Mangrove forests are recognized as one of the most biodiverse and productive marine ecosystems globally. This study investigated Beibu Gulf, Guangxi Province. Using Landsat, Sentinel, and PALSAR SAR images from 1985 to 2019 as data sources, as well as the Google Earth Engine (GEE) cloud platform, this study established a multisource dataset by integrating spectral bands, spectral indices, texture features, digital elevation models (DEMs), and backscatter coefficients. Furthermore, 14 classification schemes were developed, and a mangrove remote sensing recognition model was built using an object-based random forest (RF) algorithm. Accordingly, the long-time-series spatiotemporal dynamics of mangrove forests in Beibu Gulf were monitored. The monitoring results show that the object-based RF algorithm demonstrates a high ability to identify mangrove forests. Specifically, Scheme 3 combined with data from 2019 yielded the highest overall accuracy (96.3%) and a kappa coefficient of 0.956, which are 16.3% and 0.195 higher than those of Scheme 1 combined data from 1995, respectively. The classification schemes differed in the producer’s and user’s accuracy of different surface features in the Beibu Gulf. Specifically, these schemes yielded the highest user’s and producer’s accuracy of mangrove forests exceeding 94.6% and 92.0%, respectively. From 1985 to 2019, the area of mangrove forests in Beibu Gulf showed an increasing trend, with an annual changing rate of 6.63%, and the area expanded from inland to coastal areas. The results of this study provide a reference for the protection and sustainable management of mangrove forests while also verifying the feasibility of monitoring long-term spatiotemporal dynamics of mangrove forests based on the GEE platform.

To address the accuracy reduction in the semantic segmentation of remote sensing images due to insufficient extraction of contextual dependencies and loss of spatial details, this study proposed a semantic segmentation method based on context- and class-aware feature fusion. With ResNet-50 as the backbone network for feature extraction, the proposed method incorporates the attention module during downsampling to enhance feature representation and contextual dependency extraction. It constructs a large receptive field block on skip connections to extract rich multiscale contextual information, thereby mitigating the impacts of scale variations between targets. Furthermore, it connects a scene feature association and fusion module in parallel behind the block to guide local feature fusion based on global features. Finally, it constructs a class prediction module and a class-aware feature fusion module in the decoder part to accurately fuse the low-level advanced semantic information with high-level detailed information. The proposed method was validated on the Potsdam and Vaihingen datasets and compared with six commonly used methods, including DeepLabv3+ and BuildFormer, to verify its effectiveness. Experimental results demonstrate that the proposed method outperformed other methods in terms of recall, F1-score, and accuracy. Particularly, it yielded intersection over union (IoU) values of 90.44% and 86.74% for building segmentation, achieving improvements of 1.55% and 2.41%, respectively, compared to suboptimal networks DeepLabv3+ and A2FPN.

To address the issues of missing small-scale surface features and incomplete continuous features in segmentation results, this study proposed a densely connected multiscale semantic segmentation network (DMS-Net) model for land cover segmentation. The model integrates a multiscale densely connected atrous spatial convolution pyramid pooling module and strip pooling to extract multiscale and spatially continuous features. A position paralleling Channel attention module (PPCA) is employed to assess feature weights for high-efficiency expression. A cascade low-level feature fusion (CLFF) module is applied to capture neglected low-level features, further complementing details. Experimental results demonstrate that the DMS-Net model achieved an overall accuracy (OA) of 89.97 % and a mean intersection over union (mIoU) of 75.59 % on an iteratively extended dataset, outperforming traditional machine learning methods and deep learning models like U-Net, PSPNet, and Deeplabv3+. The segmentation results of the DMS-Net model reveal structurally complete surface features with clear boundaries, underscoring its practical value in multiscale extraction and analysis of remote sensing information for land cover.

Landslide identification has always been a research topic in the study of geological disasters, playing a significant role in emergency rescue and command. To address the limitations in landslide identification, such as missed/false detection, and low identification accuracy, this study proposed an improved YOLOv7 model that enables simultaneous object detection and image segmentation for landslides. The improved model optimized its core network by integrating data, adding the convolutional block attention module (CBAM), and changing the intersection over union (IoU) loss function. Its effectiveness was verified using the landslide dataset of Bijie City, Guizhou Province, and the 0.2 m high-resolution digital orthophoto map (DOM) of historical landslides in Sichuan Province. The results indicate that the improved model performed well in landslide detection and segmentation, achieving more efficient and accurate landslide identification compared to the conventional YOLOv7 model, and other prevailing models like Fast RCNN and Mask RCNN. Taking the Baige area in Sichuan Province as an example, this model can effectively enhance the automation level of landslide disaster information acquisition while improving accuracy.

To address the issues of low detection accuracy caused by significant scale variations, complex scenes, and limited feature information of small targets in remote sensing images, as well as low detection efficiency resulting from the large parameter size and high complexity of current object detection models, this study proposes a lightweight YOLOv7-tiny model for remote sensing image detection. First, the network neck was improved by incorporating group shuffle convolution (GSConv) and VoV-GSCSP modules. This allows for sufficient detection accuracy while reducing computational costs and network complexity. Second, a dynamic head (DyHead) combined with an attention mechanism was adopted during prediction. The performance of the detection head was enhanced using multi-head self-attention across scale-aware feature layers, spatially-aware positions, and task-aware output channels. Finally, the loss function of the original model was optimized by integrating the normalized Wasserstein distance (NWD) metric for small-target assessment and a bounding box regression loss function based on the minimum point distance IoU (MPDIoU). This assists in enhancing robustness for small target detection. The experimental results demonstrate that the proposed algorithm achieved mAP@50 scores of 87.7% and 94.7% on the DIOR and RSOD datasets, respectively, indicating increases of 2.7 and 5.1 percentage points compared to the original YOLOv7-tiny model. Furthermore, the frames per second (FPS) increased by 12.2% and 11.9%, respectively. Therefore, the proposed algorithm can effectively enhance both the accuracy and real-time performance of small target detection from remote sensing images.

Hazes reduce the quality of remote sensing images while limiting the performance of back-end visual applications. Hence, this study proposed a multi-scale residual dehazing network based on dual attention. First, an atmospheric scattering model was constructed to combine the atmospheric light value and transmissivity to derive the atmospheric power of light. Second, an end-to-end deep learning model was used to clarify remote sensing images with hazes. The dehazing network consists of a shallow feature extraction module, a deep data extraction module, a dual mapping network, and a parallel convolution reconstruction module. Finally, the proposed dehazing network was compared with CARL-net, DFAD-net, SRBFP-net, and AMGP-net through subjective and objective comparison experiments. The results indicate that the proposed dehazing network obtained a visual state close to the original haze-free scene, exhibiting high contrast, bright chroma, corresponding saturation, and clear transmission map details. Moreover, it effectively removed image noise while maintaining the edge of the foreground part. Compared to the above four networks, the proposed dehazing network achieved superior peak signal-to-noise ratio (PSNR) and structure similarity index measure (SSIM), higher algorithm processing efficiency, and stable algorithm processing time with the increase of image resolution.

Interferometric Synthetic Aperture Radar (InSAR) faces the challenges of the insufficient number of monitoring points and low monitoring accuracy when applied to complex environments with dense vegetation and large-gradient surface deformation in a mining area. To address these challenges, this study proposed an improved distributed scatterer InSAR (DS-InSAR) method assisted by stacking technology. This method identified statistically homogenous pixels using a confidence interval hypothesis test and achieved phase optimization utilizing a phase triangulation algorithm. Subsequently, the residual phases were derived by removing the linear deformation phases determined via stacking-based simulation. This step contributed to sparse deformation phase fringes, thereby enhancing the accuracy of spatiotemporal filtering and three-dimensional phase unwrapping within the subsequent DS-InSAR processing framework. Finally, the simulated phases were compensated, and thus complete deformation fields were determined. By processing Sentinel-1A SAR images covering the Xinjulong Coal Mine from October 2015 to March 2016, this study interpreted the time-series surface deformation characteristics in the mining area during this period. The findings revealed three significant deformation sites in the mining area, with a maximum cumulative radar line-of-sight (LOS) deformation of up to -313 mm. Compared to conventional small Baseline Subset (SBAS) InSAR, the proposed method yielded more uniformly distributed monitoring points via inversion, with a density approximately 12.9 times higher. The root mean squared error (RMSE) of the inversion was approximately 6.82 mm relative to leveling data, representing an accuracy improvement of about 3.0 mm compared to the SBAS results.

Satellite altimetry enables non-contact，large-scale Earth observation，providing technical support for monitoring changes in water levels of lakes where there is a lack of ground-based hydrological stations. The ICESat-2 laser altimeter features small footprints and high measurement accuracy，enjoying advantages in monitoring small-to medium-sized lakes. Therefore，this study extracted water level data from October 2018 to August 2022 for 1248 lakes across China based on ICESat-2 ATL08 data. The extracted data were validated using measured water level data from 18 lakes and Hydroweb data from 36 ones. Subsequently，based on the division of China's five major lake regions，this study analyzed variations in water levels of 957 lakes that were observed for over two years in at least four campaigns. The results show that the root mean square errors （RMSEs） between ICESat-2-derived and measured lake levels showed a minimum of 0.097 m. The cross-validation with Hydroweb data yielded a correlation coefficient of 0.95 and a minimum RMSE of 0.085 m. These results demonstrate the high precision and accuracy of the water level retrieval based on the ICESat-2 data. The lake levels on the Tibetan Plateau exhibited a slow rising trend，while those in northwestern China showed a declining trend. In eastern China，the water levels of large lakes displayed no significant variation trend，whereas those of small lakes showed pronounced fluctuations. Overall，the lake levels across China exhibited a gently rising trend. This study achieved high-precision measurement and monitoring of variations in lake levels across China，providing a scientific basis for water resource protection，ecological management，and the exploration of the responses of lake levels to human activities and climate change.

Compared to natural surfaces, urban surfaces have more complex geometric structures, leading to significant impacts of the multiple scattering effect within pixels and the neighborhood effect on the inversion results of urban land surface temperature (LST). This study proposed a novel urban LST inversion algorithm that integrates machine learning and an enhanced temperature and emissivity separation (TES) method. Finally, the proposed algorithm was applied to China’s SDGSAT-1 thermal infrared data. The algorithm comprises three key steps: First, the inversion of urban canopy brightness temperature from SDGSAT-1 data was conducted using the eXtreme Gradient Boosting (XGBoost) algorithm. Second, an enhanced TES algorithm based on the sky view factor (SVF) was developed to account for urban geometry, enabling high-precision urban LST inversion. Third, the accuracy of the inversion algorithm was assessed and applied to the urban area of Beijing. The results demonstrate that inversion using an XGBoost algorithm and a split-window algorithm yielded root mean squared errors (RMSEs) of approximately 0.2 K and 1.2 K, respectively. The LST RMSEs with and without available water vapor data were determined at 0.36 K and 0.73 K, respectively; and the LSE RMSEs under three bands were 0.020/0.026, 0.018/0.023, and 0.020/0.023, respectively. The differences in the LST inversion results derived using the original and improved TES algorithm ranged from 0 to 1.86 K.

Resource development in mining areas alters land use patterns and causes ecological damage. This renders land use identification crucial to ecological restoration and management in mining areas. Although remote sensing imagery is widely used for land use classification, the use of a single data source has limitations in the classification for mining areas. Additionally, it is difficult for conventional machine learning algorithms to effectively perform the classification. To improve classification accuracy, this study investigated the eastern part of Dongsheng District, Ordos City as an example to conduct land use classification for mining areas using a convolutional neural network (CNN) combined with multi-source remote sensing data. First, a multi-source remote sensing time series feature set was developed using data from Sentinel-1/2, Luojia-1 01, and the NASA digital elevation model (DEM). Next, optimal features were selected using the Relief-F algorithm combined with a random forest algorithm. Finally, information on surface features was extracted using the ResNet50 CNN model. This facilitated land use classification in the mining area. The results show that the proposed method achieved an overall land use classification accuracy of 95.36% and a Kappa coefficient of 0.942 1, outperforming conventional methods such as the random forest approach. Furthermore, selecting optimal features using Relief-F combined with the random forest approach enhanced the classification accuracy of various classifiers. This study offers a methodological reference for land use classification of mining areas.

Wetlands, hailed as the "kidneys of the Earth", hold great significance for maintaining the stability of ecosystems. This study investigated 10 important wetland reserves along the Silk Road. Based on remote sensing data from the ZY3 satellite, it extracted the wetland types in 2015 and 2020 through interactions between object-oriented analysis and manual interpretation. As a result, a dataset of wetland distribution and its dynamic changes in the reserves was established. By combining topography, hydrological conditions, ecological importance, and wetland type transition, this study proposed a method for assessing the spatial potential of returning farmlands to wetlands. The results of wetland information extraction show that from 2015 to 2020, the wetland area in the 10 reserves exhibited a net increase of 238.04 km² thanks to both natural and anthropogenic factors. Such an increase was dominated by lacustrine wetlands, with the wetland rate rising by 0.58% generally. This demonstrates that the establishment of ecological reserves posed a positive impact on regional wetland protection. However, in local regions, wetlands still showed a trend of degradation, covering an area of 77.00 km². The potential analysis results of returning farmlands to wetlands indicate that a total of 325.13 km² of farmlands should be returned to wetlands, consisting of 10.63 km² requiring high-priority restoration, 167.02 km² subjected to medium-priority restoration, and 147.48 km² requiring low-priority restoration. The proposed region-specific scheme for ecological restoration in wetlands can provide decision-making support for wetland protection and management along the Silk Road.

A timely and accurate understanding of the spatial extents and distributions of industrial solid waste and open pits in mining areas is significant for the precise control of solid waste contamination and the ecosystem conservation. Remote sensing technology is an effective monitoring method. However, single-scale sample sets fail to fully represent the features of industrial solid waste yards and open pits with different shapes and sizes. Constructing multiscale sample sets may be effective in solving the problem of incomplete feature representation for different industrial solid waste yards and open pits, thereby enhancing the identification accuracy and generalization capability of models. By fully considering the differences in the shape and size of different industrial solid waste yards and open pits, this study proposed a remote sensing identification method for industrial solid waste and open pits based on the multiscale sample set optimization strategy. In the proposed method, a multiscale sample set was prepared based on the preprocessed data of the GF-1B, GF-1C, and GF6 satellite remote sensing images. Subsequently, a U-Net deep learning network model was constructed to identify industrial solid waste and open pits. Finally, the identification accuracy was compared with that of the single-scale sample set model. The results show that the U-Net deep learning network model based on the multiscale sample set achieved identification accuracy of 81.23 %, recall of 66.88 %, F1-score of 73.36 %, and average intersection over union of 73.55 %, suggesting improvements by 6.02, 1.02, 3.12, and 9.86 percentage points, respectively, compared to the single-scale sample set model. Overall, this study provides a reliable approach for precisely monitoring industrial solid waste and open pits.

Optical satellite remote sensing images of tropical and subtropical vegetation areas are often affected by cloud cover, leading to missing remote sensing information of surface features. Effectively detecting clouds, classifying clouds and objects, and extracting cloud cover information remain hot topics and challenges in remote sensing research. Many optical cameras in domestic satellites lack the short-wave and thermal infrared spectral bands, which are used in prevailing cloud detection algorithms, reducing the image data availability after cloud removal. Hence, this study suggested detecting the spatial distribution of cloud cover by utilizing only several spectral bands in the visible light - near-infrared range (400 nm to 1 000 nm). Based on the hyperspectral remote sensing images from the Zhuhai-1 satellite, this study constructed feature space scatter plots using spectral indices, including normalized difference vegetation index (NDVI) and normalized differential water index (NDWI), for cloud/object classification and detection. Moreover, this study extracted the cloud, water, and vegetation cover information from mixed pixels. The results demonstrate that compared to conventional cloud detection methods using spectral index thresholds, the cloud detection algorithm under the NDWI-NDVI feature space used in this study exhibited a superior cloud-water separation capability and simple operability. It can precisely describe the spatial distribution characteristics of cloud cover by suppressing the shadow effect on cloud cover. Overall, this study offers a novel technical approach for further developing cloud detection, cloud-water separation, and cloud cover information extraction algorithms for domestic optical satellite remote sensing data.

Exploring land use evolution and its impact on carbon storage is significant for mitigating climate change and promoting green and low-carbon development in metropolitan circles. Under the carbon peak and neutrality goals, this study implemented dual-constraint transition matrix optimization using point-of-interest (POI) data and the patch-generating land use simulation (PLUS) model, followed by the coupling with the integrated valuation of ecosystem services and trade-offs (InVEST) model. Based on the InVEST-PLUS coupled model, this study analyzed the land use evolution in the Jinan metropolitan circle from 2000 to 2020 and its impact on ecosystem carbon storage. Considering natural development, urban development, and ecological conservation as three distinct scenarios, this study simulated and predicted the land use change in the Jinan metropolitan circle in 2030 and 2060. Moreover, this study estimated the corresponding ecosystem carbon storage and analyzed the shift of the carbon storage center. Finally, this study explored the factors driving the spatial differentiation of carbon storage using the optimal parameters-based geographical detector (OPGD). The results indicate that from 2000 to 2020, the Jinan metropolitan circle saw a continued decrease in arable land, grassland, and unused land; a fluctuating increase in forest land; and a rapid increase in water area and construction land. The carbon storage and land use pattern in the Jinan metropolitan circle showed similar distributions characterized by higher values in the southeast and lower values in the northwest, with the main body of the Yellow River as the dividing line. The carbon storage in arable land served as the primary source of carbon storage in the Jinan metropolitan circle since it represented over 80 % of the total carbon storage. The simulation results reveal decreased carbon storage under the three scenarios, primarily due to the conversion from arable land in high carbon-density areas to construction land in low carbon-density areas. The ecological conservation scenario shows the highest total estimated carbon storage, which is 4 226.86×10⁶ t in 2030 and 3 967.94×10⁶ t in 2060. The carbon storage center in the Jinan metropolitan circle displays a certain shift in different development periods and scenarios due to land use change. However, the carbon storage center remains located in Licheng District, suggesting that the development of the Jinan metropolitan circle is relatively comprehensive and balanced. Various driving factors manifest significant impacts on the spatial distribution of carbon storage in the Jinan metropolitan circle. Notably, population density shows the greatest explanatory power for the spatial differentiation of carbon storage. Additionally, the interactions of various factors enhance their explanatory power for carbon storage.

This study investigated the coastal wetland of Jiangsu Province. Using methods such as satellite remote sensing and airborne multi-parameter remote sensing, this study estimated the biomass of dominant plants and estimated their carbon sequestration capacities. Based on fine-scale classification of surface features achieved using airborne hyperspectral data, this study extracted 11 land cover types. The vegetation cover was approximately 76%, and zones with human activities accounted for about 1.5% of the study area. The model for vegetation biomass inversion using the multi-parameter airborne remote sensing demonstrated higher accuracy than that based on satellite remote sensing, with a coefficient of determination (R²) greater than 0.8 and a root mean square error (RMSE) of 0.25. As calculated using the multi-parameter airborne remote sensing, Spartina alterniflora and reed within the study area exhibited aboveground carbon sequestration capacities of 0.41 kg/m² and 0.58 kg/m², respectively. This study demonstrates that the multi-parameter airborne remote sensing method can accurately determine vegetation types in wetlands and carbon sequestration capacity, thus providing crucial assessment parameters for research on the carbon cycle of the ecosystem and the current status of habitats within wetlands and precisely serving wetland resource surveys.

Plastic greenhouses have gained extensive application in modern agriculture. This, however, gives rise to ecological issues. Remote sensing data enable effective extraction and identification of plastic greenhouses on a large scale. Existing studies largely focus on plastic greenhouse extraction using either classification or spectral indices methods. However, there exists a lack of the combination and comparative analysis of both methods. This study proposed a method for plastic greenhouse extraction that integrates multiple Sentinel-2 spectral indices and a one-class classification method (improved one-class random forest). Furthermore, this study extracted information on plastic greenhouses using an improved one-class random forest method, as well as six spectral indices of plastic greenhouses as classification features. The extraction results were then compared with those of the proposed method to demonstrate the effectiveness of the latter. The results indicate that the proposed method yielded an overall accuracy of above 97% across four seasons, with kappa coefficients exceeding 0.82 and F1 scores of over 0.84. These metrics all were better than those yielded using the six spectral indices. Furthermore, the proposed method exhibited differences in the overall accuracy, kappa coefficient, and F1 score across four seasons of less than 1%, under 0.1, and below 0.1 respectively. This suggests the high seasonal stability of the method, outperforming the extraction results obtained by using spectral indices alone. This study provides a method for accurately monitoring the spatial distribution of plastic greenhouses.

At its 16th Plenary Session and Ministerial Summit, the Group on Earth Observations (GEO) proposed a new goal to build a “digital library for Earth observation applications”, highlighting the transition from “open data” to “open science”. It aims to achieve the management and sharing of knowledge resources, including data, algorithms, literature, and cases, thereby facilitating the comprehensive application and knowledge service provision of Earth observations in fields such as global change. Under this research background, this study systematically examined Earth observation data resources, including the conceptual system of Earth science variables, Earth observation satellites and payloads, observational and simulated data products, and open knowledge bases of academic literature. Based on the theories and techniques related to the Semantic Web and Knowledge Graph, this study established the Earth observation knowledge ontology with corresponding instances, involving Earth science variables, remote sensing satellites, observation payloads, observational and simulated datasets, journals, and academic literature. The knowledge representation results of this study will contribute to the representation, management, and integration of data and knowledge in the field of Earth observation applications. Moreover, they facilitate the discovery of potential associations between data and knowledge, enhancing the efficiency of scientific research and advancing scientific discovery.

Land use serves as the primary cause of global environmental changes. Therefore, investigating its spatiotemporal changes and corresponding driving factors is significant for promoting the sustainable development of regional socioeconomics and ecosystems. Based on nine stages of remote sensing monitoring data on land use/land cover from 1980 to 2020, this study analyzed the spatiotemporal changes in land use types in the Golmud River basin. By combining the analysis of significant correlations, this study explored the major factors driving changes in land use within the basin. The results indicate that over the past 40 years, unused land and grassland proved to be dominant land use types in the Golmud River basin. The areas of cultivated lands, water bodies, and construction lands exhibited an increasing trend, while those of forest lands, grasslands, and unused lands trended downward. The period from 2015 to 2020 witnessed significant changes in both the areas and the dynamic degrees of various land use types within the basin. During this period, spatial changes in land use transition predominately occurred in the central and northern parts of the basin. Between 1980 and 2020, the unused land showed significant fragmentation. Human activities, particularly total population and regional gross domestic product, were identified as the main factors driving changes in the land use type within the basin.

Hyperspectral remote sensing （HRS） technology，with its high spectral resolution and extensive spectral coverage，demonstrates significant potential in geological prospecting. Focusing on the Qianhongquan gold deposit in the Beishan orogenic belt，Gansu Province，this study conducted altered mineral mapping and component analysis，using HRS data from the AHSI sensor on the ZY-1 02D satellite and the self-developed hyperspectral mineral mapping technique，GeoAHSI，revealing their spatial distribution characteristics. Besides，ground-based spectral measurements were conducted on typical profiles to validate the spectral data，thereby assessing the reliability of the hyperspectral mineral mapping results. The results indicate that the primary altered minerals in the Qianhongquan gold deposit and its surrounding rocks include sericites （low-aluminum，medium-aluminum，high-aluminum，and iron-rich muscovites），calcites，dolomites，epidotes，and chlorites. Their distribution is closely related to ductile shear zones，with the distribution of sericites，chlorites，and epidotes being particularly significant within these zones. This spatial correlation provides critical indicators for regional prospecting. Additionally，it was observed that the 2 200 nm absorption feature of sericites and the 2 250 nm absorption feature of chlorites exhibit marked enrichment in silicon （Si） and iron （Fe） around ore bodies，which is closely correlated to the chemical compositions of the minerals. By enhancing the identification of weak spectral features，this study successfully applied HRS technology to mineral identification and spatial distribution analysis. These findings provide a scientific basis for further exploration of the Qianhongquan gold deposit and offer valuable references and guidance for the application of HRS in similar deposits.

Accurately identifying plant communities in coastal wetlands is critical for strengthening the ecological quality monitoring and enhancing the ecosystem functions of coastal wetlands. With the Yellow River Delta as the study area, this study constructed a feature vector set including phenological, optical, red-edge, and radar features based on Sentinel-1/2 image data using the Google Earth Engine (GEE) platform. It classified the wetland plant communities in the Yellow River Delta in 2021 using the random forest algorithm. Moreover, it explored the effects of phenological features in classification. The results reveal an overall classification accuracy of 97.91 % and a Kappa coefficient of 0.97. In 2021, the distribution areas of Phragmites australis, Suaeda glauca, Spartina alterniflora, and Tamarix chinensis were 49.91 km², 39.91 km², 79.36 km², and 20.86 km², respectively. The phenological features of typical plant communities in the Yellow River Delta wetlands were effectively extracted based on the normalized difference vegetation index (NDVI) time-series fitting curves. The highly distinguishable features included the maximum value date, base value, growth amplitude, beginning-of-season growth rate, and end-of-season decline rate. Compared to other feature variables, phenological features contributed more significantly to the overall classification accuracy, suggesting their prominent role in classification. The results of this study provide a methodological reference and scientific basis for the plant community monitoring and ecological conservation of coastal wetlands in the Yellow River Delta.

Given its all-day availability, all-weather adaptability, and high spatial resolution, the interferometric synthetic aperture radar (InSAR) technique has been widely applied in multiple fields, demonstrating strong adaptability and high practical value. Focusing on two typical landslide areas in Zhouqu County, Gansu Province, this study compared small baseline subset InSAR (SBAS-InSAR) monitoring results and Kalman filter prediction results with monitoring data from the global navigation satellite system (GNSS), confirming the reliability and accuracy of the SBAS-InSAR technique in monitoring landslide deformations. The results indicate that the SBAS-InSAR technique exhibited significant advantages in monitoring areas with deformations induced by geologic disasters, effectively overcoming the limitations of traditional monitoring means. This technique can provide critical technical support and scientific basis for early warning and management of geologic disasters in Zhouqu County and other areas prone to suffer these disasters.

Scientifically and rationally demarcating urban and town areas is a fundamental task during China’s rapid urbanization stage. It serves as a critical basis for promoting the optimization and quality improvement of urban and rural spaces, scientifically coordinating urban and rural planning and construction management, and implementing territorial spatial planning. However, there is neither a unified concept nor a universal delimitation method for urban and town areas in China, hindering their planning, construction, development, and public management. Based on defining the relevant concepts of urban and town areas, this study proposed a people-centered method for determining town areas with no cities and counties set using geographic information system (GIS) technology, considering the characteristics and spatial relationships of land types. The data sources of this study include the results of the third national land resource survey, statistical bulletins, remote sensing image interpretation, and point-of-interest (POI) data. Finally, the proposed method was applied to demarcate the town areas in Xiapu County, Ningde City. The empirical study results demonstrate the effectiveness of the proposed method, which features a scientific and concise technology roadmap and strong operability. Therefore, the proposed method can provide a theoretical foundation for the rational territorial spatial planning.

The ecological security pattern serves as an indicator of ecosystem health and sustainability, playing a crucial role in enhancing human well-being. This study identified ecological source areas in the Guanzhong Plain based on three ecosystem services, including water conservation, soil conservation, and habitat provision. Considering regional characteristics, this study selected soil erosion sensitivity index, normalized difference vegetation index (NDVI), and nighttime lighting as disturbance factors to correct the basic resistance surface and identify ecological corridors. The results indicate that the primary and secondary ecological source areas in the Guanzhong Plain cover 3 011.85 km² and 8 434.51 km², respectively, corresponding to 5.22% and 14.62% of the total area. These areas, characterized by mountainous terrain and high vegetation cover, are primarily distributed in the Qinling Mountains in the south, the hilly and gully regions in northern Baoji City, and the junctions of Xianyang, Tongchuan, and Weinan cities. The resistance surface correction for Guanzhong Plain reduced 61 ecological corridors (total length: 1 613.4 km), leading to significant changes in their distribution, and ultimately rationalizing corridor identification. Overall, this study provides a novel case for constructing regional ecological security patterns and data support for ecological conservation and urban planning in the Guanzhong Plain.

Understanding the characteristics of urban expansion and corresponding spatial changes serves as a prerequisite for optimizing urban spatial structure and resisting disorderly urban land expansion. This study focuses on the Chengdu-Chongqing economic circle. Using multi-source data fusion, this study extracted information on urban built-up areas from 2000 to 2020. From the aspects of urban expansion characteristics, spatial changes, and intercity spatial interaction intensity, this study analyzed the spatiotemporal evolution during urban expansion at both the regional and county scales. The results indicate that incorporating impervious surface information into multi-source data fusion improved the information extraction accuracy of built-up areas, achieving an overall classification accuracy of 98% and an average Kappa coefficient of 0.75. Urban expansion from 2000 to 2020 transitioned from low to medium-high speed and then to low speed. The dominant expansion type was edge expansion, accompanied by decreased spatial compactness. Within the Chengdu-Chongqing economic circle, the strongest spatial interaction intensity occurred between Chengdu and Chongqing. The urban spatial pattern exhibited a “dual cores with two wings” pattern, highlighting the pivotal role of Chengdu and Chongqing in driving the development of surrounding cities. These findings reveal the urban development patterns and spatial change characteristics within the Chengdu-Chongqing economic circle. They will facilitate the rational optimization of land use and territorial spatial patterns, promoting coordinated urban-rural development.

The abundant contextual information in synthetic aperture radar (SAR) images remains underutilized in deep learning-based ship detection. Hence, this study proposed a novel method for detecting ships from SAR images based on high-dimensional contextual attention and dual receptive field enhancement. The dual receptive field enhancement was employed to extract multi-dimensional feature information from SAR images, thereby guiding the dynamic attention matrix to learn rich contextual information during the coarse-to-fine extraction of high-dimensional features. Based on YOLOv7, a YOLO-HD network was constructed by incorporating a lightweight convolutional module, a lightweight asymmetric multi-level compression detection head, and a new loss function,XIoU. A comparative experiment was conducted on the E-HRSID and SSDD datasets. The proposed method achieved average detection accuracy of 91.36 % and 97.64 %, respectively, representing improvements by 4.56 and 9.83 percentage points compared to the original model, and outperforming other classical models.

Water bodies serve as one of the three major elements in maintaining wetlands. Their dynamic monitoring can effectively protect wetland ecosystems. Conventional methods for monitoring water bodies in wetlands employ field surveys or manual interpretation of remote sensing images, which are costly and inefficient, and inapplicable to continuous dynamic monitoring. In recent years, using methods like machine and deep learning to extract water body features from satellite remote sensing images has developed into an effective means for monitoring water bodies in wetlands. Based on the hyperspectral images from the ZY1-02D satellite, this study classified the water bodies in the Baiyangdian wetland using machine learning, convolutional and transformer neural networks. The accuracy and computational efficiency of water body classification under different spectral preprocessing methods and different image neighborhood sizes in training were compared to explore the optimal data preprocessing method and classification model for water bodies in wetlands. The results indicate that deep learning significantly outperformed machine learning in classification accuracy and computational efficiency. In particular, the spectral-spatial residual network (SSRN) model based on the convolutional neural network achieved the highest classification accuracy (OA: 99.09 %, Recall: 99.62 %, F1-score: 0.99) under conditions of all spectral bands and a 9×9 neighborhood size. Besides, despite a low signal-to-noise ratio, the atmospheric water vapor absorption band contained significant information, assisting in improving the classification accuracy of water bodies in the wetland during model training and prediction. The results of this study are expected to provide methodological support for the business operation of water body classification of wetlands.

The Yangtze River economic belt (YREB), spanning the east, middle, and west regions of China, serves as a pioneering demonstration belt for the construction of ecological civilization. Hubei Province, a member of the YREB, holds abundant mineral resources. Its long-term exploitation of mineral resources has caused ecosystem damage and social stability disruption, necessitating ecological restoration of abandoned open-pit mines. However, few studies concern the systematic tracking, monitoring, and analysis of the ecological restoration of abandoned open-pit mines in the YREB within Hubei Province. Based on the remote sensing data from the domestic high-resolution satellite from 2017 to 2022, and utilizing human-computer interaction interpretation, this study monitored the distributions and ecological restoration of abandoned open-pit mines within 10 km on both sides of the mainstems and major tributaries (Hanjiang River and Qingjiang River) in the YREB within Hubei Province. The results indicate that by the end of 2017, 537 patches of abandoned open-pit mines had been identified in the study area, involving 283 mines with a total area of 2 225.19 hm². The monitoring results of dynamic changes from 2018 to 2022 show that the ecological restoration and treatment area of abandoned open-pit mines increased from 291.01 hm² in 2018 to 1 741.19 hm² in 2022, manifesting a continuously rising treatment rate, suggesting overall improved ecological restoration and treatment results. Using the remote sensing ecological index (RSEI) values from 2017 to 2022, this study assessed the ecological restoration effects of abandoned open-pit mines. The assessment results reveal that the average RSEI value for the patches increased from 0.397 7 in 2017 to 0.423 9 in 2022, with a growth rate of 6.59 %, suggesting significantly improved ecological conditions and restoration effects. Overall, the monitoring and assessment results of the dynamic changes in ecological restoration of abandoned open-pit mines in the study area provide valuable data and methodological insights for monitoring abandoned open-pit mines in Hubei Province and other regions in China, highlighting the significance of this study.

Geological disasters, influenced by natural and human factors, directly threaten the safety of people’s lives and property. Exploring the spatiotemporal variations and development mechanisms of geological disaster risk can enhance disaster prevention and mitigation. This study examined 31 factors such as topography, rainfall, and social economy from the perspectives of nature and humanity. Based on the four-factor risk theory, this study investigated the variations of geological disaster risk in the western Sichuan region using methods like the analytic hierarchy process, principal component analysis, information value model, entropy weight method, and hot/cold spot analysis. Employing the obstacle degree model, this study explored the degrees of influence of various factors on geological disaster risk in the western Sichuan region. The results indicate that from 2007 to 2022, the geological disaster risk in the western Sichuan region was generally characterized by higher levels in the west and lower levels in the east. Kangding and Maerkang were the concentrated distribution areas of perennial cold spots. The area of extremely low and low risk levels increased by 8 871.1 km² and 12 478.6 km² respectively at growth rates of 1.056%/a and 1.485%/a respectively. The area of high and extremely high risk levels decreased by 10 127.8 km² and 9 880.1 km² respectively at growth rates of -0.02484 km²/a. The degrees of influence of various factors on risk levels exhibited temporal heterogeneity. The dominant obstacle factors (obstacle degree: above 5 %) were concentrated in risk and disaster prevention and mitigation indicators. Factors including rainfall, topography, and medical resources contributed significantly to geological disaster risk.

Nonnegative matrix factorization (NMF) is commonly used in hyperspectral image (HSI) unmixing due to its high interpretability and computability. To effectively address HSI noise and improve unmixing efficiency, this study proposed a method for hyperspectral unmixing based on spectral-spatial total variation nonnegative matrix factorization (SSTVNMF) with feature space augmentation. First, the original data space was converted to the feature space through feature extraction, allowing the unmixing process to be performed in the feature space for enhanced unmixing efficiency. Second, to reduce the impact of noise, the spatial information was extracted using the bilateral filtering (BF) method for enhanced feature extraction, thereby ensuring the accuracy of extracted features. Third, to ensure the effectiveness of the unmixing method, total variation (TV) regularization that considers both spatial and spectral features was established based on the NMF method. The spatial TV promotes abundance smoothing by calculating the horizontal and vertical differences in abundance between neighboring pixels. Based on the minimum-volume TV, the spectral TV enhances endmember extraction by applying constraint forces between endmembers to minimize the volume. Finally, the proposed method was verified using the synthetic data from the USGS spectral library as simulated data and the Jasper Ridge, APEX, and Cuprite datasets as actual data. The experimental results demonstrate that the proposed method outperformed other improved NMF-based methods in terms of qualitative and quantitative assessments.

The AMS-3000 wide-field three-linear-array stereoscopic aerial survey system (hereafter referred to as the AMS-3000 system) is China’s first independently developed airborne linear-array aerial survey system. It can obtain panchromatic and R-, G-, and B-band multispectral images. However, the data quality of these images lacks quantitative assessments and analyses. Focusing on the area along the Jinsha River within western Panzhihua City, Sichuan Province, this study assessed the spectral quality of the data obtained from the AMS-3000 system in terms of grayscale, texture, and energy features, and noise level. Moreover, this study compared the AMS-3000 system with the internationally recognized ADS100 aerial photography system and assessed the geometric accuracy of the AMS-3000 system using the 1∶2 000-scale terrain data. Additionally, this study analyzed the effectiveness of the AMS-3000 system in the geological industry by applying it to the investigation of the mineral resource exploitation status and geologic hazards. Overall, this study serves as a reference for the application promotion and improvement of the AMS-3000 system.

To quantify the mining disturbance of open-pit coal mines to surrounding ecosystems, this study investigated the Pingshuo mining area in Shanxi Province. Based on the pressure-state-response (PSR) model, seven types of assessment indicators were selected to construct the remote sensing ecological index of open-pit coal mining area (OMRSEI) through combination weighting. The validity of the OMRSEI was verified through correlation and comparative analyses. Moreover, the trend of ecological evolution in the study area for the next two years was predicted using the exponential smoothing method. The results indicate that the OMRSEI exhibited significant spatial correlation and validity, establishing it as an effective remote sensing indicator for ecological assessment in open-pit coal mining areas. The study area manifested an overall enhanced ecological quality from 2013 to 2023. Specifically, the Antaibao and Anjialing open-pit coal mines witnessed continuously improved ecological quality due to the progressive restoration of waste dumps. In contrast, the Dong open-pit coal mine displayed an ecological quality trend characterized by a first decline and then recovery. The average OMRSEI of the study area is predicted to continuously rise from 2025 to 2027, indicating sustained enhancement in ecological quality.

To explore the spatiotemporal evolution of the bottomland in Dongting Lake since the middle stage of the Republic of China, this study examined the historical topographic maps and aerospace remote sensing data concerning the study area for over 10 time periods since the 1930s. Based on remote sensing image interpretation, statistical data analysis, and historical comparison, this study analyzed the temporal variations in the bottomland area of Dongting Lake in various periods to infer the corresponding spatial distributions of the bottomland. The results show that the spatial development of the bottomland in Dongting Lake was primarily characterized by the rapidly advancing delta at the mouth of the east branch of the Ouchi River and Piaowei Islet in East Dongting Lake, the alluvial deposits along the Caowei and Songzhu rivers in the north of South Dongting Lake, and the “Jiangnan Grassland” landscape formed by the bottom uplift of Qili and Muping lakes. The bottomland area in Dongting Lake expanded from 1 622.17 km² in 1 938 to 1 962.28 km² in 2018, coupled with the 980.96 km² of reclaimed high bottomland, suggesting a net increase of 1 321.07 km². In terms of spatial distribution, the bottomland area exhibited an undulating trend rather than a continuous increase. It manifested a significant expansion from 1938 to 1948 and 1958 to 1998 but a slow shrinkage from 1948 to 1958 and 1998 to 2018. Overall, the results of this study provide objective data for preserving lakeshore ecosystems and biodiversity and serving ecological restoration and environmental conservation in the Yangtze River basin.

In satellite image-based law enforcement for minerals, using unmanned aerial vehicles (UAVs) for investigation and aerial photography in mining areas with complex geographic conditions and unknown staffing situations, and photogrammetry for three-dimensional modeling of mining areas serves as an effective means to ensure the personal safety of field workers and improve work efficiency and accuracy. Lightweight UAVs are more applicable to satellite image-based law enforcement for highly mobile minerals due to their flexible take-off and landing conditions and high maneuverability while ensuring shooting clarity and modeling data requirements. Based on the UAV aerial survey results of several mining faces in Liaoning Province, this study demonstrates that satellite image-based law enforcement for minerals assisted by lightweight UAVs can significantly improve the efficiency and safety of fieldwork. Moreover, the modeling results provide data support for subsequent accurate survey and analysis and multi-temporal monitoring in mining areas.

Riparian zones have been extensively used in non-point source pollution control projects worldwide, and remote sensing has gradually become a significant means of non-point source pollution research. However, combining remote sensing technology with riparian zones for efficient pollution interception effects is still a challenge. With the Xingyun Lake basin in Yunnan Province as the study area, this study established a soil and water assessment tool (SWAT) model by coupling with remote sensing. It created a riparian zone by changing the land use type for scenario simulation, investigating the different effects of various widths and vegetation types on pollutant reduction. The key findings are as follows: ①The created riparian zone exhibited better interception effects for nitrogen compared to phosphorus; ② Concerning different vegetation types in the riparian zone, forest land manifested significantly better pollution interception effects than grassland. Moreover, the pollutant reduction rate gradually increased with an increase in the width of the riparian zone; ③A riparian zone consisting of 30-m-wide forest land and 30-m-wide grassland can reduce total nitrogen production by 5.20% and total phosphorus production by 6.03% while intercepting 19.83% of organic nitrogen and 21.30% of organic phosphorus into the lake, demonstrating the optimal pollution interception effects.

Investigating land-use-related carbon emissions (LCE) plays a vital role in achieving goals of peak carbon dioxide emissions and carbon neutrality (also referred to as the “dual carbon” goals). Research on the changes and prediction of LCE in Xiangxi Tujia and Miao Autonomous Prefecture (also referred to as the Xiangxi Prefecture) can provide a theoretical reference for the region to develop policies on the achievement of the “dual carbon” goals and for local balanced development and protection. Based on five sets of land use data from 2000 to 2020, this study analyzed the land use conditions and the spatiotemporal evolution of historical carbon emissions in Xiangxi Prefecture. The factors influencing LCE were determined using a decoupling model and a logarithmic mean Divisia index (LMDI) model. Furthermore, three land use scenarios were established: natural development, priority of cultivated land protection, and ecological protection priority. Using these scenarios, this study predicted the land use and carbon emissions in Xiangxi Prefecture in 2030. The results indicate that forest land represents the dominant land use type in Xiangxi Prefecture. Regarding land use transition, the period from 2000 to 2020 witnessed a significant increase in construction land, which encroached into substantial areas of forest land and cultivated land. Concurrently, water bodies and grassland decreased in area, being converted into forest land and cultivated land. From the perspective of carbon emissions, land use in the region exhibited a transformation from carbon sinks to carbon sources in general. During the 20-year span, the total LCE continued to increase. Construction land was identified as the primary land type as a carbon source, while forest land was the main land type as a carbon sink. Within the 20 years, carbon emissions decreased only in Huayuan County but increased in all other counties and cities. After 2010, the original regions with elevated carbon emissions showed a decrease in carbon emissions, while other regions witnessed growing carbon emissions to varying degrees. These regional changes in carbon emissions were largely attributed to the increased carbon emissions from construction land. Xiangxi Prefecture maintained a weak decoupling effect generally, with counties and cities fluctuating between weak decoupling and strong decoupling states. The economic output effect and energy efficiency effect served as the primary factors influencing carbon emissions. The overall land pattern remained relatively stable across the three scenarios. The carbon emissions of the three scenarios increased in the order of ecological protection priority, natural development, and priority of cultivated land protection. In the future, construction land will still represent the dominant factor causing overall changes in carbon emissions, while forest land will remain as the primary carbon sink.

The change in carbon stocks is recognized as an important indicator of the carbon pool function. The effective, accurate assessment of carbon stocks is of great significance for research on regional carbon cycle and carbon source/sink dynamics, climate change mitigation, and regional sustainable development. Based on multi-time series remote sensing images (Sentinel-1 and Sentinel-2) and the Google Earth Engine (GEE) cloud computing platform, this study matched the photon point data of ICESat-2-derived vegetation canopy for the inversion of mangrove forest heights. Then, the inversion of the biomass of mangrove forests was conducted using a large-scale tree height-biomass model. Consequently, the heights, above-ground biomass, and carbon stocks of mangrove forests in Hainan Island were obtained, and their distribution and variations were further analyzed. The results indicate that in 2016, 2019, and 2022, mangrove forests in Hainan Island exhibited average heights of 6.99 m, 7.26 m, and 7.71 m, respectively, with an increasing trend observed in the highlights across all regions in the three years. Their total above-ground biomass was 400 939.48 t, 411 928.64 t, and 458 759.02 t, respectively, with average densities of 110.23 t/hm², 114.61 t/hm², and 120.02 t/hm², respectively. The above-ground biomass of Dongzhai Port and the Bamenwan area, both located in the northeastern part of Hainan, accounted for about 80% of the total. The carbon stocks of mangrove forests exhibited rates of increase ranging from 1% to 4.45% over the three years, with the top two growth rates occurring in Dongzhai Port and the Bamenwan area, respectively (4.45% and 3.17%). The results of this study can provide foundational data and a methodological reference for assessing carbon stocks of large-scale mangrove forests and serve as important parameters for mangrove forest management and protection in Hainan Island, holding THE value of widespread applications.

Surface deformations pose significant threats to the normal operation of railways. Investigating the spatial distribution of surface deformations along the China-Laos railway holds great significance for disaster prevention and mitigation. Based on 36 scenes of ascending orbit and 50 scenes of descending orbit images from Sentinel-1A satellite from December 2021 to August 2023, this study conducted deformation inversion using the small baseline subset interferometric synthetic aperture Radar (SBAS-InSAR) technique. Besides, this study conducted spatial distribution statistics and analysis of surface deformations along the Jinghong section of the China-Laos railway. The results indicate that the overall deformation along the railway exhibits a heterogeneous distribution, with multiple potential hazards in the northern mountainous area. Among the selected typical deformation zones, the maximum subsidence rate in the northern mountainous area reaches -108.718 mm/a, whereas the southern plain area shows significant uplift with a rate of 227.315 mm/a. Along the railway, the surface deformation rates in the line of sight (LOS) direction ranged from -319.811 mm/a to 321.638 mm/a. Obvious subsidence occurred in Puwen Town and Dadugang Township. Conversely, minor subsidence was observed in urban areas like Mengyang town, Yunjinghong subdistrict, and Gasa town, with pronounced uplifts in the southeastern part of Menghan town. Along the railway, deformations in mountainous areas were primarily concentrated at elevations ranging from 800 m to 1400 m, with soft rocks prevailing in these deformed areas. InSAR-based analysis of the spatial distribution of the surface deformations along the railway is of significant value for the safe operation of the railway.

Land use demands vary under different development objectives. Scientifically and rationally regulating changes in land use are crucial to efficient land resource utilization and achieving ecological, developmental, and economic coordination in the Pearl River Delta urban agglomeration. Based on the land use data of the urban agglomeration of 1990, 2000, 2010, and 2020 and using the FLUS-Markov model, this study predicted the quantity and spatial changes in land use in the Pearl River Delta urban agglomeration by 2035 under three scenarios: natural development, ecological protection, and development priority. Furthermore, this study determined the differences in land use change under the three scenarios. Additionally, a simulation analysis of the land use in 2035 was conducted to facilitate the optimized land and space allocation under varying developmental objectives. The results indicate significant changes in the use of construction land in the Pearl River Delta urban agglomeration. From 1990 to 2020, the area of construction land, including urban land and infrastructure land increased by 4 945.25 km², representing an increase of 2.8 times. The simulations and predictions under three land use scenarios reveal that the urban land area will trend upward by the end of 2034, with its expansion speed being restricted under the ecological protection scenario, while the ecological land, such as forest land, grassland, and water area, will maintain an increasing trend until 2035. From 1990 to 2020, the arable land area decreased by 3 759.5 km². Under the three land use scenarios, the trend of arable land reduction will continuously decrease until 2035, with the decreasing trends slowing down from 2020 to 2035. Especially, under the development scenario, the area of construction land will continue to increase, the decreasing trend of the arable land area will be somewhat curbed, while the area of grassland and forest land will undergo a more serious decrease. Although dominant factors affecting arable land protection in the Pearl River Delta urban agglomeration vary across different development stages, the main factor is infrastructure construction such as rail transit roads.

Investigating the regional economic development of Yunnan Province-a radiation center facing Southeast Asia-and Myanmar-a country along the Belt and Road Initiative-is of great significance for promoting the construction of a China-Myanmar community with a shared future. Based on NPP/VIIRS nighttime light data，as well as spatial analysis methods including the centroid model，standard deviation ellipse，and Moran's I index，this study analyzed the spatiotemporal characteristics of economic development in the Yunnan-Myanmar region from 2013 to 2022. The results indicate a significant correlation between nighttime light and gross domestic product （GDP） data in the Yunnan-Myanmar region. From 2013 to 2022，the total nighttime light intensity in the Yunnan-Myanmar region showed a steadily increasing trend. From the perspective of the characteristics of economic development direction in the region，the economic centroid generally shifted southwestward first and then northeastward. The area of the standard deviation ellipse trended upward from 2013 to 2020 but trended downward in 2022. The long axis of the ellipse showed an increasing trend before 2020 but decreased slightly thereafter，while the short axis showed a stable increasing trend. The azimuth remained largely unchanged. In terms of the spatial correlation of economic development in the region，areas with high nighttime light intensity were primarily concentrated in the central Yunnan urban agglomeration，while those with low nighttime light intensity were mainly distributed in the eastern and western parts of Myanmar. This study can provide a valuable reference for economic and trade exchanges between China and Myanmar，as well as for the implementation of the Belt and Road Initiative.

Interferometric synthetic aperture Radar (InSAR) technology serves as a significant approach for analyzing surface subsidence in large-scale mining areas. Accurately predicting surface subsidence plays a significant role in preventing geological disasters. The long short-term memory (LSTM) network model faces challenges in parameter selection, while the rime optimization algorithm (RIME) is susceptible to local optimum and dependence on the initial solution. Considering these challenges, this study proposed a surface subsidence prediction model with LSTM optimized by the improved rime optimization algorithm (IRIME). The IRIME incorporated chaotic mapping, the improved Lévy flight mechanism, and the global exploration strategy of the hunter-prey optimizer (HPO). The proposed model is also referred to as the IRIME-LSTM model. With the Honghui coal mine as the study area, this study obtained the subsidence time series of highly coherent points in the mining area using small baseline subset (SBAS)-InSAR technology. Subsequently, this study conducted multi-step predictions of these highly coherent points using the IRIME-LSTM model, with the prediction results compared with the InSAR monitoring data. The results of this study indicate that the IRIME-LSTM model yielded a root mean square error (RMSE) of 2.65 mm, a mean absolute error (MAE) of 1.59 mm, and a mean absolute percentage error (MAPE) of 3.92 % in the overall test set. Compared to the RIME-LSTM and GS-LSTM models, the IRIME-LSTM model reduced the RMSE by 37.20 % and 51.73 %, the MAE by 42.60 % and 56.32 %, and the MAPE by 35.63 % and 50.51 %, respectively, demonstrating its high reliability and feasibility.

The Baige landslide occurred twice in October and November 2018, causing huge economic losses and extensive social impact. Monitoring the activity characteristics of the Baige landslide in various stages based on multisource data is significant for understanding the failure mechanism of this landslide. With Sentinel-1, ALOS-2, and Landsat8 data as data sources, this study derived the deformation characteristics of the Baige landslide before, during, and after two slide events using techniques, such as small baseline subset-interferometric synthetic aperture radar (SBAS-InSAR), SAR offset tracking, and optical offset tracking. The optical offset calculation results show that from November 2014 to March 29, 2018 in the pre-sliding stage, the cumulative displacement of the Baige landslide reached 40 m, with deformation concentrated in the middle of the landslide. The SAR offset results indicate that the cumulative displacement of the landslide reached 6.4 m in May and July 2018 in the pre-sliding stage, with deformation also concentrated in the middle of the landslide. The InSAR-based monitoring results reveal that after the two failures of the Baige landslide in October and November 2018, significant residual deformation remained in the trailing edge and upper left side of the landslide. From November 2018 to November 2021 in the post-sliding stage, the Baige landslide exhibited a deformation rate of -140 mm/y at the high trailing edge of the landslide, and the deformation range on the upper left side continued to expand. All the results of this study reconstructed the whole sliding process of the Baige landslide subjected to large displacements, providing a valuable reference for the monitoring and early warning of landslides.

Landslide disasters are frequent and widespread in southwestern China. The accurate identification and mapping of landslides using remote sensing imagery are of great significance for disaster prevention and mitigation. However，in complex environments，traditional remote sensing detection methods are often prone to misidentification due to background noise in the imagery. This paper proposed a dual-fusion landslide detection network （DLDNet） to improve landslide detection accuracy under challenging conditions. First，based on existing landslide samples，landslide simulation was conducted in complex environments using data augmentation techniques. Second，the ConvNeXt was adopted as the feature extraction backbone of DLDNet to capture more complex landslide features. Then，an attention module enhanced with deformable convolution was introduced to better focus on landslide-related information. Finally，a dual-fusion feature pyramid network （DFPN） was designed to thoroughly integrate feature information across different scales and receptive fields. The experimental results show that the proposed DLDNet achieved average precision （AP） scores of 56.9% for bounding box detection and 52.5% for segmentation，10.4 and 10.7 percentage points higher than those of the baseline model （Mask R-CNN）. Compared with other landslide detection models，the DLDNet demonstrates higher detection accuracy and a lower false alarm rate. The method，characterized by accurate landslide detection in complex environments，can support rapid landslide identification and emergency response.

Siling Co, the largest lake in Xizang Autonomous Region, has expanded significantly in the past few years, threatening surrounding pastoral activities, infrastructures, and even the ecological environment. This study systematically reconstructed the time series of changes in the lake area, water level, and water volume of Siling Co from 1995 to 2023 using optical images from satellites Landsat and GF, as well as altimeter data from satellites ERS-2, ICEsat, Cryosat-2, and ICEsat-2. Through Mann-Kendall trend analysis, the study determined the stages of the lake area changes and revealed the key characteristics of various stages. Furthermore, it also made a preliminary judgement on the inundation trend and its impacts. The results indicate that from 1995 to 2023, Siling Co experienced an increase in water area of 676.75 km² (with an annual average of 24.17 km²/a), a water level rise of approximately 13.32 m (with an annual average of 0.48 m/a), and a water volume growth of 28.45 Gt (with an annual average of 1.02 Gt/a). The changes in Siling Co from 1995 to 2023 can be divided into four stages: the fluctuating growth stage from 1995 to 2000, the rapid expansion stage from 2000 to 2011, the relatively stable stage from 2011 to 2017, and the re-expansion stage from 2017 to 2023. The inundated areas during the fluctuating growth and rapid expansion stages were primarily concentrated in the northern and southern parts of the lake. During the relatively stable stage, no significant expansion was observed in the inundated areas. In the re-expansion stage, the inundated areas were distributed in the eastern part of the lake. The continuous rise in the water level of Siling Co led to an annually increasing risk of surrounding inundation. Currently, the areas exposed to a high inundation risk are primarily concentrated along the south bank of the lake, which should be the focus in future monitoring and research.

Rapid and uneven land subsidence severely threatens human life and production activities. Understanding the spatiotemporal evolutionary patterns of land subsidence is crucial for the precise prevention and control of geological disasters. Employing the persistent scatterer interferometric synthetic aperture Radar (PS-InSAR) technology, this study obtained the information of monthly surface deformation in Dezhou City to calculate the multi-scale subsidence vulnerability indices (SVI). Combining time series cluster analysis, space-time cube, spatiotemporal hot spot analysis, and spatiotemporal outlier analysis, this study explored the spatiotemporal distribution characteristics of multi-scale SVI in Dezhou City from July 2017 to December 2021. The time series cluster analysis reveals inconspicuous trend clustering on a monthly scale, and significant clustering characteristics on quarterly and semi-annual scales, with large subsidence fluctuations on a semi-annual scale. The space-time cube model presents poor continuity of SVI and subtle subsidence variations on a monthly scale. In contrast, the subsidence on quarterly and semi-annual scales exhibited relatively close occurrence time, showing a significant pattern of subsidence from March to August and rebound from September to February of the ensuing year. The spatiotemporal hot spot analysis of SVI in Dezhou City for 54 months shows that enhanced and continuous subsidences occurred primarily in the northwest of Wucheng County and Decheng District. Oscillatory subsidence and rebound occurred principally in Linyi, Yucheng, and Qihe counties in the southeast. The local outlier analysis of multi-scale SVI shows nonsignificant subsidence characteristics on a monthly scale but similar subsidence conditions on quarterly and semi-annual scales. Seasonal subsidence and semi-annual subsidence related to crop growth in Linyi and Qihe counties gradually weakened or even rebounded. Notably, the high-high clustering range on a semi-annual scale was broader, accompanied by a more significant rebound.

Constructing a regional green infrastructure (GI) network can alleviate the contradiction between land use and ecological development in the process of rapid urbanization, playing a significant role in future urban planning. This study investigated Lincang City, a typical mountain city in Southwest China. Employing the patch-generating land use simulation (PLUS) model, this study predicted the land use and land cover (LULC) in Lincang City in 2030 under the ecological priority scenario. Furthermore, this study extracted information about the ecological source areas and corridors by integrating the morphological spatial pattern analysis (MSPA), minimum cumulative resistance (MCR) model, and circuit theory. Finally, this study constructed an optimized GI network for 2030 adapted to the sustainable expansion of Lincang City. The results show that from 2020 to 2030, the construction land area in Lincang City is projected to expand by about 23 %, while forest land and grassland will decrease by 0.2 % and 1.3 %, respectively. The water area is expected to increase by 46.9 % under reasonable management and protection. The core zone of GI landscape elements will represent 56.12% of the total area, while the edges will make up 21.3%. The spurs, bridging zones, islets, perforations, and circuits will constitute the rest 22.6%. Under the sustainable urban expansion scenario, the GI scale remains overall stable, with a relatively scattered distribution in built-up areas. The optimized GI network will involve 12 ecological source areas and 24 ecological corridors. The GI network of Lincang City in 2030 constructed based on the MSPA-PLUS model strengthens the understanding of the GI network for the sustainable development of a mountain city, adapting to future urban development. This study provides novel insights into the urban planning of mountain cities including Lincang and critical implications for GI protection and regional ecological security maintenance.

Forest above-ground biomass （AGB） is recognized as an important indicator of forest productivity. Rapid and accurate estimation of forest AGB is crucial for sustainable forest management and carbon cycle research. Based on spaceborne light detection and ranging （LiDAR） data from the global ecosystem dynamic investigation （GEDI） and Sentinel-2 optical data，this study extracted GEDI L2B，Sentinel-2 remote sensing features，and topographic factors （elevation，aspect，and slope） in the study area. Among them，variables were determined through Pearson correlation analysis. Then，this study constructed the partial least squares regression （PLSR），gradient boosting regression tree （GBRT），and random forest （RF） models for forest AGB inversion. Consequently，this study estimated these models’ potential for forest AGB estimation and analyzed the spatial distribution differences of forest AGB. The results indicate that the estimation using multi-source data consistently outperformed that using single-source data. Among them，the RF model based on GEDI and Sentinel-2 data exhibited the best performance （R²=0.76，root mean square error （RMSE）=23.02 t/hm²），followed by the GBRT model，while the PLSR model performed the worst （R²=0.26）. In terms of spatial distribution，within the elevation range of 1 200~1 800 m，forest AGB density increased with elevation. Slope variation had little effect on forest AGB density，but a pronounced decrease in AGB density was observed on steep slopes. Aspect analysis showed that semi-shaded and sunny slopes exhibited high forest AGB density，while shaded and semi-sunny slopes presented similar values. Slope-aspect interaction analysis revealed that sunny and semi-sunny slopes displayed the highest total forest AGB on gentle and moderate slopes，respectively. In contrast，forest AGB significantly decreased across all orientations on flat and steep slopes，with a more significant decline observed on shaded and semi-shaded slopes. These findings provide a scientific basis for formulating forest protection and cultivation policies at the provincial level.

In recent decades, the surface mass balance (SMB) and the calving of outlet glaciers have accelerated the mass loss of the Greenland Ice Sheet (GrIS), with SMB’s contribution continuing to increase. Therefore, determining SMB’s spatiotemporal distribution is critical for understanding the mass balance of the GrIS. Currently, the regional climate model and the remote sensing observation of outlet glacier flux gates serve as two primary calculation methods for the GrIS’s SMB. However, the former method results in large uncertainties in the SMB simulation. The latter method can only indirectly estimate the overall SMB value for the upper reaches of the flux gate, failing to reflect the spatial distribution of SMB. This study proposed a method for estimating the GrIS’s SMB based on remote sensing data and ice flux divergence, obtaining the relatively accurate spatial distribution of SMB. First, the interannual variation in the elevation of the GrIS was derived from ICESat-2 laser altimetry data. Second, based on MEaSUREs-derived glacier flow velocity data and BedMachine-derived ice thickness data, the ice flux divergence was calculated using the pixel-based finite difference method to estimate the GrIS’s elevation changes caused by glacier flow. The GrIS’s elevation changes caused by SMB were then obtained by subtracting the elevation changes caused by glacier flow from the ICESat-2 elevation data. Third, through the firn densification model, the elevation changes caused by SMB were converted into mass changes to reflect the interannual spatial distribution of the GrIS’s SMB. The proposed method was applied to estimate the spatial distribution of the GrIS’s SMB in 2019 and 2020, yielding relatively high accuracy (RMSE=0.519 m w.e.) in comparison with the measured SMB from the observation station, and outperforming the regional climate model (RMSE=0.565 m to 0.877 m w.e.), ultimately demonstrating its reliability.

The analysis of patches showing changes in coastal areas of the Chinese mainland tends to encounter challenges such as low image resolution, long time intervals, and limited spatial coverage. This study aims to obtain high-frequency, accurate information on changes in coastal areas nationwide. This will facilitate the dynamic monitoring of marine resources and the implementation of relevant protection policies for coastal areas in China. To this end, using domestic high-resolution remote sensing data of 15 days (i.e., one cycle), as well as the iteratively reweighted multivariate alteration detection (IR-MAD) algorithm combined with visual interpretation, this study extracted patches reflecting 2020—2023 changes along the coasts of 11 provinces and cities in the Chinese mainland. Accordingly, this study analyzed their spatiotemporal characteristics, landscape patterns, and spatial correlation. The results indicate distinct directional changes in the patches. The patches reflecting changes from sea enclosure to reclamation exhibited the largest areas across various investigated areas. Except for Hainan Province, the area of this type of patches exceeded 1 000 km². The proportions of patches reflecting different types of changes gradually tended to be balanced. In the winter of 2022, the proportion of patches showing changes in the reclamation dropped below 50% for the first time. The aggregation degree of patches reflecting various types of changes showed increasing trends, suggesting that patches reflecting various changes will become more concentrated in the future. The centroids of these patches of various regions shifted in varying directions, and these patches exhibited significant spatial correlation within a 20 km range.

To determine the origin of surface substrate for soil salinization and alkalization in the Songnen Plain, this study investigated Songyuan City based on Sentinel-2 multispectral images. Considering various commonly used indices like the soil salinity index (SSI), soil water index (SWI), and vegetation index (VI), this study constructed the optimal 3D spectral feature model to calculate the soil salinization-alkalization index (SSAI) for inversion of the soil salinization-alkalization status. Surface water and groundwater in areas subjected to soil salinization and alkalization were sampled to test their salt ion concentrations, followed by the analysis of salt ion sources according to the groundwater levels. The surface substrate was explored through planar grid layout and vertical stratified sampling. A total of 2 362 soil samples were collected in various layers within a depth of 5 m to test their pH and texture for the construction of a 3D surface substrate model. The results of this study reveal a positive linear correlation between the inversion result of remote sensing data and the topsoil salt content (coefficient of determination: R²=0.74). The study area was characterized by alkalization of sodium bicarbonate, with soil salt ions originating primarily from groundwater. The deep multilayer argillaceous soils acted as an aquiclude to prevent the downward infiltration, migration, and dilution of salt ions along with water. This surface substrate condition serves as the objective cause of soil salinization and alkalization in the study area.

The deep learning-based models currently available for detecting changes in remote sensing images face several challenges, including limited multi-granularity, poor classification performance of networks, high sensitivity to parameters, and great efforts in parameter adjustment. To address these challenges, this study proposed a deep forest-based model for detecting changes in remote sensing images. Initially, preliminary results were determined using a basic change detection method. Then, the results were optimized using the multi-granularity scanning characteristics and strong data classification of deep forest sub-networks. In this manner, the final change detection results were obtained. Verification experiments conducted on the LEVIR-CD and SYSU-CD datasets using various common change detection models indicated that the proposed deep forest-based model significantly outperformed other models in terms of precision, F1 score, and recall. Additionally, the proposed model exhibited strong adaptability on small datasets, as verified by loss function comparison, small-sample experiments, and ablation studies. This adaptability can reduce the complexity of parameter adjustment and address the issues that other deep learning sub-networks fail to be applicable to medium and small datasets.