, but ERA5 shows no significant trend, and neither of them reproduces temporal variability of ground measurements. Data accuracy of ERA5 is more robust than JiEA in time but less in space. Latitudinal dependency is noted for ERA5 while not for JiEA. In addition, spatial distributions of Rdif from ERA5 and JiEA show pronounced discrepancy. Neglect of adjacency effects caused by horizontal photon transport is the main cause for Rdif underestimation of ERA5. Spatial analysis calls for improvements to the representation of clouds, aerosols and water vapor for reproducing fine spatial distribution and seasonal variations of Rdif.

Based on the ensemble precipitation forecast data in the summers of 2014−2018 from the Observing System Research and Predictability Experiment (THORPEX) Interactive Grand Global Ensemble (TIGGE), a comparative study of two multi-model ensemble methods, the Bayesian model average (BMA) and the logistic regression (LR), was conducted. Meanwhile, forecasts of heavy precipitation from the two models over the Wujiang River Basin in China for the summer of 2018 were compared to verify their performances. The training period sensitivity test results show that a training period of 2 years was the best for BMA probability forecast model. Compared with the BMA method, the LR model required more statistical samples and its optimal length of the training period was 5 years. According to the Brier score (BS), for precipitation events exceeding 10 mm with lead times of 1−7 days, the BMA outperformed the LR and the raw ensemble prediction system forecasts (RAW) except for forecasts with a lead time of 1 day. Furthermore, for heavy rainfall events exceeding 25 and 50 mm, the RAW and the BMA performed much the same in terms of prediction. The reliability diagram of the two multi-model ensembles (i.e., BMA and LR) was more reliable than the RAW for heavy and moderate rainfall forecasts, and the BMA model had the best performance. The BMA probabilistic forecast can produce a highly concentrated probability density function (PDF) curve and can also provide deterministic forecasts through analyzing percentile forecast results. With regard to the heavy rainfall forecast in mountainous areas, it is recommended to refer to the forecast with a larger percentile between the 75th and 90th percentiles. Nevertheless, extreme events with low probability forecasts may occur and cannot be ignored.

The rapid development of remote sensing technology has promoted the generation of different vegetation index products, resulting in substantive accomplishment in comprehensive economic development and monitoring of natural environmental changes. The results of scientific experiments based on various vegetation index products are also different with the variation of time and space. In this work, the consistency characteristics among three global normalized difference vegetation index (NDVI) products, namely, GIMMS3g NDVI, MOD13A3 NDVI, and SPOT-VGT NDVI, are intercompared and validated based on Landsat 8 NDVI at biome and regional scale over the Mongolian Plateau (MP) from 2000 to 2014 by decomposing time series datasets. The agreement coefficient (AC) and statistical scores such as Pearson correlation coefficient, root mean square error (RMSE), mean bias error (MBE), and standard deviation (STD) are used to evaluate the consistency between three NDVI datasets. Intercomparison results reveal that GIMMS3g NDVI has the highest values basically over the MP, while SPOT-VGT NDVI has the lowest values. The spatial distribution of AC values between various NDVI products indicates that the three NDVI datasets are highly consistent with each other in the northern regions of the MP, and MOD13A3 NDVI and SPOT-VGT NDVI have better consistency in expressing vegetation cover and change trends due to the highest proportions of pixels with AC values greater than 0.6. However, the trend components of decomposed NDVI sequences show that SPOT-VGT NDVI values are about 0.02 lower than the other two datasets in the whole variation periods. The zonal characteristics show that GIMMS3g NDVI in January 2013 is significantly higher than those of the other two datasets. However, in July 2013, the three datasets are remarkably consistent because of the greater vegetation coverage. Consistency validation results show that values of SPOT-VGT NDVI agree more with Landsat 8 NDVI than GIMMS3g NDVI and MOD13A3 NDVI, and the consistencies in the northeast of the MP are higher than northwest regions.