Dacheng Wang's research while affiliated with Chinese Academy of Sciences and other places

The spatiotemporal distribution of industrial heat sources (IHS) is an important indicator for assessing levels of energy consumption and air pollution. Continuous, comprehensive, dynamic monitoring and publicly available datasets of global IHS (GIHS) are lacking and urgently needed. In this study, we built the first long-term (2012–2021) GIHS dataset based on the density-based spatiotemporal clustering method using multi-sources remote sensing data. A total of 25,544 IHS objects with 19 characteristics are identified and validated individually using high-resolution remote sensing images and point of interest (POI) data. The results show that the user’s accuracy of the GIHS dataset ranges from 90.95% to 93.46%, surpassing other global IHS products in terms of accuracy, omission rates, and granularity. This long-term GIHS dataset serves as a valuable resource for understanding global environmental changes and making informed policy decisions. Its availability contributes to filling the gap in GIHS data and enhances our knowledge of global-scale industrial heat sources.

Industrial heat sources (IHSs) are key contributors to anthropogenic heat, air pollution, and carbon emissions. Accurately and automatically detecting their production areas (IHSPAs) on a large scale is vital for environmental monitoring and decision making, yet this is challenged by the lack of high-resolution thermal data. Sustainable Development Science Satellite 1 (SDGSAT-1) thermal infrared spectrometer (TIS) data with the highest resolution (30 m) in the civilian field and a three-band advantage were first introduced to detect IHSPAs. In this study, an IHSPA identification model using multi-features extracted from SDGSAT-1 TIS and Landsat OLI data and support vector machine (SVM) was proposed. First, three brightness temperatures and four thermal radiation indices using SDGSAT-1 TIS and Landsat OLI data were designed to enlarge the temperature difference between IHSPAs and the background. Then, 10 features combined with three indices from Landsat OLI images with the same spatial resolution (30 m) and stable data were extracted. Second, an IHSPA identification model based on SVM and multi-feature extraction was constructed to identify IHSPAs. Finally, the IHS objects were manually delineated and verified using the identified IHSPAs and Google Earth images. Some conclusions were obtained from different comparisons in Wuhai, China: (1) IHSPA identification based on SVM using thermal and optical features can detect IHSPAs and obtain the best results compared with different features and identification models. (2) The importance of using thermal features from the SDGSAT-1 TIS to detect IHSPAs was demonstrated by different importance analysis methods. (3) Our proposed method can detect more IHSs, with greater spatial coverage and smaller areas, compared with the methods of Ma and Liu. This new way to detect IHSPAs can obtain higher-spatial-resolution emissions of IHSs on a large scale and help decision makers target environmental monitoring, management, and decision making in industrial plant processing.

The prevalent high-energy, high-pollution and high-emission economic model has led to significant air pollution challenges in recent years. The industrial sector in the Beijing–Tianjin–Hebei (BTH) region is a notable source of atmospheric pollutants, with industrial heat sources (IHSs) being primary contributors to this pollution. Effectively managing emissions from these sources is pivotal for achieving air pollution control goals in the region. A new three-stage model using multi-source long-term data was proposed to estimate atmospheric, delicate particulate matter (PM2.5) concentrations caused by IHS. In the first stage, a region-growing algorithm was used to identify the IHS radiation areas. In the second and third stages, based on a seasonal trend decomposition procedure based on Loess (STL), multiple linear regression, and U-convLSTM models, IHS-related PM2.5 concentrations caused by meteorological and anthropogenic conditions were removed using long-term data from 2012 to 2021. Finally, this study analyzed the spatial and temporal variations in IHS-related PM2.5 concentrations in the BTH region. The findings reveal that PM2.5 concentrations in IHS radiation areas were higher than in background areas, with approximately 33.16% attributable to IHS activities. A decreasing trend in IHS-related PM2.5 concentrations was observed. Seasonal and spatial analyses indicated higher concentrations in the industrially dense southern region, particularly during autumn and winter. Moreover, a case study in Handan’s She County demonstrated dynamic fluctuations in IHS-related PM2.5 concentrations, with notable reductions during periods of industrial inactivity. Our results aligned closely with previous studies and actual IHS operations, showing strong positive correlations with related industrial indices. This study’s outcomes are theoretically and practically significant for understanding and addressing the regional air quality caused by IHSs, contributing positively to regional environmental quality improvement and sustainable industrial development.

Global warming is currently an area of concern. Human activities are the leading cause of urban greenhouse gas intensification. Inversing the spatial distribution of carbon emissions at microscopic scales such as communities or controlling detailed planning plots can capture the critical emission areas of carbon emissions, thus providing scientific guidance for intracity low-carbon development planning. Using the Sino—Singapore Tianjin Eco-city as an example, this paper uses night-light images and statistical yearbooks to perform linear fitting within the Beijing—Tianjin—Hebei city-county region and then uses fine-scale data such as points of interest, road networks, and mobile signaling data to construct spatial characteristic indicators of carbon emissions distribution and assign weights to each indicator through the analytic hierarchy process. As a result, the spatial distribution of carbon emissions based on detailed control planning plots is calculated. The results show that among the selected indicators, the population distribution significantly influences carbon emissions, with a weight of 0.384. The spatial distribution of carbon emissions is relatively distinctive. The primary carbon emissions are from the Sino—Singapore Cooperation Zone due to its rapid urban construction and development. In contrast, carbon emissions from other areas are sparse, as there is mostly unused land under construction.

Accurate monitoring of bare soil land (BSL) is an urgent need for environmental governance and optimal utilization of land resources. High-resolution imagery contains rich semantic information, which is beneficial for the recognition of objects on the ground. Simultaneously, it is susceptible to the impact of its background. We propose a semantic segmentation model, Deeplabv3+-M-CBAM, for extracting BSL. First, we replaced the Xception of Deeplabv3+ with MobileNetV2 as the backbone network to reduce the number of parameters. Second, to distinguish BSL from the background, we employed the convolutional block attention module (CBAM) via a combination of channel attention and spatial attention. For model training, we built a BSL dataset based on BJ-2 satellite images. The test result for the F1 of the model was 88.42%. Compared with Deeplabv3+, the classification accuracy improved by 8.52%, and the segmentation speed was 2.34 times faster. In addition, compared with the visual interpretation, the extraction speed improved by 11.5 times. In order to verify the transferable performance of the model, Jilin-1GXA images were used for the transfer test, and the extraction accuracies for F1, IoU, recall and precision were 86.07%, 87.88%, 87.00% and 95.80%, respectively. All of these experiments show that Deeplabv3+-M-CBAM achieved efficient and accurate extraction results and a well transferable performance for BSL. The methodology proposed in this study exhibits its application value for the refinement of environmental governance and the surveillance of land use.

As dust source bare soil land (BSL) contributes to air pollution and affects the photosynthesis of green plants and carbon absorption, it is the objective of this study to develop an approach for monitoring the changes of BSL using remote sensing technology. Unlike other land use/cover types, the classification of BSL as well as its change detection is often ignored. For traditional convolutional neural networks, deep layers cause a long range between input and output, inevitably leading to the loss of information and computational costs. To alleviate this problem, transformer is available to model the global dependencies. Bi-temporal association, HHaassociationa which is described as subtraction or attention mechanism, is not fully considered by current methods. Therefore, we proposed a spatial-temporal association enhanced mobile-friendly vision transformer (STAE-MobileVIT) for change detection of high-resolution images with light weight and high efficiency. On the one hand, a temporal association enhanced MobileVIT block is employed to strengthen the association of bi-temporal images during feature extraction. On the other hand, a multi-scale feature difference aggregator enhanced by spatial association is designed to fuse semantic and detailed information. Since the lack of binary change detection dataset for BSL, we established a small dataset named BSL-CD, consisting of 1083 pairs of 0.8m bi-temporal images with the size of 256×256 pixels, along with the corresponding labels. The experiments on BSL-CD show that our light-weight model surpass seven common methods by 3.48, 5.05 and 1.44 percent on F1, IoU and OA, which proves the efficiency and accuracy of STAE-MobileVIT.

The water use efficiency (WUE) is an essential indicator of carbon–water coupling between terrestrial ecosystems and the atmosphere, and it is an important parameter for studying ecosystem responses to global climate change. A comprehensive understanding of the water–carbon coupling process in the Loess Plateau can reflect the balance between the “carbon absorption” and “water consumption” in vegetation, which drives the ecosystem succession process. In recent years, scholars have gained a more comprehensive understanding of the WUE and the driving factors of the Loess Plateau. However, there is still a need to study the carbon and water coupling mechanisms of different land use types in the Loess Plateau region. In this article, based on the gross primary productivity (GPP), evapotranspiration (ET), surface cover remote sensing products, and meteorological observation data, the trend of WUE changes for different vegetation types in the Loess Plateau from 2001 to 2020 and the correlations with the Normalized Difference Vegetation Index (NDVI), precipitation, and temperature values were analyzed using the Theil–Sen median (SEN) trend analysis method and correlation coefficient analysis method. The spatial distribution patterns of the changes with the drought index showed that the multi-year average WUE value of the Loess Plateau was 1.24 g C mm−1 H2O, and the mean WUE values in different seasons were ranked as follows: summer > autumn > spring. The WUE growth rates of all vegetation types showed a decreasing trend with the increase in drought index, and the size of the WUE response rate for each vegetation type to drought was ranked as follows: grassland > forest > shrub > crop. The annual average WUE increase rate of the Loess Plateau was 0.02 g C mm−1 H2O yr−1, of which 93.36% of the area showed an increasing trend. The NDVI was the dominant factor affecting the spatial and temporal variations in WUE rates in the Loess Plateau, and the correlation between the NDVI and WUE was strongest in summer. In the more arid regional ecosystems, the WUE was negatively correlated with the precipitation and temperature, but in summer the precipitation had a positive effect on the WUE. The correlation of grassland and shrub WUE rates with temperature was more sensitive to the drought index than that of the forest and crop areas, but there was also a threshold effect. Therefore, when vegetation restoration is carried out in arid and semi-arid regions, the carbon and water coupling mechanisms of different vegetation types and the reasonable allocation of regional water resources should be fully considered.

The Grain for Green Project (GGP) has considerably improved the vegetation cover of the Loess Plateau, as well as changed the carbon and water coupling process of local vegetation to a certain extent. Water use efficiency (WUE) is a crucial measure for evaluating ecosystem responses to global climate change and is a key indicator of the carbon–water coupling between terrestrial ecosystems and the environment. A comprehensive understanding of the impact of vegetation reconstruction on WUE on the Loess Plateau is of great significance to the vegetation growth and contribution to sustainable of the Loess Plateau. In recent years, scholars have gained a more comprehensive understanding of the distribution and drivers of WUE on the Loess Plateau. However, through the study of carbon and water coupling in the Loess Plateau, it is found that the effects of different vegetation restoration levels on WUE are still to be studied in depth in terms of spatial and temporal heterogeneity and long timeseries. In this paper, we analyzed the trends of Normalized Difference vegetation cover (NDVI) and WUE from 2001 to 2010 and 2011 to 2020, respectively, to research at the WUE of the vegetation in this area in relation to vegetation restoration. It was found that the Loess Plateau’s vegetation WUE rose from 2001 to 2020 at a rate of 0.023 g C kg−1 H2O per year, and that the increase from 2011 to 2020 was more significant than the growth from 2000 to 2010. The Loess Plateau’s area with a growing trend in vegetation water use rate increased from 77.12% in 2001–2010 to 88.63% in 2011–2020, with the majority of the increased area occurring in the northeastern Inner Mongolia region. After 20 years of the reforestation project, the area where NDVI and WUE increased simultaneously accounted for 71.54% of the Loess Plateau, the area where NDVI increased but WUE decreased accounted for 10.95% of the Loess Plateau, and the area where NDVI increased but WUE decreased accounted for 7.15% of the Loess Plateau. The correlation between temperature precipitation and WUE was not significant for the whole Loess Plateau, further indicating that the increase in vegetation cover was the main reason for the increase in vegetation water efficiency. Therefore, the effect of vegetation cover on WUE should be fully considered when vegetation restoration is carried out on the Loess Plateau.

Accurate mapping of dams can provide useful information about geographical locations and boundaries and can help improve public dam datasets. However, when applied to disaster emergency management, it is often difficult to completely determine the distribution of dams due to the incompleteness of the available data. Thus, we propose an automatic and intelligent extraction method that combines location with post-segmentation for dam detection. First, we constructed a dataset named RSDams and proposed an object detection model, YOLOv5s-ViT-BiFPN (You Only Look Once version 5s-Vision Transformer-Bi-Directional Feature Pyramid Network), with a training method using deep transfer learning to generate graphical locations for dams. After retraining the model on the RSDams dataset, its precision for dam detection reached 88.2% and showed a 3.4% improvement over learning from scratch. Second, based on the graphical locations, we utilized an improved Morphological Building Index (MBI) algorithm for dam segmentation to derive dam masks. The average overall accuracy and Kappa coefficient of the model applied to 100 images reached 97.4% and 0.7, respectively. Finally, we applied the dam extraction method to two study areas, namely, Yangbi County of Yunnan Province and Changping District of Beijing in China, and the recall rates reached 69.2% and 81.5%, respectively. The results show that our method has high accuracy and good potential to serve as an automatic and intelligent method for the establishment of a public dam dataset on a regional or national scale.