Daily mean temperatures, as a primary indicator reflecting climatic characteristics, play a decisive role in monitoring urban heat island effects and agroecological environments. However, daily mean temperatures measured at meteorological stations lack spatial representativeness in regional-scale models. By contrast, the inversion results of daily mean temperatures using remote sensing data can better accommodate the large-scale monitoring needs, but with insufficient accuracy and quality. This study presented several common estimation methods for daily mean temperatures using remote sensing data, including multiple linear regression, machine learning, and feature space-based extrapolation. Then, based on the principle and process for estimation of daily mean temperatures using remote sensing data, this study systematically analyzed the effects of uncertainties such as clouds and aerosols and offered corresponding solutions. Finally, this study predicted the development trend of such estimation methods. Additionally, this study posited that image fusion and multi-source data fusion at different transit times can significantly improve the estimation accuracy under cloud interference.

Cloud detection, as a crucial step in preprocessing optical satellite images, plays a significant role in the subsequent application analysis. The increasingly enriched optical satellite remote sensing images pose a challenge in achieving quick cloud detection of numerous multi-source satellite remote sensing images. Given that conventional cloud detection exhibits low accuracy and limited universality, this study proposed a multi-scale feature fusion neural network model, i.e., the multi-source remote sensing cloud detection network (MCDNet). The MCDNet comprises a U-shaped architecture and a lightweight backbone network, and its decoder integrates multi-scale feature fusion and a channel attention mechanism to enhance model performance. The MCDNet model was trained using tens of thousands of globally distributed multi-source satellite images, covering commonly used satellite data like Google and Landsat data and domestic satellite data like GF-1, GF-2, and GF-5 data. Several classic semantic segmentation models were used for comparison with the MCDNet model in the experiment. The experimental results indicate that the MCDNet model exhibited superior performance in cloud detection, achieving detection accuracy of over 90% for all types of satellite data. Additionally, the MCDNet model was tested on the Sentinel data that were not used in training, yielding satisfactory cloud detection effects. This demonstrates the MCDNet model’s robustness and potential for use as a general model for cloud detection of medium- to high-resolution satellite images.

Hyperspectral data, exhibiting technical advantages in the spectral dimension, have been extensively used for accurately identifying surface features, particularly mineral information. Mineral identification relies on hyperspectral reflectance products, necessitating the application of proper atmospheric correction methods to obtain high-precision surface reflectance products that meet application requirements. Hence, three commonly used atmospheric correction models, ATCOR, FLAASH, and QUAC, were utilized to correct the hyperspectral data acquired by the ZY-1 02D satellite. Moreover, a comparative analysis was conducted on their visual effects, spectral analysis of typical surface features, and extraction of mineral information. The results are as follows: ① All three atmospheric correction models can effectively enhance image clarity in terms of visual effects. Specifically, the ATCOR model slightly outperformed the FLAASH and QUAC models; ② The correlation coefficients (R²) between the typical surface feature spectra of the three models and the ASD-measured spectra showed average values exceeding 0.7, suggesting high consistency and accuracy. Especially, the imaging spectra derived from the inversion results of the ATCOR model were more similar to the ASD-measured spectra; ③ The three models yielded relatively consistent results in chlorite identification but divergent results in sericite identification. Comparatively, the FLAASH and QUAC models exhibited high omission rates in surface regions with low sericite content. Overall, all three models can achieve satisfactory atmospheric correction effects, but the ATCOR model is superior to the other two models in mineral identification.

Interferometric synthetic aperture Radar (InSAR) can extract three-dimensional information about a ground target from the phase information in an interferogram. Phase unwrapping is an important step in the InSAR process, and its accuracy dictates the accuracy of the digital elevation model (DEM) or the ground deformation information. To overcome the serious phase decorrelation and phase noise in complex mountainous areas, this study divided the study area according to the quality of interference phases and proposed a quality-guided least squares phase unwrapping algorithm. Then, the algorithm was employed for the phase unwrapping of simulated low-noise interferometric phase data and Sentinel-1A InSAR interferometric images of the Qinling area of China. The results show that the proposed algorithm can effectively improve the phase consistency among high- and low-quality zones and the overall accuracy of phase unwrapping.

The combination of remote sensing and deep learning is efficient in the monitoring and evaluation of natural resources. Based on the comprehensive consideration of the characteristics of Zhuhai-1 OHS hyperspectral images, this study established the 3D-HRNet architecture by introducing the 3D convolutional module into the HRNet architecture and applied it to the semantic segmentation model for natural resources survey and monitoring. Using remote sensing images, this study established an ecological index (EI) evaluation model by calculating the species richness index, vegetation index, water network density index, land stress index, pollution load index, and environmental regulation index. Then, the model was employed to monitor and evaluate the natural resources in partial areas in the northern part of Beijing. The results show that: ① when being used to extract the natural resources, the 3D-HRNet model yielded average overall accuracy, a F1 score, and a Kappa coefficient of 0.83, 0.83, and 0.73, respectively, which were 0.04, 0.04, and 0.06 higher than those of the HRNet model, and 0.04, 0.05 and 0.06 higher than those of 3D-CNN model, respectively. This suggests that the 3D-HRNet model can extract natural resources from hyperspectral images more effectively than the HRNet model. In other words, the 3D convolutional module can utilize the inter-hyperspectral features to extract information more effectively; ② The eco-environment of partial areas in north Beijing in 2020 was evaluated using the EI evaluation model, with an average EI value of 68.2. This reflects a good ecological state in the study area, highly consistent with the conclusion of the Report on the State of the Ecology and Environment in Beijing, demonstrating the feasibility of remote sensing for ecological assessment. Therefore, this study provides an innovative method for the spatio-temporal analysis of the regional ecological state.

In recent years, the research on residents’ carbon emissions has mostly focused on the economic level and direct energy consumption, and less involved in the area of residential areas, and most of the research has relied on traditional surface measured data. In order to improve data accuracy and make more targeted policies, this paper selected China as the research object by taking advantage of the features of strong timeliness, wide coverage and small constraints of remote sensing images, and analyzed the correlation between residential area and residential carbon emissions in China in 2019. After determining the significance of the two, combined with the influencing factor of GDP, a multiple linear regression model was established between residents’ carbon emissions and residential area and GDP. The results show that there is a linear correlation between residents’ carbon emissions and the area of residential areas and GDP. With the development of economic level, the expansion of residential area is the main driving force for the increase of residential carbon emissions, and the driving effect of GDP on the increase of residential carbon emissions has decreased. Therefore, it is necessary to reasonably control the expansion of residential areas while considering economic development, so as to make more refined emission reduction policies and achieve the country's future green and low-carbon goals.

Granite pegmatites serve as a significant carrier and prospecting marker of granite pegmatite-type lithium deposits. The southeastern Zhaka area in Tianjun County, Qinghai Province demonstrates considerable prospecting potential for lithium deposits. Nevertheless, its high altitudes and deep cross-cutting characteristics pose challenges in surface surveys. Hence, this study extracted the granite pegmatite information within the study area from remote sensing images using the random forest algorithm. With high-spatial-resolution GF-2 remote sensing images as the primary data source, it extracted the spectral, texture, exponential, topographic, and edge features from various ground objects within the study area. These features, together with the newly introduced contrast limited adaptive histogram equalization (CLAHE) features, constituted 25 feature variables, forming a feature subset. Then, feature variables in the subset were evaluated for their feature importance, and their importance scores were used for feature selection, determining the optimal feature combination for extracting granite pegmatite information. Ultimately, 16 feature variables were chosen for random forest classification, with the accuracy of the classification results assessed. The study indicates that: ①The CLAHE feature variables emphasize the tonal variations among ground objects, thereby enhancing the classification accuracy, with the overall accuracy increased by 2.7 percentage points and the Kappa coefficient increased by 0.035; ②The classification results for granite pegmatites based on GF-2 images and the random forest algorithm exhibited overall accuracy of 93.1%, with a Kappa coefficient of 0.902, user accuracy of 94.24%, and producer accuracy of 98.00%, confirming the effectiveness of the method used in this study. Moreover, this study provides reliable data for future research in the study area.

However, since GF-5 launch in 2018, few studies regarding the application of GF-5 AHSI data for uranium deposit exploration have been reported. In this study, with the Weijing area of Inner Mongolia as the study area, the spectral hourglass technology was applied to extract alteration anomalies of goethite and low-, medium-, and high-aluminum sericite from corresponding GF-5 AHSI data. Then, the principal component analysis (PCA) and the LINE module in PCI Geomatica software were employed for the automatic extraction of linear structures in the study area, with a linear structure density map created. Finally, a uranium mineralization potential map of the study area was generated by integrating all proof layers based on the ArcGIS software. The results indicate that the extraction of alteration information and linear structures, and the integration of multiple proof layers are feasible, and the obtained uranium mineralization potential map exhibits high reliability. One uranium deposit prediction zone was identified based on the study results and geological data. The study results will guide the subsequent uranium deposit exploration in the study area while providing a reference for the geological application of GF-5 AHSI data.

Against the backdrop of global warming, increasingly frequent floods become a primary agricultural disaster that causes reduced crop production in China. Radar remote sensing technology, possessing all-weather earth observation capabilities, serves as a critical means for rapid monitoring of regional flood information. With the advancement in artificial intelligence, machine learning methods have been extensively applied in the remote sensing-based monitoring of floods. Despite the high accuracy of their algorithms, their training processes often entail extensive field investigations or numerous samples for remote sensing image interpretation. This study aims to overcome sample labeling limitations and improve regional flood monitoring accuracy. Based on the catastrophic flood that occurred in northern Henan on July 20, 2021, this study constructed a flooded crop monitoring method based on the weakly supervised Gaussian mixture model (GMM) using domestic GF-3 HH-HV radar images. Then this method was applied to extract the flooding range of farmland in some areas of northern Henan. Compared to four typical machine learning methods, i.e., random forest (RF), support vector machine (SVM), K-nearest neighbor classification, and parallelepiped classification, the weakly supervised GMM in this study enjoyed the highest accuracy, with overall precision of 0.95 and a Kappa coefficient of 0.90. This study holds great significance for enhancing the accuracy and universality of regional crop flooding monitoring based on remote sensing technology and synthetic aperture radars (SARs).

Assessing the landslide susceptibility of highways in precarious mountainous areas can provide crucial information for the geologic route selection of highways. Conventional landslide susceptibility assessment methods ignore the application of surface deformation data and other dynamic data, leading to low-accuracy assessment results. Hence, this study proposed a landslide susceptibility assessment method for mountain highways based on optical and SAR remote sensing images. With the Longwuxia-Gongboxia section of the Yanhuang Highway in Qinghai Province as the study area, this study extracted various static factors of landslides from high-resolution QuickBird satellite images and calculated the initial risk level of landslide susceptibility within the route area using a random forest model. Afterward, this study obtained the surface deformation factors, which directly reflect the dynamic changes of landslides, based on the long-time-series Sentinel-1A images. Finally, this study corrected the initial landslide susceptibility risk level based on the surface deformation factors, generating a landslide susceptibility assessment zoning map. As demonstrated by engineering practice, the method proposed in this study yielded a high-accuracy landslide susceptibility assessment zoning map for the mountain highway by combining data on both static and dynamic factors of landslides, thus providing accurate information for subsequent geologic route selection of the highway.

In response to the detection and positioning demands for land for photovoltaic development due to the rapid growth of the photovoltaic industry, this study proposed a YOLOv5-pv algorithm for the detection of land for photovoltaic development based on the improved YOLOv5. For quick and accurate detection and positioning of land for photovoltaic development in complex scenes, the YOLOv5-pv algorithm adopted a weighted bi-directional feature pyramid based on YOLOv5 to achieve simple and fast multi-scale feature fusion, thereby enhancing the ability to detect small targets. Subsequently, the Ghost convolution was employed to retain valuable feature map information in redundant information. Finally, a co-attention mechanism was integrated to improve the algorithm's attention on the land for photovoltaic development, increasing its capacity to resist background interference. The experimental results demonstrate that YOLOv5-pv outperformed YOLOv5, with the recall rate and average accuracy improved by 6.68 percentage points and 4.43 percentage points, respectively. Therefore, the method proposed in this study can effectively detect the land for photovoltaic development, holding referential significance for relevant detection research.

Currently, most change detection algorithms for remote sensing images fail to effectively process images polluted by Gaussian, impulse, or mixed noise. To address this problem, this study presented five fuzzy C-means (FCM) clustering algorithms (FCM_S1, FCM_S2, KFCM_S1, KFCM_S2, and FLICM) based on neighborhood space information. These algorithms, which can efficiently decompose mixed pixels in the presence of noise pollution, were combined with a simple Bayesian network (SBN). Under the framework of change vector analysis in posterior probability space (CVAPS), this study developed five change detection methods for remote sensing images, exhibiting high resistance to Gaussian, impulse, and mixed noise. Comparative experiments demonstrate that the change detection algorithms proposed in this study manifest high robustness against the above-mentioned noise.

A timely and accurate assessment of the spatial precipitation distribution holds great significance to the development of the national economy. At present, most remote sensing-based precipitation products improve their accuracy using multiple regression models and physical models rather than deep learning models. This study improved a long short-term memory neural network (LSTM) deep learning model, yielding an optimized LSTM deep learning model. With the Fujian-Zhejiang-Jiangxi area as the study area, this study conducted downscaling for an integrated multi-satellite retrievals for global precipitation measurement (IMERG) product based on the daily precipitation data of 69 meteorological stations from 2015 to 2019 by introducing multiple factors controlling precipitation such as vegetation, slope aspect, slope gradient. Finally, this study assessed the reliability of the optimized model through verifications based on high-density meteorological stations and individual years. The results show that the downscaling results are consistent with the spatio-temporal distribution of precipitation measured at meteorological stations and, thus, can better reflect the spatial distribution of precipitation in the study area than the original IMERG. Furthermore, underestimated and overestimated precipitation data of the study area from the GPM product were corrected. As indicated by the verification based on high-density meteorological stations, the downscaled model yielded correlation coefficients of 0.9 or above for July and October, which were followed by April. The correlation coefficient was the lowest of 0.7 in January. As shown by the verification based on individual year data, the correlation coefficient between the daily precipitation downscaling results and the measurement results in 2020 was above 0.8, with a root mean square error of 5.23 mm and an average relative error of 9.43%. Therefore, the deep learning-based downscaling model enjoys high accuracy on both daily and monthly scales and can be widely applied in the assessment of both spatial and temporal precipitation distributions.

Superpixel segmentation-based image processing has been extensively used for the classification of hyperspectral images (HSI) in recent years. However, it fails to fully extract the HSI information at a single scale, and its classification process highly depends on parameters. Given the insufficient spatial information utilization by the superpixel segmentation-based HSI classification technology, this study proposed an HSI classification method that combines the superpixel segmentation method and the extended multi-attribute profile (EMAP) method. First, the superpixel segmentation and EMAP methods were employed to extract superpixel-level and pixel-level HSI features, respectively. By fusing the two types of features, the resulting images displayed complete HSI structural characteristics. To eliminate information redundancy, the fused images were subjected to spectral filtering through the recursive filtering method. Finally, the features were input to the support vector machine (SVM) for pixel tag determination. Experiments on the Indian Pines and University of Pavia datasets analyzed the effects of parameter variations on classification accuracy. Compared with the S3-PCA algorithm, the method proposed in this study exhibited superior classification accuracy and Kappa coefficient, which were improved by 3.55 and 2.88 percentage points, respectively.

This study aims to investigate the fundamental facts concerning slope geological hazards in Longgang District, Shenzhen City, as well as the distributions of disaster-prone zones in the district. Based on the multi-source remote sensing satellite data, this study interpreted the slope geological hazards using the expert interpretation method on a geological hazard interpretation platform. Furthermore, some interpreted geological hazards were verified through field verification combined with Baidu Street View data. Finally, the distributions of zones susceptible to slope geological hazards in Longgang District were determined using the information value method, with the slope height, slope gradient, rainfall, surface lithology, and land cover as assessment factors. Additionally, existing geological hazard sites were superimposed with the susceptibility assessment results for analysis, yielding completely consistent results. This confirms the effectiveness of the method used in this study for assessing the susceptibility of slope geological hazards, as well as the accuracy of remote sensing interpretation of slope geological hazards.

Delving into the urban heat island effects caused by urban expansion holds crucial significance for addressing urban thermal environment challenges. Based on the Landsat remote sensing images of Nanjing in 2000, 2010, and 2020, this study obtained Nanjing’s surface temperatures through inversion using the radiative transfer equation and extracted the impervious surface information using the biophysical composition index (BCI). It analyzed the urban expansion directions and types of Nanjing from 2000 to 2020 by employing the standard deviation ellipse and the landscape expansion index. Moreover, it investigated the effects of urban expansion types on the thermal environment through statistical analysis. The results are as follows: ① From 2000 to 2020, Nanjing experienced an increase in surface temperatures from 29 ℃ to 30 ℃ and an expansion of the heat island area from 2 248 km² to 3 051 km². The urban heat island expanded towards the south between 2000 and 2010 and spread to the surrounding areas between 2010 and 2020; ② The urban land of Nanjing expanded outwards from its center, mainly towards the south. The expansion types were dominated by edge expansion, succeeded by infilling and exclave expansions. The proportion of edge expansion between 2000 and 2010 was slightly higher than that between 2010 and 2020; ③ The urban expansion exhibited the same direction as the urban heat island expansion, and edge expansion resulted in the most intense urban heat island effects, followed by exclave and infilling expansions. This study can provide a scientific basis for ameliorating Nanjing’s thermal environment based on the urban expansion types and directions.

The Yellow River basin has a weak eco-environment due to severe water-soil erosion. Research on ecological security state and primary driving factors holds great significance for the high-quality development of the basin. Based on the social statistical data and remote sensing data of counties/banners in Ordos in the basin from 2010 to 2020, this study established an ecological security index framework for these areas using the pressure-state-response (PSR) model by integrating 19 indices. Furthermore, this study calculated key driving factors using the gray relational model and forecast the ecological security evolution trend in the coming five years. The results are as follows: ① From 2010 to 2020, the comprehensive ecological security index of Dalad, Jungar, and Otog banners all showed a rising trend, while that of Hanggin Banner showed a downward trend following a rise. The ecological security level fluctuated around the critical state during this period. This result suggests an improvement in the ecological security of the basin; ② The dynamic change of ecological security in the study area was closely related to the changes in regional population, urbanization rate, grassland area, water resources area, Engel coefficient and proportion of tertiary industries; (iii) As indicated by the results of the ecological security forecast model of various counties/banners, the ecological security index of this area will show an upward trend in the coming five years. Finally, based on the above results, this study put forward relevant policies and suggestions, with a view to providing a quantitative reference and decision-making basis for ecological management and risk prevention and control of the Yellow River basin in Ordos.

Nature Reserves are effective means to protect biodiversity and improve the ecological environment. However, frequent human activities have threatened the ecosystem quality and stability of them. In order to study the use of remote sensing means to monitor and evaluate of human activities influence in the protected area, Dongzhaigang, Tongguling, Sanya Coral Reef and Dazhou Island National Marine Nature Reserves in Hainan Provinces were taken as research areas, and the high spatial resolution remote sensing images from 2016 to 2020 were adopted, to obtain the transformation information of artificial and natural factors through image reprocessing, classification system and interpretation signs establishment, human-computer interpretation and other steps. By collecting the topographic and meteorological data, considering the characteristics of human activity and ecological sensitivity of the local area, 11 evaluation factors including topographic, meteorological and land use types were selected to establish the evaluation index system. Analytic Hierarchy Process method was used to evaluate and grade the degree of human activity influence, and the distribution results of severely affected areas, moderately affected areas, mildly affected areas and non-affected areas were obtained. The results were analyzed and the conclusions can be drawn. The results show that from 2016 to 2020, in Dongzhaigang Nature Reserve, human disturbance is strong, but has a tendency to reduce. Severely and moderately affected areas are distributed in Beigang Island, north of the reserve and mildly affected areas are distributed in the edge of the protected area. Destructions in these areas are mainly caused by construction activities of the village. As to Tongguling Nature Reserve, even though it also has human disturbance, but the overall protection is good. Severely and moderately affected areas are mainly located in the rock park in the north and the Tongguling scenic spot in the east. It is mainly caused by the construction of tourism and transportation facilities. Mildly affected areas are located along the coast of Qishui Bay in the west, mainly caused by real estate development. Human activity disturbance in Sanya Coral Reef Nature Reserve is serious, but has a tendency to decrease. Severely affected areas are mainly located in Luhuitou Peninsula and Yulinjiao area, where vigorous development of tourism real estate destroy forests. Moderately affected areas are located in Ximao Island, mainly caused by the construction of residential areas in the north. No land cover changes caused by human activities are found in the Dazhou Island Nature Reserve. The results can provide a scientific basis for the management and protection of national Marine nature reserves in Hainan Province.

To explore the ecosystem health of Anhui Province in the context of modern urbanization, this study established an ecosystem health evaluation system for Anhui Province based on time-series land use type data using the vigour-organization-resilience (VOR) model. Then, this study defined the weights of indicators in the evaluation system using the entropy weight method and evaluated the ecosystem health of Anhui Province and its spatio-temporal evolution over the period from 1980 to 2015 on the county scale. The results show that: ① Anhui Province was dominated by counties with an unhealthy ecosystem, followed by counties with a subhealthy ecosystem, both of which accounted for 56.37% and 16.83%, respectively. Spatially, the comprehensive average multiyear health index and average multiyear health grade of Anhui Province’s ecosystem were higher in the south and lower in the north, strongly correlating with the landform type. Areas with Healthy and very healthy ecosystems were mostly distributed in the hilly and mountainous areas in the southern and western parts of the province. In contrast, areas with an unhealthy ecosystem were contiguously distributed in the Huai River Plain, the Jianghuai hilly area, and the plain area along the Yangtze River. Among, Qimen County, Huangshan City exhibited the highest comprehensive ecosystem health index of 0.92, and Luyang District of Hefei City displayed the lowest comprehensive ecosystem health index of about 0.17; ② The overall ecosystem of Anhui Province showed a healthy trend from 1980 to 2015. With 24 counties exhibiting increasing health grades and no counties displaying decreasing health grades, the average provincial health grade increased from unhealthy to subhealthy grades. Nevertheless, 13 counties exhibited decreasing comprehensive health index values, indicating unstable ecosystem health levels and risks of decreasing health grades. The findings of the study can, to some extent, provide a reference for Anhui Province to formulate reasonable land use policies, protect and manage the eco-environment, and optimize ecosystem service functions.

The net primary productivity (NPP) of vegetation, exhibiting regional differentiation, serves as a crucial parameter for determining the carbon source/sink of ecosystems. Based on the MOD17A3HGF, topography, and human activity data, this study delved into the spatio-temporal differentiation of vegetation NPP in Henan Province from 2010 to 2020 and its response to driving factors using methods like the gravity center model, trend analysis, and the geographical detector model. Moreover, it revealed the explanatory power and interactions of the driving factors. The results are as follows: ① Temporally, the vegetation NPP from 2010 to 2020 displayed a slightly fluctuating upward trend, averaging 424.89 gC·m^-2·a^-1. Its gravity center exhibited significant temporal differentiation, with the average center of gravity closer to the geometric center. ② Spatially, the vegetation NPP values increased from the northeast to the southwest and were dominated by median values (300~600 gC·m^-2·a^-1). ③ In terms of influencing factors, the vegetation NPP showed a higher correlation with precipitation compared to temperature. Moreover, it first increased and then decreased with an increase in altitude and slope. The areas with altitudes below 200 m and slopes less than 2° contributed the most to NPP in the study area. The vegetation NPP on sunny slopes was higher than that on shady slopes. In the case of land use changes, the shift to arable land plays a significant role in the increase of total NPP. ④ The geographical detection results indicate that precipitation exhibited the highest explanatory power for changes in vegetation NPP. The two-factor interactions all showed an enhanced relationship, with the q value of precipitation ∩ longitude being the highest. These findings provide data support for ecological protection and high-quality development of Henan Province.

This study aims to compare the accuracy of random forest(RF) and ordinary kriging(OK) model for predicting spatial distribution of soil pH in arable land of Jiangxi Province using different covariates combination, and assess the feasibility and potential of RF method for improving the prediction accuracy of soil pH value. The RF algorithm is used to predict the pH value of cultivated soil in Jiangxi Province based on environmental covariate such as climate, topography and vegetation, combined with soil properties and cultivated land use conditions, identify the main influencing factors. The results produced by the RF was compared with the classical OK interpolation model. Our results showed that the accuracy of RF-A model with soil properties and cultivated land use conditions as environmental variables is better than that of RF-B model which only including terrain, climate and vegetation attributes as environmental variables. Climatic condition is the dominate factor which control the spatial variation of soil pH. the topographic factors and anthropogenic factors also have essential effect on spatial variability of soil pH. Thus, this study proved RF method has theoretical and practical significance for improving the accuracy of soil pH prediction at large-scale.

Fugong County of Yunnan Province, located near the Yunnan-Myanmar border, is a typical alpine gorge area with a fragile geological environment. Geological hazards, including landslide, collapse, and debris flow, occur frequently in this area, thus posing a severe threat to the safety of people's lives and properties, economic development, and even national defense security. However, high mountains and thick forests in this area complicate manual investigations and increase the risk. Hence, this study conducted surface deformation monitoring and geological hazard detection in this alpine gorge area by employing the small baseline subset - interferometric synthetic aperture radar (SBAS-InSAR) technology based on the Sentinel-1A satellite data. The identification results were verified by combining ground surveys and the interpretation of optical remote sensing images. The findings indicate that most of the deformation zones in this area exhibit fragile geological conditions and are prone to landslide instability in the case of heavy rainfall, thus requiring continuous observation. This study offers a valuable reference for investigating and monitoring landslides in alpine gorge areas.

It has been proposed to develop natural resource survey and monitoring systems. Under this background, the integration of the new generation of digital technologies and the survey and monitoring is suffering from the following problems in the unified natural resource survey and monitoring: the lack of collaboration among spatio-temporal data on natural resources, the inadequate automation of data processing, the limited level of refined management of achievements, and inadequate social services and data security. In the context of the rapid development of modern survey technologies and new-generation digital technologies such as artificial intelligence, big data, and cloud computing, as well as their in-depth combination with natural resources survey and monitoring, this study proposed a technical framework of the unified natural resource survey and monitoring system. Accordingly, this study conducted research on five aspects: air-space-land-people-network collaborative perception, the automatic processing of spatio-temporal data on natural resources, the construction of spatio-temporal database of natural resource entities, the sharing service of survey and monitoring achievements, and data security. Finally, with the air-space-land-people-network collaborative perception network, the R&D of a universal survey and monitoring data acquisition system, and the comprehensive monitoring of natural resources in Guangxi Province as examples, this study introduced the achievements made in the construction of the technical framework of the natural resources survey and monitoring system.

Land use/land cover (LULC) simulation is essential for research on changes in land use. Based on the Google Earth Engine (GEE) platform, this study extracted the high-precision LULC information of Luquan County from 1991 to 2021 and analyzed the spatio-temporal evolution pattern. Then, this study analyzed the factors driving LULC changes using a random forest model and compared the simulation results of Luquan County obtained using the cellular automata-Markov (CA-Markov), land change modeler (LCM), future land use simulation (FLUS), and patch-generating land use simulation (PLUS). Finally, this study forecast the LULC change scenario in Luquan County in 2027 using the optimal model. The results show that: ① From 1991 to 2021, the spatial LULC pattern of Luquan County was dominated by forestland, grassland, and farmland. The areas of farmland and waterbodies increased by 89.26 km² and 27.72 km², respectively, the areas of forestland, construction land, and bare land increased continuously by 724.25 km², 21.08 km², and 13.67 km², respectively, and the grassland decreased at an annual average rate of 29.20 km²; ② The LULC in Luquan County was primarily influenced by topographic conditions (elevation and slope); ③ The simulation effects of the four LULC models were in the order of PLUS > FLUS > CA-Markov > LCM, with Kappa coefficient of 0.63, 0.58, 0.46 and 0.35, respectively and the overall accuracy of 0.78, 0.75, 0.66 and 0.58, respectively; ④ The spatial LULC pattern in Luquan County in 2027 will share similarities with that in 2021. From 2021 to 2027, the areas of farmland land, grassland, and water bodies will decrease at a rate of 40.21 km²/a, 4.51 km²/a, and 0.70 km²/a, respectively, while the forestland, construction land, and bare land will expand by 265.52 km², 4.85 km², and 2.08 km², respectively.

This study aims to analyze the changes in the land use and vegetation cover in the Xiong'an New Area from 1991 to 2021. To this end, this study explored the characteristics of the land use changes in the area over the 30 years based on the Landsat TM\OLI data of five periods using the GIS technology and map fusion method. Then, it extracted the vegetation cover information using the dimidiate pixel model and analyzed the changes in the vegetation cover. Furthermore, this study explored the potential factors driving the vegetation cover change in the area using the geographic detector model and analyzed the impact of land use change on vegetation cover change by referencing the existing map fusion method. The results show that: ① From 1991 to 2021, the construction land in Xiong’an New Area increased by 108.09 km², primarily transformed from farmland and other types of land; other types of land reduced by 108.17 km², predominantly transformed to farmland; forestland and grassland increased by 11.56 km², mainly transformed from water areas and other types of land; the water area decreased by 38.76 km², mainly transformed to farmland and other types of land; and the area of farmland roughly remained unchanged; ② Over the 30 years, the Xiong’an New Area generally exhibited high vegetation coverage, and the area with moderate and high vegetation coverage and above accounted for more than 50.00%. The vegetation coverage in the Xiong’an New Area presented an overall spatial distribution pattern characterized by high in Anxin County, moderate in Rongcheng County, and low in Xiong County. Regarding the phased changes, this area showed a degradation trend from 1991 to 2001, and the area with degraded vegetation cover accounted for 39.15%. From 2001 to 2021, this area exhibited an improvement trend, the area with improved vegetation cover accounted for up to 47.55%; ③ The vegetation cover change showed spatial differentiation, significantly affected by the population density, GDP, soil type, and soil quality but slightly affected by the elevation and slope. The transformation of construction land and other types of land to farmland acted as an important reason for the improvement in vegetation cover, while the transformation of farmland to construction land and other types of land served as an important reason for vegetation degradation. The results of this study can, to some extent, provide a scientific basis and suggestions for the sustainable development of Xiong’an New Area.

Soil salinization is an important reason for land degradation and desertification and has a huge impact on the eco-environment. Coastal wetlands are typical areas subjected to a weak eco-environment and severe salinization, and there is an urgent need to investigate the impact of soil salinization on their eco-environment. This study proposed the baseline-based soil salinity index (BSSI), which can effectively suppress the influence of complex features on surface salinization monitoring and improve the accuracy of saline soil extraction by 10% compared to other salinity index models. Furthermore, this study proposed the optimized water benefit-based ecological index (OWBEI) by optimizing the water benefit-based ecological index (WBEI), which can effectively increase the accuracy of eco-environment quality assessment to 87%. Finally, this study explored the mechanical processes of the influence of soil salinization on the eco-environment quality based on the distribution of soil salinization and eco-environment quality obtained from the Yellow River Delta. The results show that the deterioration of soil salinization has led to an increase in the soil vulnerability of coastal wetlands, indirectly resulting in a continuous decrease in eco-environment quality. Although eco-environment protection measures have been continuously proposed, few of them are tailored to the solving of salinization. This leads to the deterioration of the ecological quality, which then yields negative feedback to the soil and eventually forms a vicious circle. This adversely affects local production, life, and social development.

Ecological restoration zoning is a prerequisite for effective ecological restoration, and currently, the most commonly used zoning method is based on ecosystem services (ES). Most of the previous studies merely focus on the current ES status but ignore its changes, thus failing to reflect the potential and direction of ecological restoration. This study proposed a two-level zoning method for ecological restoration based on ES changes and applied this method to the Hexi region, Gansu Province. First, through Landsat image classification based on the Google Earth Engine (GEE) platform, this study obtained two periods (2005 and 2020) of land use data and calculated the ES sub-value and total value of the two periods of data using the equivalent factor method. Then, this study constructed the level-1 ecological restoration areas (priority restoration area PrR, important restoration area ImR, potential restoration area PoR, important protection area ImP, and priority protection area PrP) through the clustering and outlier analyses of the total ES value changes. Finally, the level-2 ecological restoration zones were determined based on the combined characteristics of changes in the values ES subitems. The results show that: ① Various level-1 areas, i.e., the PrR, ImR, PoR, ImP, and PrP areas account for 0.9%, 7.2%, 78.0%, 13.0%, and 0.9%, respectively. Most of the PoR areas are distributed in the Gobi desert of Hexi region, the PrR and PrP areas are sporadically distributed in the transition zone from the Qilian Mountains to the piedmont grassland, the ImR and ImP zones are mostly distributed in the Qilian Mountains, corridor plains, and mountains in the north. There exit greater potential for the restoration of the ImR areas and the protection of the ImP areas but limited potential for ecological restoration. Furthermore, there is a more urgent need for protection than for restoration; ② The level-2 areas can be classified into 10 categories of restoration areas and six categories of protection areas. The level-2 areas of the ImR and ImP areas are primarily determined based on the synergistic changes in sub-services. Both ecological restoration and protection measures for the Hexi region should focus on the comprehensive enhancement of ES.

Accurately identifying the evolutionary trend and influencing factors of ecological environment quality in new urban agglomerations holds crucial significance for scientifically guiding urbanization and achieving sustainable development. Existing research on the spatio-temporal evolutionary characteristics of ecological environment quality in new urban agglomerations ignored the interactions of multiple factors on ecological environment quality. Based on the Google Earth Engine (GEE) platform, and long-time-series Landsat TM/OLI remote sensing images as the fundamental data source, this study delved into the spatio-temporal variations of ecological environment quality in the Changsha-Zhuzhou-Xiangtan urban agglomeration from 1990 to 2020 using methods including the remote sensing ecological index (RSEI), Sen’s slope estimator, and Mann-Kendall test. Moreover, the geographical detector was employed to quantitatively measure the effects of various factors on the urban agglomeration’s spatial heterogeneity. The results indicate that the Changsha-Zhuzhou-Xiangtan urban agglomeration exhibited generally high ecological environment quality, with a spatial distribution pattern of higher quality in marginal areas and lower quality in core areas. The average proportion of areas with ecological environment quality graded as “excellent” and “good” exceeds 60% in the urban agglomeration. The sustainable development strategy altered the urban sprawl in this urban agglomeration, leading to a decline followed by an increase in RSEI, with an inflection point in 2000. From 1990 to 2020, the ecological environment quality significantly deteriorated in central urban areas while improvement was observed in non-central urban areas. Physical and geographical conditions significantly influenced the ecological environment quality of the urban agglomeration in the early stages. With socio-economic progression, the influence of socio-economic factors like nighttime lighting on ecological environment quality gradually intensified, assuming a dominant role over time. Besides, the interactions among factors significantly enhanced the effects of individual factors on ecological environment quality. Before 2010, the interactions between human and natural factors exerted considerable influences on the ecological environment. After 2015, the interactions among human factors yielded more pronounced effects on ecological environment quality. These findings serve as a foundational guide for the integrated high-quality development of the Changsha-Zhuzhou-Xiangtan urban agglomeration and a reference for the advancement of other comparable urban agglomerations.

The coordinated economic and ecological development and the cultural landscape preservation of the Great Wall cultural zone are crucial for regional social sustainability. To meet the need for integrated monitoring and evaluation of large-scale linear cultural heritage, this study proposed a remote sensing dynamic monitoring method that integrates object-oriented change vector analysis and U-net deep learning. Based on the suppression of classified scattered noise and the accurate dynamic description of key regional environmental components, this study achieved the interpretation and information mining of the factors driving cultural landscape changes by combining socio-economic data and remote sensing change detection. Building on the 2-m-resolution GF-2 fused images from 2015 to 2020, the Beijing section of Ming Great Wall was examined through remote sensing change detection of surface elements and quantitative analysis of the land cover change matrix for its landscape corridor using methods including multiresolution segmentation, change vector analysis and extraction, and U-net image classification. The study reveals that the land cover along the Beijing section of Ming Great Wall cultural zone yielded a change rate of 0.098%, primarily manifested in the shift from bare land and farmland to forests and the growth of artificial land. Meanwhile, the ecological environment of the cultural zone exhibited positive development and an overall favorable protection state. The research results will provide technical support for the coordinated economic and ecological development and the sustainable preservation of the cultural landscape along the Beijing section of Ming Great Wall.

Power grid projects in mountainous regions have encountered numerous landslides in recent years, leading to collapsed transmission towers and power outages. Hence, early identification of potential landslides is crucial for ensuring the safety of power engineering. For this purpose, this study conducted early identification of potential landslides along the Sichuan-Chongqing power grid based on optical remote sensing and the small baseline subset (SBAS) - interferometric synthetic aperture radar (InSAR) technology. The interpretation of high-resolution optical remote sensing images revealed 28 potential landslide sites near the transmission towers along the power grid. Based on this, this study detected the study area’s surface deformation using the SBAS-InSAR technology, identifying 27 potential landslide sites. Except for 15 repeated results, the above two methods identified a total of 40 potential landslide sites. Finally, through field check and the qualitative analysis of deformation signs and stability, this study determined that seven potential landslide sites threaten the safety of transmission towers, with two of them presenting higher risks. These findings provide valuable guidance and references for the prevention and control of landslides along the Sichuan-Chongqing power grid.

The simulation-based estimation and spatio-temporal variations of the net primary productivity (NPP) of regional vegetation hold critical significance for analyzing regional vegetation quality and carbon balance. This study investigated the Jianghe River basin in western Hubei Province. First, it pre-processed the input data, including land cover, topography, soil, meteorology, and vegetation indices. Based on this, it estimated the NPP of vegetation in the Jianghe River basin from 1986 to 2017 using the BEPS-TerrainlabV2.0 model, with the model's simulation accuracy evaluated. Moreover, this study explored the spatio-temporal variations of the NPP and its response to environmental changes. The results are as follows: ① The NPP of vegetation in the Jianghe River basin exhibited a unimodal distribution, with higher values in summer and lower values in winter, on an intra-annual scale, and a fluctuating rising trend on an inter-annual scale; ② The spatial distribution of the NPP manifested higher values in the north and lower values in the south; ③ The NPP values of different land cover types followed the sequence below: broad-leaved forests > mixed forests > coniferous forests > farmland > urban areas. The NPP rose with an increase in elevation. The NPP values of different soil textures rank below: sandy soil > sandy loam > loamy sand > silty loam; ④ Radiation and temperature manifested the strongest impact on NPP on a daily basis, and the leaf area index (LAI) exhibited the most significant influence on NPP on an annual basis, both passing the 0.01 significance test.

Identifying land use conflicts holds critical significance for sustainable socio-economic development and ecological civilization construction. Since Urumqi City is situated in the core region of the Silk Road Economic Belt, investigating the causes and manifestations of its land use conflicts arising from urban development, oasis agriculture, and ecological environment becomes an urgent and necessary task. With Urumqi as the study area, this study analyzed its land use characteristics in 2000, 2010, and 2020, as well as those in 2030 simulated from the FLUS model. Based on this analysis and the pressure-state-response (PSR) model, a land use conflict intensity measurement model was constructed to evaluate the land use conflicts over the four periods. Finally, a geographic detector was employed to quantitatively analyze the factors driving land use conflicts in the study area. The findings indicate that: ① The land use between 2000 and 2030 exhibited significant spatial differentiation, showing increased construction land, forest land, and water areas, but decreased grassland, arable land, and unused land; ② The comprehensive indices of land use indicate low to medium utilization degrees but an overall rising trend, suggesting land use in a development stage; ③ Significant spatial changes occurred in land use conflicts between 2000 and 2030. The conflict-free and mild conflict zones occupied the largest proportions, the moderate conflict zones showed normal distributions, and severe and high-level conflict zones increased annually, with the highest increase observed in high-level conflict zones; ④ From 2000 to 2010, the hotspots of land use conflicts were distributed primarily in the north and southwest of the central urban area. From 2010 to 2020, they spread to the periphery of forest land in the southern and northern mountainous areas, and the areas near the alluvial fans on both sides of the salt lake in the Dabancheng District. From 2020 to 2030, the hotspots are still mainly located around the land for construction and near the forest land in mountainous areas but significantly decreased in the mountainous areas; ⑤ As demonstrated by one-way influence analysis of spatial differentiation drivers on land use conflicts, the influences of factors are in the order of patch density > population density > GDP > slope > elevation > distance from districts and counties > distance from rivers > distance from roads. Additionally, the interaction detection analysis indicates (patch density ∩ elevation) > (patch density ∩ average land population)>(patch density ∩ distance from roads). This study serves as a reference for effectively managing the conflicting demands between economic development and ecological conservation in Urumqi and enhancing the future land use composition.

Focusing on Quanjiao County in Chuzhou City, this study determined 90 features, including spectral, traditional vegetation index, red-edge vegetation index, and texture features, from Sentinel-2 satellite data on the GEE platform. This study examined the effects of diverse feature optimization algorithms combined with a random forest classifier on identifying crop planting types in the study area. These algorithms included the random forest-recursive feature elimination (RF_RFE) algorithm, the Relief F algorithm based on Relief expansion, and the correlation-based feature selection (CFS) algorithm. On this basis, this study further analyzed the classification effects of the optimal feature optimization algorithm in various machine learning classification approaches. The study demonstrates that: ① Spectral features proved to be the most crucial for crop identification, followed by red-edge index features, and texture features manifested minimal effects; ② RF_RFE-based remote sensing identification results exhibited the highest accuracy, with overall accuracy of 92% and a Kappa coefficient of 0.89; ③ Under the RF_RFE feature optimization method, the RF’s Kappa coefficient was 0.01 and 0.41 higher than that of the support vector machine (SVM) and the minimum distance classification (MDC), respectively. This indicates that the RF_RFE feature optimization method based on multiple features, combined with the RF algorithm, can effectively enhance the accuracy and efficiency of remote sensing identification of crops.

The rapid growth of China’s photovoltaic (PV) industry is accompanied by unplanned construction of PV power plants. Ascertaining the regional PV development suitability, power generation potential, and emission reduction effects holds critical significance for the sound development of the PV industry. Based on remote sensing, meteorological, and fundamental geographic data, this study constructed an evaluation index system for PV development suitability. Using this system, it assessed the zones suitable for PV development in the Western Sichuan Plateau and estimated the PV power generation potential and emission reduction effects. The results are as follows: ① The zones suitable for PV development account for 57.43% of the entire plateau, with highly suitable zones covering an area of approximately 2.07×10⁴ km², which are distributed primarily in the southwestern and northwestern portions of the plateau; ② The plateau exhibits significant power generation potential, reaching 17 197.97×10⁸ KWh in highly suitable zones under a full development scenario, which is equivalent to 6.52-fold Sichuan Province’s total electricity consumption in 2019 before the COVID-19 outbreak; ③ Contrasting with conventional thermal power generation, PV power generation in highly suitable zones can achieve annual CO₂ emission reduction of 12.45×10⁸ t, which is about 12.71% of China’s total CO₂ emissions in 2019 and 3.95-fold Sichuan Province’s CO₂ emissions. Moreover, PV power generation can diminish the emissions of coal and conventional pollutants as well as heavy metals. The findings offer a scientific reference and guidance for selecting sites for PV power plants in the Western Sichuan Plateau and promoting the sustainable growth of the PV industry.