Change detection refers to the technology of extracting land cover changes by comparing and analyzing multi-temporal remote sensing images acquired at different periods covering the same area. With the advancements in satellite and sensor technologies， there has been a significant increase in Earth observation data. Change detection plays an important role in various fields， including geoinformation survey and ecological environment protection. In recent years， deep learning technology has become an advanced method for change detection due to its powerful feature mining ability. This paper provides a comprehensive overview of deep learning-based change detection methods from three aspects： pixel-level， object-level， and scene-level. Furthermore， this paper discusses the practical implementation of deep learning in change detection through three research examples. Finally， the study concludes by outlining future development trends in deep learning-based change detection.

Pine wilt disease is a devastating pine disease， which seriously threatens forest ecological security. Timely and reliable acquisition of the extent and severity of pine wilt disease is very important for forest management and disease prevention and control. However， pine wilt disease spreads rapidly and is difficult to control， and the traditional manual survey methods can hardly meet the demand. Unmanned Aerial Vehicle （UAV） remote sensing can quickly and accurately obtain the extent and severity of forest diseases， and provide reliable information support for forest pest control and management. In this study， UAV was used to acquire high-resolution Red-Green-Blue （RGB） visible light images. Firstly， object-oriented multi-scale segmentation algorithm was used to extract the crown of a single tree， and Vegetation Index （VIs） and texture （GLCM） features were calculated. Then the feature selection algorithm was used to optimize the feature set， and Random Forest （RF） classification and Support Vector Machine （SVM） classification algorithm were used to construct the pine wilt disease classification model based on different feature sets. Through the ablation experiment， the optimal classification model was selected and the object-oriented method was used to monitor the disease degree and spatial distribution of pine wilt disease. The results show that the vegetation index and texture features of pine canopy with different disease degrees are different on the object-oriented single tree crown scale， and the accuracy of classification results using vegetation index was better than texture characteristics （VIs RF：OA=76.52%，Marco-F1=0.77；SVM：OA=79.68%，Marco-F1=0.79）. Compared with a single feature set， the combination of vegetation index and texture features can significantly improve the classification accuracy （VIs&GLCM RF：OA=79.47%，Marco-F1=0.80；SVM：OA=85.45%，Marco-F1=0.85）， indicating that multi-feature combination can effectively improve the pine wilt disease classification. The SVM model outperforms the RF model for classification， both for single feature set modeling and combined feature set modeling. This study provides timely and reliable information to support a comprehensive grasp of the extent and severity of pine wilt disease， and helps to promote the construction of a major forestry pest control system and maintain ecological security.

The super-resolution reconstruction of remote sensing images is a type of method that uses image analysis methods to reconstruct high-resolution images from one or more low resolution images， in order to restore high-frequency details lost during sensor imaging， storage， and transmission， and improve the quality of remote sensing image data. The core lies in constructing a mapping relationship between high- and low- resolution images. This paper reviews the mainstream methods and representative research works in image super-resolution reconstruction and focuses on analyzing the recent advances in traditional methods and deep learning methods in the field of remote sensing image super-resolution reconstruction. The results indicates that： （1） Methods based on deep learning frameworks are the main focus and frontier of research in remote sensing image super-resolution reconstruction methods； （2） Model lightweighting and real-time performance are the main challenges faced by super-resolution reconstruction methods for multispectral remote sensing images in complex scenes； （3） There is an urgent need to construct public datasets for research on remote sensing image super-resolution reconstruction methods and to improve the evaluation index system. In addition， this paper also discusses the effects of methods based on bicubic interpolation， CNNs， GANs， and DPMs frameworks on remote sensing image super-resolution reconstruction in complex scenes through experiments.

The terrain of the “Asian Water Tower” area centered on the Qinghai-Tibet Plateau is complex and the weather is changeable. Especially in the alpine areas， ground observations are scarce. The existing remote sensing observation energy and water cycle elements lack a comprehensive systematic framework， and the accuracy is not high， making it difficult to obtain the understanding of the temporal and spatial distribution and variation characteristics of water balance in the Asian Water Tower area. Relying on the special project of "The Second Comprehensive Scientific Expedition to the Qinghai-Tibet Plateau"， systematic satellite remote sensing observation research on energy and water cycle in the Qinghai-Tibet Plateau was carried out. The development and sharing of satellite remote sensing datasets for 15 key elements of energy and water cycle were completed. The retrieval methods and technologies for each element were improved and developed. A satellite networking observation system for key elements of regional energy balance （6 types） and water cycle （9 types） in the Asian Water tower has been formed （including cloud， surface temperature， emissivity， radiation， albedo， Precipitation- Evapotranspiration， atmospheric water vapor， soil moisture， soil freeze-thaw， snow cover and snow depth， surface water change and terrestrial water storage， etc.）， and a systematic remote sensing dataset of energy and water cycle elements with high precision， spatiotemporal continuity and high spatiotemporal resolution has been produced. The systematic analysis based on this dataset indicates that the multi-year average spatial distribution pattern of surface evapotranspiration on the Qinghai-Tibet Plateau is controlled by the precipitation distribution and shows a decreasing trend from southeast to northwest. Evapotranspiration is mainly controlled by moisture conditions in most areas （arid and semi-arid regions）， and is dominated by radiation factors only in a few humid areas. The energy parameters that affect the evapotranspiration of the plateau and other factors of the water cycle， such as the downward long-wave radiation on the surface， show an increasing trend， while the surface reflectivity shows a decreasing trend， which has a significant correlation with the decreasing trend of snow coverage. In addition， key parameters of the water cycle such as atmospheric water vapor， soil moisture， lake area and land water storage all show a significant increasing trend， while snow coverage and the annual freezing days of the surface show a decreasing trend. Overall， since 2000， the Qinghai-Tibet Plateau has generally shown a trend of warming and humidification. The dataset provided in this paper is conducive to supporting the combination of remote sensing and models， and promoting the verification and improvement of models for regional climate-land surface and multi-sphere hydrology.

Over the past two decades， studies about the spectral invariant theory have been developed rapidly in vegetation remote sensing. The theory has been widely used in vegetation parameter measurement， canopy reflectance modeling， and biophysical parameters retrieval. A review paper titled "Photon recollision probability and the spectral invariant theory： Principles， methods， and applications" was published by the author in “Remote Sensing of Environment” in 2023 （DOI： 10.1016/j.rse.2023.113859）. The current paper provides a comprehensive overview of the background of spectral invariant theory， general principles of the theory， determination and applications of spectral invariants. Recent progresses of the theory are summarized and potential future developments are discussed. A special section is dedicated to the researches made by Chinese scholars. The goal is to provide a synthetic overview of the theory. Some new thoughts about the theory are also given in the paper.（1） The spectral invariant theory evolves from the successive orders of scattering approximation method of the multiple scattering process of photons in vegetation canopy. The theory faciliates the conversion of spectral parameters between different spectral bands and different scales （mainly between leaves and canopy） and provides new means for calculating the directional reflectance， albedo， and fluorescence escape probability.（2） Spectral invariants can be obtained through empirical methods， spectral methods， and structural methods. The spectral method is divided into single scale spectral method and spectral scaling method， and the structural method is divided into the Silhouette to Total Area Ratio （STAR） method， clumping index method， Stenberg method， and approximation method. Different methods can be cross-validated.（3） The spectral invariant theory has been applied in a number of canopy reflectance models. Based on these models， researchers have carried out a large number of inversion studies for vegetation structural parameters and physiological parameters. The principles， methods and applications of the theory can be further explored in the future.（4） New approximation methods for the spectral invariants are proposed and new formulae for the visible sunlit leaf area index and the hemispherical directional area scattering factor are summarized. Chinese researchers have made significant contribution to the development of the theory， especially in calculating the escape probability of the solar induced fluorescence using the theory.

As the main surface water resource carrier， lakes play an important role in regulating river water volume and meeting water demand for life， production and ecology. Similarly， artificial lakes （reservoirs） also play a key role in water supply， irrigation， power generation and ecological protection. In recent years， affected by climate change and human activities， the water volume and quality of lakes and reservoirs （referred to as "lakes and reservoirs"） have changed greatly. Long-term remote sensing monitoring can more objectively and accurately understand the temporal and spatial changes. The study combined multi-source remote sensing image data and satellite altimetry data to extract the area and water level of China's key monitored lakes （83） and reservoirs （118） from 2013 to 2022， estimated the changes in the water volume of lakes and reservoirs， and sorted out the monitoring data of key lakes and reservoirs through the monthly water quality reports issued by relevant national departments， analyzed the changes in their eutrophication levels， and finally conducted a comprehensive analysis of the two. The results show that： in terms of water volume， the water volume of China's key lakes and reservoirs showed an overall increasing trend during 2013~2022， with the water volume of lakes increasing by 10.65 km³ and the water volume of reservoirs increasing by 2.50 km³； in terms of water quality， the water quality of most monitored lakes was poor， ranging from mesotrophic to mildly eutrophic， and the number of lakes with mesotrophic to mildly eutrophic levels generally increased year by year； the water quality of reservoirs was relatively good， but the water quality of some reservoirs showed a trend of deterioration. The combined analysis of water volume and water quality change information found that the water volume of lakes and reservoirs changed greatly when they were at the mesotrophic level， and there were different correlations between water volume and water quality. Among them， the water volume and water quality changes of 33 lakes showed a negative correlation， especially Chengxi Lake and Shijiu Lake in the eastern plains， with correlation coefficients of -0.67 and -0.90， respectively. In addition， the water volume and water quality changes of nearly half of the reservoirs showed a positive correlation. By combining the water quantity and water quality monitoring data of lakes and reservoirs and exploring the relationship between the two， we can provide an important scientific basis for the comprehensive water resources management and ecological protection of lakes and reservoirs.

The classification of wetlands with high-resolution images is one of the research hotspots of remote sensing classification. Aiming at the complex mottling of high-resolution image wetlands and wetland hydrological boundaries fluctuate seasonally， the traditional classification of high-resolution image wetlands adopts the manual extraction feature interpretation method， which is time-consuming， laborious and has low accuracy. Therefore， how to achieve automatic and efficient interpretation of wetlands is an urgent problem to be solved. In recent years， with the rapid development of artificial intelligence technology， the use of deep learning to achieve high-resolution image wetland classification has become a new research direction. In order to promote the development of high-resolution image wetland classification technology， the latest research results of deep learning models commonly used in high-resolution image wetland classification， including deep neural networks， convolutional neural networks， and generative adversarial networks， are summarized， and the application and innovation of various deep learning models in high-resolution image wetland classification were analyzed and discussed， such as the application of ensemble learning and the construction of semi-supervised models. Finally， from the aspects of samples and models， the problems of deep learning in the classification of wetlands with high-resolution images and the possible research trends in the future are prospected.

Hongze Lake serves as the primary water supply source in northern Jiangsu and functions as the water storage reservoir for the eastern route of the South-to-North Water Diversion Project. The quality of its water is crucial to ensuring safe water supply and the sustainable utilization of water resources. Secchi Disc Depth （SDD） is an important index to measure water environment quality and plays an important role in water ecosystem. In this study， Hongze Lake was taken as the research area， and a remote sensing estimation model suitable for Hongze Lake SDD was constructed based on Landsat 8 OLI remote sensing data by using field measured SDD data and spectral data. The verified results were the Mean Absolute Percentage Error （MAPE） and Root Mean Square Error （RMSE） as 19.8% and 0.07m respectively. The constructed model was applied to Landsat 8 OLI images of Hongze Lake from 2013 to 2022， and the following conclusions were obtained： （1） The average inter-annual variation range of SDD was 0.18~0.25 m， with the highest value and the lowest value appearing in 2019~2020 and 2013~2014， respectively. The overall variation trend of the three lakes was consistent， with the highest SDD in Chengzi Lake Bay and the lowest in Huaihe Lake Bay. The main factors affecting the SDD of the lake area are wind speed， among which the sediment discharge is the main influencing factor of the Huaihe River Bay. （2） The monthly variation of SDD increased from January to August， and decreased month by month after August， with the highest value of 0.36 m in August and the lowest value of 0.20 m in May. The SDD of the lake area is mainly affected by wind speed. Chengzi Lake Bay and Lihe Lake Bay are obviously affected by meteorological factors， and Huaihe Lake Bay is affected by sediment discharge and shipping， and the factors are complicated.

Leaf Area Index （LAI） is an important parameter for studying vegetation canopy structure and physiological and biochemical characteristics. Due to the high complexity and heterogeneity of mountain’s surface and forest canopy structure， there is no unified standard for mountain LAI ground measurement methods， resulting in significant differences in ground measurement between different methods. In order to analyze the effect of various factors on the ground measurement， and improve the accuracy and reliability of LAI ground measurement data in mountainous areas， this article uses LAI2200 Plant Canopy Analyzers （LAI2200） and Digital Hemisphere Photography （DHP） to conduct LAI ground observation experiments in typical mountain forest scenes， quantitatively analyzing different optical measurement instruments， vegetation clumping effects， the impact of terrain and other factors on the LAI ground measurement in mountainous areas. The results showed that both LAI2200 and DHP optical instrument could be used to measure LAI in mountain forests. The experiment found that LAI2200 was more sensitive to LAI of coniferous forest than DHP； Regarding the impact of terrain factors on LAI ground measurement， the hinder of incident radiation by surrounding terrain fluctuations is an important factor that causes measurement errors and needs to be eliminated in the measurement； In addition， the vegetation clumping effect has a significant impact on the measurement results， and it is necessary to correct the effective LAI by introducing the Clumping Index （CI） to obtain the true LAI. This article provides effective reference suggestions for improving the accuracy and precision of LAI ground measurement in mountainous forests.

Terrestrial Gross Primary Production （GPP） is the total amount of organic carbon fixed by plant photosynthesis， and it is also the start of terrestrial carbon cycles. The remote sensing data-driven GPP models can accurately monitor the spatio-temporal pattern of GPP at the regional scale. The remote sensing data-driven GPP products support the studies of terrestrial carbon cycles， ecosystem responses to climate change and ecosystem service. However， large discrepancies in absolute magnitude， spatial distribution， and interannual variability in remote sensing data-driven GPP models lead to a large uncertainty in the estimation of global GPP. The uncertainty mainly stems from the fact that different researchers have adopted different modeling frameworks and assumptions， different remote sensing data sources， and different research periods， and thus researchers have come up with diverse conclusions. For this reason， it is necessary to summarize the results of existing representative studies to form a knowledge base and understanding of the remote sensing data-driven GPP models and their applications at the macro-scale level. Meanwhile， this study also discusses the common problems in current remote sensing data-driven GPP models and provides an outlook for future model developments.

Domain adaption transfers the source domain knowledge to the target domain to improve the classification accuracy of hyperspectral image classification model for features in different scenes. The development of domain adaptation classification methods for hyperspectral images is rapid， however， there is a lack of comparative analysis for domain adaptation methods. Therefore， the domain adaptation classification methods are classified into four categories： Distribution Adaptation， Feature Selection， Subspace Learning， and Deep Domain Adaptation. In this paper， eight typical methods are selected and three standard hyperspectral datasets from Pavia Center， Pavia University and HyRANK are used to design the comparison experiments. The experimental results show that the deep domain adaptation methods are more advantageous， among which the overall classification effect and computational efficiency of the topological structure and semantic information transfer network method are the best overall.

Based on six indexes of greenness， humidity， heat， dryness， salinity and soil and water conservation factors， a remote sensing ecological index URSEI suitable for arid areas was constructed by using PCA， and the tem-spatial characteristics of ecological quality changes in Shiyang River Basin from 2000 to 2018 were explored by using trend analysis and Hurst persistence analysis. The findings are as follows： （1）URSEI considers salinity and soil and water conservation indicators， and the contribution rate of the first principal component is higher than 80%， integrating the main information of each indicator， which is conducive to improving the comprehensiveness of ecological quality assessment. （2） From 2000 to 2018， the URSEI in Shiyang River Basin increased from 0.288 to 0.316， which tended to improve on the whole， mainly due to the strengthening of ecological management and control policies and the implementation of water transfer projects such as Jingdian Phase II. Among them， the area of ecological quality improvement is about 10 495.75 km²， which is distributed in a belt along Jinchang City， Liangzhou District and Gulang County； The degraded area covers 2 388.50 km²， which is concentrated in the main urban area of Liangzhou District in the middle reaches， the Longshoushan mining and the middle oasis area of Minqin in the lower reaches， and there is a risk of continuous degradation in the future. （3） In the future， the ecological quality of the basin will be mainly unchanged and continuously improved， accounting for about 72.55%， but at the same time， about 2 231.25 km² of the area will develop from improvement to degradation， distributed in Minquan Township of Gulang County， Huangyang Township of Liangzhou District， Yongfengtan Township， Jiaojiazhuang Township of Yongchang County， etc.， basically corresponding to the surrounding areas of the main urban areas of each district and county. It is the main expansion area of urbanization in the future.

Accurate monitoring and assessment of vegetation restoration effectiveness are essential for ecological conservation， sustainable development， and environmental management. Previous studies have primarily relied on remote sensing imagery， using two-dimensional monitoring indicators such as Fractional Vegetation Cover （FVC） to assess vegetation restoration effectiveness. However， these studies have focused solely on changes in vegetation coverage area， neglecting structural measures， which limits the precise understanding of ecological governance effects. To address these limitations， this study proposes the integration of LiDAR technology with two-dimensional optical remote sensing for a more comprehensive and accurate monitoring approach in ecological restoration areas. This method combines two-dimensional indicators derived from optical remote sensing imagery with three-dimensional indicators obtained from UAV LiDAR to capture both changes in vegetation coverage and structural differences， enabling comprehensive monitoring and evaluation of vegetation restoration effectiveness. To validate the feasibility of this method， four mine restoration areas in Beijing were selected as examples. Using UAV LiDAR point cloud data in 2022 and Sentinel-2 time series remote sensing imagery data from 2018 to 2022， four vegetation structure indicators （canopy height， canopy cover， leaf area index， and canopy entropy） and two-dimensional plane indicators of FVC were calculated for the study areas. Various analytical methods， including comparative analysis and trend analysis， were employed to evaluate the vegetation restoration situation in the study areas. The results indicate a significant increasing trend in vegetation coverage area in all mining areas from 2018 to 2022. However， when evaluating vegetation structure indicators， only one mining area exhibited consistency between structural indicators and two-dimensional evaluation results. This suggests that both increased vegetation coverage area and improved vegetation structure in this specific mining area. In contrast， other mining areas only showed an increase in vegetation coverage area， emphasizing the need for subsequent efforts to enhance the restoration and management of vegetation structure. The introduction of LiDAR technology alongside optical remote sensing provides a more comprehensive assessment approach， offering more accurate references for ecological restoration effectiveness evaluation and the further implementation of ecological restoration projects.

Soil Organic Matter （SOM）， a vital component of the soil solid phase， provides essential nutrients for plant growth and serves as a key indicator of soil fertility. Recent advancements in remote sensing technology have introduced novel approaches for efficient SOM estimation and mapping， yet challenges persist due to environmental interference and data complexity. This review systematically examines the applications of multispectral and hyperspectral data in SOM inversion and mapping， alongside critical data processing methodologies. Comparative analyses demonstrate that laboratory-acquired spectral data under controlled conditions exhibit significantly higher model accuracy and robustness compared to field-collected data， attributed to stable measurement environments. Feature selection and extraction， particularly for hyperspectral datasets， enhance inversion precision by mitigating data dimensionality and multicollinearity. Ensemble modeling frameworks integrating machine learning and deep learning outperform single-model approaches by effectively characterizing the nonlinear complexity of soil systems. Multi-temporal datasets further improve predictive capabilities by incorporating seasonal vegetation dynamics and temporal evolutionary patterns. However， optical data remain susceptible to atmospheric disturbances， especially in cloud-prone regions such as southern China， while microwave remote sensing emerges as a complementary solution for its all-weather operability and topographic adaptability. Future research should prioritize multi-source synergy strategies， including optical-SAR synergies， multi-sensor platform integration， and physics-informed machine learning to address confounding factors like crop residue cover and soil moisture. Advanced preprocessing techniques， such as wavelet analysis and blind source separation， are essential for isolating soil-specific spectral signatures. Spatiotemporal modeling frameworks that integrate soil types， agronomic practices， and climatic variables will enhance prediction generalizability. Concurrently， developing interpretable artificial intelligence models and geographically adaptive spatial interpolation methods is crucial to ensure scientific rigor and global scalability. This study provides theoretical and practical insights for leveraging multi-source remote sensing in precision agriculture and sustainable land management.

Accurately and rapidly assessing soil salinity is crucial for land quality evaluation and agricultural development. Hyperspectral remote sensing， as an effective monitoring technology， offers new avenues. Optimizing hyperspectral data processing to enhance features is critical for accurately estimating soil parameters. However， the impact mechanisms of different spectral combination treatments on soil salinity estimation need further study. This study aims to investigate the potential of hyperspectral data for estimating soil salinity under nonlinear transformation and fractional derivative combination treatments. Based on 60 typical soil samples from the Yulin area， the spectral characteristics of saline soils under different spectral transformation combinations were analyzed. The Competitive Adaptive Reweighted Sampling method （CARS） and Successive Projections Algorithm （SPA） were used to select characteristic bands as input variables. Partial Least Squares Regression （PLSR） and Random Forest （RF） models were established to compare and analyze the estimation capabilities of different spectral processing and modeling strategies for soil salinity. The results indicate that fractional derivatives， compared to integer derivatives， better reflect the absorption characteristics of saline soil spectra. Reciprocal and logarithmic transformations effectively enhance the correlation between spectral data and soil salinity information， especially in the 1905–2078 nm range. The PLSR model generally outperforms the RF model， with the CARS-PLSR achieving optimal modeling accuracy （R²=0.966）. These findings can provide a theoretical basis for the implementation of saline soil prediction and precision agriculture.

Black and odorous water is an extreme water pollution phenomenon that is a common problem in many cities， seriously affecting the well-being and satisfaction of residents. Therefore， addressing the issue of urban black and odorous water is urgent. Scientifically monitoring the urban black and odorous condition of water is the first step in the remediation of black and odorous water， providing accurate targets for subsequent assessment. Compared to traditional ground-based monitoring methods， remote sensing is a more effective way to discover black and odorous water， allowing for a wider and more rapid identification of black and odorous water areas and timely comprehensive representation of the spatiotemporal evolution of black and odorous water. This paper reviews the main research advances in remote sensing monitoring of urban black and odorous water. First， it briefly introduces the overview of black and odorous water， summarizes the remote sensing feature， and common data sources for urban black and odorous water. Second， it discusses in detail the methods for remote sensing identification and classification of urban black and odorous water and compares the applicability and advantages and disadvantages of each method. Finally， it summarizes the current status of remote sensing research on urban black and odorous water and its shortcomings and looks forward to future development directions. The aim is to provide reference and thinking for related research and better decision support for the long-term management of black and odorous water.

Fire threatens the safety of human life and property and causes great damage to ecosystems. The study of remote sensing response characteristics of burned area is important for the accurate extraction of area， quantitative assessment of fire damage and vegetation restoration. Based on Sentinel-1 SAR remote sensing images， the characteristics of unburned forest， burned area， buildings， water bodies were analyzed in six fire cases. The time series of burned area from one year before the fire to two years after it was analyzed. The results shows that the cross-polarization ratio and the backward scattering intensity of VH polarization are lower in the burned area compared to the unfired area， the backward scattering intensity of VH and VV polarization of buildings are much higher than those of other features， and the backward scattering of water bodies in both polarizations is very low， while the cross-polarization ratio is higher. From time series perspective， the backward scattering intensity for VV polarization shows obvious seasonal variations. In most cases， the backward scattering intensity for VV polarization is significantly higher within one month after the fire， and the cross-polarization ratio rapidly decreases. The time series variation of Normalized Burned Ratio index （NBR） calculated from Sentinel-2 MSI follows a consistent pattern with SAR images， showing obvious seasonal changes. It rapidly decreases within half a month after the fire and gradually recovers.

The identification of urban functional zones can assist in the decision-making process in urban construction. This paper proposes a multi-source scene feature fusion method utilizing the Transformer model for urban functional zone identification. Firstly， the Traffic Analysis Zone （TAZ） is constructed based on the road network. The graph structure of POI （Point of Interest） data is created using the Delaunay Triangulation （DT）. Additionally， remote sensing data is utilized to obtain the corresponding image objects for each TAZ. Subsequently， the POI graph structure is processed using a Graph Convolution Network （GCN） to extract social scene features. Meanwhile， the natural scene features of remote sensing data are obtained through encoding with ResNet-50. Finally， the multi-head attention mechanism of Transformer decoder is utilized to fuse multi-dimensional feature vectors， facilitating accurate identification of urban functional zone with SoftMax. Taking the main urban area of Shenyang as an example， multi-source data such as OSM （Open Street Map）， POI and remote sensing data in 2021 are used as experimental data. The results indicate that the overall accuracy and Kappa coefficient of this method are 82.2% and 70% respectively. Furthermore， the Kappa coefficient is at least 18% higher than that of the single data method and at least 9% higher than that of other fusion methods. This study innovatively employs the Transformer model to integrate social and natural scene features， effectively addresses the challenge of combining diverse features from multiple sources into an integrated representation， and provides a new technical approach for urban functional zones identification.

The Qinba Mountain region， encompassing the transition zone between North and South China， holds significant ecological importance. Investigating the phenological changes in vegetation and their response to climate change in this region is crucial for understanding the complexity of the ecological environment in the transitional zone and reconstructing historical climate patterns. Based on MODIS MCD12Q2 data and meteorological remote sensing data， this study employs trend analysis and correlation analysis methods to explore the spatiotemporal characteristics of vegetation phenology in the Qinba Mountain region from 2001 to 2020 and its relationship with climate change. Results indicate that the start， end， and length of the growing season in the Qinba Mountain region exhibit distinct vertical zonal distribution characteristics from east to west. The start of the growing season is primarily distributed from mid-to-late March to late April （70-110 days）， while the end of the growing season is concentrated from late October to late November （290-320 days）， with the majority of growing seasons falling between 180 and 260 days. Over the 20-year period， the overall characteristics of phenological interannual variation in the Qinba Mountain region show an average advancement of 0.38 days per year. The end of the growing season exhibits an average delay of 0.43 days per year. The length of the growing season displays an average extension of 0.80 days per year. Significant trend of change. Regarding the time lag response to climate factors， the start of phenology in the Qinba Mountain region shows the highest correlation with monthly temperature and potential evapotranspiration without significant time lag effects， while precipitation exhibits a time lag of approximately 1.73 months. Altitude to some extent determines the response relationship between the start of phenology and various meteorological elements in the Qinba Mountain region.

Solar-Induced chlorophyll Fluorescence （SIF）， a proxy for vegetation photosynthetic activity， has gained widespread applications. This review synthesizes the principles， progress， and key frontiers in satellite SIF remote sensing. Firstly， we introduced the SIF retrieval principles and algorithms at both ground and spaceborne platforms. The SIF retrieval methods can be divided into two categories： physically-based inversion and data-driven algorithms. The accurate separation of SIF from reflected radiance in upwelling radiation is the key challenge. The current ground-based retrievals are still limited by spectrometer resolution and Signal-to-Noise Ratio （SNR）， developing instrument-agnostic， high-robustness algorithms remains a research priority. Data-driven approaches dominate satellite SIF retrievals. However， huge uncertainties persist in red-band SIF retrieval， demanding transformative algorithmic breakthroughs. Second， we analyze global SIF satellite developments over 30 years， highlighting China’s rapid progress （e.g.， successful experiments with TanSat and Goumang）. Nevertheless， gaps persist in satellite longevity， data sharing， and scientific utilization compared to international counterparts. The next-generation TanSat-2 （to be launched in 2026） is expected to revolutionize SIF remote sensing， offering 2-km resolution， global daily coverage， and dual-band （red/far-red） SIF data—resolving critical limitations of low resolution， SNR， and revisit frequency. Finally， we investigated the progresses of spatiotemporal fusion of SIF satellite data. Machine Learning （ML）-based simulation methods have achieved high-precision simulation of SIF data and have been widely used. However， the ML-based SIF datasets represent the modeled signals driven by reflectance and meteorology， not observations. Spatial downscaling of satellite SIF products preserves observational fidelity despite lower spatiotemporal continuity than ML counterparts. Emerging multi-sensor， long-term （1995—2024）， 0.05°-resolution downscaled products hold potential to accelerate SIF science applications. Therefore， despite the inherent challenge of high-precision retrieval of weak SIF signal， advances in payload technology and quantitative remote sensing are rapidly transforming SIF monitoring capabilities. China’s SIF remote sensing program （TanSat-2） is positioned to play an increasingly pivotal role in guiding global SIF science and applications.

In order to realize more accurate mangrove extraction and monitoring， four mangrove plantation areas， including the Aojiang River coast in Wenzhou City， were taken as the study area， and the distribution of Mangrove forests was extracted and accuracy verified based on the DeepLabV3+ semantic segmentation model using Sentinel-2 remotely sensed imagery data， which was applied to analyze the spatial change of mangrove forests in the period of 2019~2023. The results show that： ① the Mangrove information extraction model constructed by DeepLabV3+ network can better distinguish Mangrove and Non-Mangrove areas， with fewer mis- and omissions； ②the semantic segmentation algorithms are significantly better than traditional machine learning methods， with the DeepLabV3+ method having the highest accuracy， with an precision of 84.89% and a Kappa coefficient of 0.82； ③The growth of Mangrove forests is greatly affected by the geographical location and growth environment， and Mangrove forests in the intertidal zone of the coast or at the mouth of the sea are more susceptible to the influence of typhoons， tides， etc.， and the encroachment of mangrove forests' growth space by exotic species， such as spartina alterniflora， etc.， are all the key factors that cause the low survival rate of Mangrove seedlings and the slow growth rate. Therefore， the semantic segmentation model based on DeepLabV3+ can better recognize and extract the Mangrove forests and provide data base support for the monitoring and assessment of Mangrove forests in Wenzhou City.

Forest ecosystems play a crucial role in regulating climate， maintaining water and soil， and balancing carbon. However， in recent years， this system has been increasingly disturbed by climate change and human activities， making precise and timely forest change monitoring urgently needed. Remote sensing technology， with its advantage of multi-temporal resolution data and automated processing capabilities， has become a key means for forest change detection. This paper focuses on multi-temporal resolution remote sensing change detection methods， systematically reviews and compares two types of technologies： bi-temporal and time series remote sensing. Bi-temporal change detection includes manual visual interpretation， traditional machine learning， and deep learning techniques； time series include research on temporal trend analysis， dynamic change monitoring， and multi-algorithm integration. By summarizing the related problems of deep learning and multi-modal time series data in practical applications， relevant solutions are proposed， providing references for improving the accuracy of forest change detection.

In this study， a comprehensive overview of snow avalanche hazards occurred in the past in Xizang area was reviewed first， followed in-depth analysis on the spatial distribution and main driving factors of snow avalanche hazards in the Xizang mountain region. Snow avalanche-prone areas for the study area were then mapped based on the spatial distribution of snow cover and DEM （Digital Elevation Model） data， and were validated using in-situ observations in southeastern Xizang. Results indicated that there are the highest frequencies of avalanche occurrences in southeastern Nyainqentanglha mountains and southern slope of the Himalayas. In the interior of plateau， avalanche occurrence is constrained due to less precipitation and flatter terrain. The perennially snow avalanche-prone areas in Xizang account for 1.6% of total area of the plateau， while it reaches 2.9% and 4.9% of total area of Xizang in winter and spring， respectively. Snow avalanche hazards and fatalities present increasing trends under global climate warming due to more human activities at higher altitudes. In addition to continuous implementation of engineering prevention and control measures in the key regions， such as in Sichuan-Xizang highway and railway sections， enhancing monitoring， early warning and forecasting services are important to prevent and mitigate avalanche hazards in the Xizang high mountain regions.

Underground coal fires are regarded as a global disaster "without geographical boundaries"， not only resulting in substantial waste of coal resources but also posing serious threats to ecological environments and the safe development of society. Remote sensing technology has demonstrated unique advantages in the long-term detection and identification of underground coal fires， providing critical technical support for monitoring and remediation efforts. Focusing on the typical surface response characteristics induced by underground coal fires， this paper elaborates on the fundamental principles of remote sensing detection and identification in coal fire areas， systematically analyzes and reviews the existing remote sensing-based methods for detecting and identifying underground coal fire zones， highlighting the strengths and limitations of each approach. Building on this analysis， the paper discusses the existing research gaps and provides a forward-looking perspective on the existing challenges and future development directions in this field.

Under the background of the “Four Million Mu” ecological project and “the community of life is the mountain， water， forest， field， lake， grass and sand”， this paper studies the habitat quality and carbon storage and influencing factors under the land use change of the Aksu River Basin and typical areas in the basin-- the Kekeya greening project， so as to provide a basis for the ecological and environmental protection policy of the Aksu River Basin， provide guidance for land use， carry forward the spirit of Kekeya， and promote the sustainable development of the Aksu River Basin. Based on the land use data from 2000 to 2020， the Habitat quality and Carbon storage module in the InVEST model were used to explore the spatial and temporal variation patterns and influencing factors of Habitat quality and Carbon storage in the Aksu River Basin from 2000 to 2020. The results show that： （1） From 2000 to 2020， the Habitat quality of the Aksu River Basin remained stable on the whole， and there was a slight increase， and the areas with better Habitat quality were distributed in the upper reaches of the basin and oases on both sides of the basin， mainly forest land and grassland land use types. In the past 20 years， the Habitat quality of Kekeya Greening Project Area has shown a steady upward trend. （2） The Carbon storage in the Aksu River Basin showed a stable growth trend during the study period from 2000 to 2020， with a total increase of 13.21×10⁶ t， and the areas with high Carbon storage values showed the distribution characteristics of "northern central cluster and southern band". The Carbon storage in the Kekeya greening project area showed an increasing trend， which was consistent with the overall trend of the Aksu River Basin. （3） Habitat quality and Carbon storage presented a significant synergic relation. The influencing factors of regional Habitat quality and Carbon storage were analyzed by using geographic detectors， land use type was the main influencing factor of Habitat quality and Carbon storage， and the interaction of influencing factors showed two-factor enhancement or nonlinear enhancement.

GRACE satellite has opened a new era of quantitative retrieval of groundwater change by remote sensing， but it has the problem of low spatial resolution. High-resolution groundwater observations will significantly improve the accuracy of local-scale hydrological process understanding， thereby offering essential data support for the development of scientifically based groundwater management policies. After sorting out the characteristic factors such as precipitation， air temperature， evapotranspiration， surface temperature， normalized vegetation index and soil water in the Yellow River Basin， partial least squares regression method was used to screen the characteristic factors respectively from January to December， and the optimal monthly characteristic factor subset was constructed. Then， the random forest algorithm was used to downscale the groundwater data of the Yellow River Basin from 0.25°× 0.25° to 1 km×1 km， and compared and verified with the measured groundwater level data. The results show that： （1） Except evapotranspiration and surface temperature， the importance of other factors changes with the change of month； （2） In the time series， the correlation coefficient and Nash coefficient of groundwater data before and after downscaling are as high as 0.95， and the root-mean-square error is 3.17 mm； （3） Spatially， compared with before downscaling， the correlation coefficient between the change data of groundwater reserves and the measured groundwater level after downscaling increased by 47.67%. The research results can meet the demand for high-resolution groundwater data in practical applications， and provide reference for the feature factor screening of groundwater downscaling research.

In response to the challenge of significantly degraded image quality and difficulties in target detection caused by heavy fog during typhoon weather， as well as the limited generalization of existing research in power scenarios， an improved algorithm that integrated dark channel defogging with YOLOv8 is proposed. The proposed algorithm employs a two-stage processing technique. Initially， color adaptive defogging is applied based on an analysis of the image's color distribution using cumulative distribution functions and the dark channel algorithm to enhance the visibility of targets within foggy environments. Subsequently， to further enhance detection accuracy for multi-scale targets in remote sensing images，A small target detection layer is introduced into the YOLOv8 network architecture. This addition facilitates deeper feature extraction for small targets while employing MPDIoU instead of CIoU to reduce computational complexity. Experimental results demonstrate that the proposed algorithm improves detection accuracy for power towers and wind turbines by 8.1% and 3.9%， respectively. These findings validate both the feasibility and effectiveness of the proposed algorithm in processing foggy remote sensing images and recognizing targets， thereby providing reliable technical support for defogging operations on such images and identifying large-scale outdoor power facilities.

Public health emergencies not only affect the characteristics of urban crowd aggregation but also lead to significant changes in emotional responses. However， current research tends to focus either on crowd aggregation characteristics or emotional responses in isolation， lacking comprehensive analysis. To address this gap， this study proposes an integrated monitoring framework that combines crowd aggregation features with emotional responses to explore the dual impact of public health events on urban psychosocial behavior. The study focuses on the main urban area of Xi’an and utilizes Baidu heat maps to calculate the crowd aggregation index and tidal index. It also integrates social media data for sentiment analysis， employing a social semantic network to reveal the main causes of emotional responses. Data analysis covers both the pandemic control period and the recovery period， comparing crowd aggregation characteristics and emotional responses using multi-source heterogeneous big data. The results indicate that the spatial intensity of crowd aggregation in Xi’an differs significantly between the pandemic control and recovery periods， with higher crowd aggregation during the recovery period. Additionally， the scope and intensity of positive emotions increased during the recovery period compared to the control period， while negative emotions were alleviated. Furthermore， the social semantic network analysis reveals a strong correlation between public sentiment and pandemic control policies. The framework developed in this study not only provides new insights into the analysis of crowd aggregation and emotions under public health events but also offers data support for government decision-making in pandemic control and emotional management. This research holds strong practical value， promoting the integration and application of multi-source spatiotemporal big data in public health and social digital governance.

The continuous changes in land use patterns have profound impacts on the ecological environment and socio-economic development of the Yellow River Basin. Understanding the dynamic changes in land use in the middle reaches of the Yellow River Basin can provide a scientific basis for achieving sustainable development in the Yellow River Basin. This study is based on four periods of Landsat remote sensing data from 1995， 2005， 2015， and 2023 in the middle reaches of the Yellow River Basin. Obtain the spatial distribution of land use in the demonstration area through Support Vector Machine （SVM） and maximum likelihood method classification， and analyze the characteristics of land use change in the study area from 1995 to 2023 using quantitative indicators such as land use change and transition matrix. The Markov model was applied to predict the land use changes in 2025 and 2030. By establishing the FLUS-Markov model， the land use changes in the study area in 2025 and 2030 were predicted. For the six types of land use changes （forest， grassland， wetland， cultivated land， construction land， and unused land）， the results show that： （1） Over the past nearly 30 years， cultivated land and forest land have decreased by 8 600 km² and 6 400 km² respectively， while the area of construction land has significantly increased by 7 500 km²； （2） The transfer directions of land use types are diverse， mainly from cultivated land to construction land and vegetation， and the landscape of each land type tends to be balanced in spatial distribution， with coordinated urban development； （3） Between 1995 and 2023， the number of forests， grasslands， and construction land showed an upward trend， while the number of wetlands， cultivated land， and unused land showed a downward trend； （4） In the next 10 years， there will be significant changes in land use in the middle reaches of the Yellow River， as humans require more land for construction.

The previous methods for hyperspectral data classification only focus on the extraction of spectral features， which often wastes some valuable spectral spatial information and leads to unsatisfactory classification results. In view of this， this paper proposes a method that combines Principal Component Analysis （PCA）， guided filtering and deep learning architecture into hyperspectral data classification. First， PCA， as a mature dimension reduction architecture， can effectively reduce the redundancy of hyperspectral information； Then， guided filtering is used to provide a simple and effective channel to obtain spatial dominant information； Finally， the stack Autoencoder model is used as a deep learning architecture to effectively process deep level multi feature image data. Train and test the algorithm using two common datasets， and then use a third common dataset to test the models trained on the other two datasets. The experimental results show that the proposed GF-FSAE algorithm achieves classification accuracy above 99%， demonstrating good classification performance and generalization ability. Compared to the CNN-AL model， the algorithm’s accuracy is slightly higher， verifying the superiority of the spectrum-spatial hyperspectral image classification framework。

The Liaohe Estuary coastal wetland is the northernmost estuary wetland in China， which is an ideal breeding and migration station for many kinds of waterfowl. In recent years， several ecological restoration projects have been carried out to improve the habitat quality in this region. Accurately mapping the landcover types using remote sensing is very important for efficiently evaluating wetland habitat quality and restoration effectiveness. However， most of the classification methods in the Liaohe Estuary were object-oriented， and the mapping results were not fine enough and needed to be updated in years. The applicability of pixel-level method and dense time-series information in this region needed to be further evaluated. This paper relied on Google Earth Engine （GEE） platform， utilized Sentinel-2， Sentinel-1， and topographic multi-source data to extract the features including spectral indices， texture， topography， backscattering， and phenology from the dense time-series vegetation indices. Multi-year sample datasets were generated by field sampling and sample migration， and the pixel-level fine classification mapping from 2018 to 2022 was carried out based on the random forest model. The effects of different features on the classification accuracy were also evaluated. The classification method combining GEE and dense time-series information got an overall classification accuracy of 95.77%. Adding phenology features improved the accuracy most obviously， especially for the mixing between suaeda salsa and reeds， rice， or aquaculture ponds. Adding texture and backscattering features significantly improved the accuracies of aquaculture ponds and construction land. In the last five years， aquaculture ponds decreased while the mudflat and suaeda salsa expanded， which indicated the effects of ecological restoration project. The results provide data and technology supports for analyzing the spatial-temporal changes and driving mechanism of coastal wetland， which is of great significance for strengthening the protection and restoration of wetland ecosystems.

In the process of obtaining agricultural planting structure through remote sensing images， the uncertainty of remote sensing images itself and the classification process will inevitably reduce the accuracy and reliability of mapping results. As the basic unit of agriculture， parcels have the advantages of classification accuracy and uncertainty control. However， current remote sensing classification uncertainty research methods mostly use pixels as the basic unit， which is difficult to directly apply to parcels crop classification. Therefore， using parcels as the basic spatial unit and selects information entropy as the posterior uncertainty evaluation index to carry out crop classification and uncertainty analysis experiments in the Ningxia Irrigation District of the Yellow River. With the help of temporal characteristics and multiple value characteristics inside the parcel， the influence effects of random uncertainty and fuzzy uncertainty in remote sensing data on posterior classification uncertainty are analyzed. The experimental results show that： （1） Compared with pixel-scale classification， parcel-scale classification can effectively weaken posterior uncertainty； （2） Random uncertainty has a significant impact on posterior uncertainty and is significantly correlated with temporal characteristics； （3） The fuzzy uncertainty introduced by mixed pixels at the edge of the parcel and internal mixed heterogeneity amplifies the random uncertainty and further affects the posterior uncertainty. Targeted reduction measures can effectively reduce classification uncertainty， and the reduction effect is basically consistent with the correlation analysis results. The research results can provide ideas and methods for subsequent improvement of crop classification process and uncertainty control.

Forest canopy closure is a crucial parameter in forest resource inventory that plays a significant role in evaluating and monitoring the stability of forest ecosystems. With the continuous development of remote sensing technology， the estimation of large-scale forest canopy closure using multi-source remote sensing data has become a hot research topic. In this study， a regression model was constructed using machine learning algorithms based on laser point cloud data and multi-source optical remote sensing data to estimate forest canopy closure in large forested areas. Firstly， the dependent variable of the regression model， which is the true values of forest canopy closure， was calculated from the Airborne Laser Scanning （ALS） point cloud data. Secondly， 18 independent variables， such as vegetation indices and texture， were extracted from Sentinel-2 MSI， Landsat-8 OLI， and Sentinel-1 SAR images. Then， taking 14 forest plots in Guangxi as an example， the impacts of different independent variable combinations on forest canopy closure inversion were experimentally analyzed using two machine learning models， Random Forest Regression （RFR） and Support Vector Regression （SVR）. Finally， the best variable combination and machine learning method were selected to map the forest canopy closure in Guangxi. The experimental results showed that RFR performed better than SVR， and the S2+S1 combination had the highest accuracy， with a correlation coefficient R² of 0.703， Root Mean Square Error （RMSE） of 0.19， and Mean Absolute Error （MAE） of 0.13. Additionally， polarization features can significantly improve the inversion accuracy of forest canopy closure.

Above Ground Biomass （AGB） is an important indicator of grassland ecosystem function and grassland productivity. Accurate estimation of AGB is of great significance for grassland management and ecological environment assessment. Taking part of grassland in Xilinhot， Hulunbuir and Ordos grassland of three different types as the research area， based on UAV multi-spectral data， LiDAR data and field measured sample data， multiple texture and vegetation index were obtained， and different feature combinations were obtained through various feature screening methods. Five regression analysis algorithms， including Random Forest（RF）， Multiple Linear Regression（MLR）， BP neural network（BP）， Support Vector Regression（SVR）， and Long Short-Term Memory neural network（LSTM）， were used to construct a grassland AGB estimation model， and the optimal estimation model was obtained after comparison and evaluation， and the spatial biomass estimation was carried out. The results indicate that： （1） The feature combination selected by the feature importance method， including spectral indices and the measured average plant height （Mean Height） within the sample plots， achieved high accuracy in multi-model comparisons for grassland AGB estimation； （2） The Random Forest model outperformed other models in estimation accuracy. Using the coefficient of determination（R²），Root Mean Squared Error（RMSE），and Mean Absolute Error（MAE） as evaluation metrics，the test sample achieved an R² of 0.859， with RMSE and MAE values of 35.17 g/m² and 28.22 g/m²， respectively. The study demonstrates that integrating multi-source remote sensing features with machine learning algorithms can effectively overcome the limitations of traditional AGB estimation methods. The Random Forest model based on optimized feature combinations provides a reliable methodological reference for accurate grassland AGB estimation.

It is of great significance to have a detailed and accurate inventory of anthropogenic CO₂ emissions and to strengthen the effective control of CO₂ emissions in order to achieve the carbon peak and neutrality targets. Therefore， this study evaluated the reliability of EDGAR and ODIAC global anthropogenic CO₂ emission inventories in China and developed a high-accuracy and high-resolution estimation model for anthropogenic CO₂ emission based on multi-source remote sensing data and the Random Forest algorithm. The anthropogenic CO₂ emissions from 2005 to 2020 at the resolution of 1 km×1 km were estimated and the clustering patterns of anthropogenic CO₂ emissions were explored using the spatial autocorrelation. The spatial and temporal variations of anthropogenic CO₂ emissions in China and key areas of clustering patterns were then analysed. Results show that the estimation model of anthropogenic CO₂ emissions in China can effectively integrate the advantages of global inventory and remote sensing data， providing valid support for the spatial and temporal estimation of anthropogenic CO₂ emissions at a fine scale； from 2005 to 2013， China's anthropogenic CO₂ emissions increased dramatically from 5.53 billion tonnes to 10.73 billion tonnes； the growth rate of CO₂ emissions declined in 2013-2020， with changes tending to flatten out； China's overall anthropogenic carbon emissions demonstrate regional characteristics of "high in the east and low in the west， high in the coastal and low in the inland"， clustering in the focus areas of Yangtze River Delta， Beijing-Tianjin-Hebei region and western region； the characteristics of anthropogenic carbon emissions in different regions are closely related to their industrialization process， level of economic development， scale of counties and so on.