Search

Rapid and accurate agricultural greenhouse extraction with remote sensing imagery is essential for providing spatial information for precision agriculture. Benefiting from local spatial perception, deep learning based object extraction methods have achieved satisfactory performances in extracting geo-objects. However, they fail to work for large-scale greenhouse extraction since the objects are sparsely distributed in the background and densely distributed in the foreground, where the local spatial perception causes redundant computation and false detection problems. In this paper, we propose a layout attention network (LANet) framework for large-scale greenhouse extraction using remote sensing imagery, which replaces the local spatial perception with spatial layout perception, i.e., a sparse global layout to identify the sparse background and a dense local layout to identify the dense foreground. To address the shortcoming of the sparse background, which leads to redundant computation, a sparse global layout awareness module is formulated as a scene classifier. This accommodates the global layout attention map of the global scene features by adopting a layout-shared convolutional neural network (CNN) backbone for generating class-agnostic layout priors and global channel attention for aggregating discriminative global layout features, ensuring robust sparse background identification. Then, to alleviate the problem of the dense foreground, which causes false detection, a dense local layout awareness module is proposed to incorporate the local layout attention map and rotated region of interest (RRoI) features. The RRoI features are then further embedded to guide the initial RRoIs for object location refinement by aligning the initial RRoI locations in a layout-sensitive attention mechanism and achieving semantic enhancement by taking the local layout density as a semantic prior to assign a reliable class score map. The experimental results obtained on an agricultural greenhouse benchmark dataset and a large-scale agricultural greenhouse extraction dataset illustrate that the proposed framework can outperform the state-of-the-art object extraction methods in both speed and accuracy, and has a high generalization ability for large-scale dense object extraction. [ABSTRACT FROM AUTHOR]

Abrupt air pollution accidents can endanger people's health and destroy the local ecological environment. The appropriate emergency response can minimize the harmful effects of accidents and protect people's lives and property. This paper provides an overview of the key emergency response technologies for abrupt air pollution accidents around the globe with emphasis on the major achievements that China has obtained in recent years. With decades of effort, China has made significant progress in emergency monitoring technologies and equipment, source estimation technologies, pollutant dispersion simulation technologies and others. Many effective domestic emergency monitoring instruments (e.g., portable DOAS/FT-IR systems, portable FID/PID systems, portable GC-MS systems, scanning imaging remote sensing systems, and emergency monitoring vehicles) had been developed which can meet the demands for routine emergency response activities. A monitoring layout technique combining air dispersion simulation, fuzzy comprehensive evaluation, and a post-optimality analysis was proposed to identify the optimal monitoring layout scheme under the constraints of limited monitoring resources. Multiple source estimation technologies, including the forward method and the inversion method, have been established and evaluated under various scenarios. Multi-scale dynamic pollution dispersion simulation systems with high temporal and spatial resolution were further developed. A comprehensive emergency response platform integrating database support, source estimation, monitoring schemes, fast monitoring of pollutants, pollution predictions and risk assessment was developed based on the technical idea of "source identification - model simulation - environmental monitoring" dynamic interactive feedback. It is expected that the emergency response capability for abrupt air pollution accidents will gradually improve in China. [Display omitted] [ABSTRACT FROM AUTHOR]

Coal fire disasters generally occur in major coal-producing countries. They destroyed many coal resources and triggered various environmental problems. The identification and state assessment of coal fire areas are paramount to coal fire management. In this paper, a method for identifying coal fires by integrating land surface temperatures (LSTs) from multitemporal thermal infrared remote sensing and subsidence information from multitemporal interferometric synthetic aperture radar (MT-InSAR) is proposed. The method primarily consists of estimating the LSTs and subsidence values, normalizing and equidistantly partitioning the LSTs, estimating the thermal anomaly frequency (TAF), extracting high-subsidence areas, determining the thermal anomaly frequency threshold (TAFT), and identifying coal fire areas according to the TAFT. The coal fire ratio index (CRI) was developed to quantitatively evaluate the severity of coal fires and its variations. Midong District, Urumqi, Xinjiang, China, was selected as the study area. Forty-five Landsat 8 images and sixty-one Sentinel-1 SAR images were used to retrieve LST and subsidence time series, respectively. The proposed method and CRI were applied to identify and evaluate coal fires in the study area. The results demonstrate the satisfactory reliability of these methods compared with field surveys. The maximum CRI was 6.145 × 10-4 (July 28, 2019), and the minimum CRI was 2.685 × 10-5 (January 13, 2020). Coal fires in summer were more severe than those in winter. The CRI profile presented seasonal and annual periodicities and was affected by the local climate. The soil cohesion and humidity are higher in winter due to possible snowmelt, which weakens the interaction between oxygen and coal seams and the subsequent combustion of coal. The LST, thermal anomaly, and deformation information were combined in a time series overlay analysis to reveal the differences in the LSTs and subsidence values among mining areas, coal fire areas, and areas with both mining and coal fires. The results verify the applicability and reliability of the proposed method. [ABSTRACT FROM AUTHOR]

The use of remote sensing technique to monitor atmospheric water vapor is significant for weather and climate studies. However, the general methods of retrieving precipitable water vapor (PWV) with high precision and high resolution using remote sensing satellite has hardly been investigated, which becomes the focus of this paper. A general remote sensing PWV retrieval (GRPR) method that uses level-1 near-infrared (NIR) data sympathized by a global navigation satellite system (GNSS) is proposed. In this method, the atmospheric transmittance coefficients are determined by combining the radiative transmission model and the central wavelength interpolation method instead of directly using the traditional empirical values. Next, the PWV derived from remote sensing NIR data is obtained using the adaptive seasonal exponent model rather than the traditional transmittance–water vapor lookup table method. Furthermore, the accuracy of remote sensing NIR PWV is further calibrated by establishing the seasonal relationship between PWV residual and elevation. The corresponding NIR data from the Fengyun-3A (FY3A) satellite over the period of 2013–2015 in China are selected to validate the proposed method. Statistical results show the good performance of GRPR method for internal and external accuracies with root mean square (RMS) improvement rates of 76.8% and 72.4%, respectively, compared with the FY3A level-2 PWV products. In addition, the proposed method has good robustness and is almost unaffected by the PWV magnitude at different seasons. Proposed GRPR method is also applied for PWV retrieval at different time scales, further showing its superiority of retrieving PWV with high precision and high resolution. These results indicate the good application prospect of the GRPR method proposed in this study for generating remote sensing PWV products. • The optimal atmospheric transmittance coefficients are determined in this study. • An adaptive seasonal exponent model is developed to retrieve remote sensing PWV. • The accuracy of remote sensing NIR PWV is further calibrated using DEM data. • GRPR method is applied for remote sensing PWV retrieval at different time scales. [ABSTRACT FROM AUTHOR]

• The ISP method allows for gathering a large number of historical training samples. • The SRVI values of the same crop planted in neighboring years range from 0.91 and 1.2. • A distinct northward expansion trend is evident for all four major crops, in Heilongjiang Province, China. Mapping agricultural information such as cropping area and type is of great significance for land use and food security and a common method to retrieve such information is satellite imagery classification. The current image classification techniques require a statistically large number of training samples to obtain representative spectral signatures, ideally collected during the satellite overpass time/date. Although this prerequisite sounds simple and easy in principle, it is extremely challenging and sometimes impossible when classifying historical satellite images, as one simply cannot go back in time to collect training datasets or ground truthing. In this paper, we introduce an iterative and subsequent proximation (ISP) method to circumvent this problem for historical image classifications for historical crop mapping. This method assumes that the same crops grown two years apart have similar spectral properties within the same growing season, with limited variation. This allows the use of the crop information classified in any given year (n) to be used as training samples for the next year (n + 1) or the previous year (n-1). Iteratively repeating n-1, n-2... process leads to the mapping of historical cropping information without a priori ground truthing data. To demonstrate the ISP feasibility, we first used the historical Landsat time-series data to map four major stable crops (rice, maize, soybean, and wheat) in Hailun County, Heilongjiang Province, China, and then further expanded the ISP application to the entire Heilongjiang Province to map cropping areas from 1982 to 2020, using the classification and regression trees (CART) algorithm. The results were validated for those years when actual cropping measurement data were available and further verified with statistical data for other years. The results indicated that the proposed ISP method was appropriate for the historical mapping of four major crops, with a mapping accuracy of approximately 80 % when validated with field data, and correlation coefficients of 0.86, 0.85, 0.91, and 0.94, respectively when compared with historical statistical cropping data. The crop mapping results showed distinct trends of northward expansion in each of the four crops in Heilongjiang Province, which agreed well with previous studies. In conclusion, the ISP method is quick, easy, and convenient for historical crop mapping, which is important in understanding the agricultural production history. [ABSTRACT FROM AUTHOR]

Co-seismic landslide mapping after earthquake event is essential for emergency rescue, geohazard prevention, and post-disaster reconstruction. Most co-seismic landslide mapping is primarily achieved via field surveys or visual interpretation of remote sensing images. However, such methods are highly labor-intensive and time-consuming, particularly over large areas. This paper proposed an automated co-seismic landslide mapping approach, which has three main advantages compared to state-of-the-art methods. First, it removes the dependence on the manual labeling for training samples through an unsupervised change detection process. Second, the approach takes the lead in introducing multi-scale extended morphological profiles to comprehensively describe the characteristics of various landslides induced by earthquake. Third, it is also the first attempt to employ ensemble strategy in landslide identification, which integrates the advantages of different machine learning-based classifiers, further improving the accuracy of recognition. Three experiments were carried out through multi-temporal Sentinel-2 and PlanetScope images acquired in China and Haiti. The results demonstrated the effectiveness and superiority of the proposed approach compared to other methods, providing an effective solution for complicated co-seismic landslide mapping task in the future. [ABSTRACT FROM AUTHOR]

Aerosol types have crucial and distinguished impacts on climate effects and cause the difference of air pollution. However, direct retrieval of aerosol type by satellite remote sensing is difficult, and the satellite products of aerosol type are seriously lacking. At present, the identification of aerosol type is mainly based on ground-based observation data, but the coverage area of ground-based stations is limited, and large-scale spatial continuous monitoring cannot be carried out. The existing aerosol type identification methods based on satellite data need to combine multiple satellites data (e.g. Moderate-resolution Imaging Spectroradiometer (MODIS), Ozone Monitoring Instrument (OMI), etc.) which leads to a low temporal resolution. The research about aerosol type identification only based on geostationary satellite data is lacking. Therefore, aerosol type identification based on the multi-parameter threshold method and the neural network method are proposed and compared in this paper. The neural network jointly using spectral and temporal information was trained and tested by the measurements from 26 sites of Aerosol Robotic Network (AERONET) and Sun-Sky Radiometer Observation Network (SONET). For the overall validation, the aerosol type identification based on the neural network is better than that based on the multi-parameter threshold method, and the accuracy reaches 71.44%. Atmospheric environment monitoring and quantitative remote sensing can be supported by this study. • Aerosol type, neural network, geostationary satellite [ABSTRACT FROM AUTHOR]

• Sensor technology developments drove advances in biophysical estimation techniques. • CubeSat constellations will increase VHR data accessibility to various users. • Sentinel-2 red-edge bands present prospects for operational crop stress detection. • RF is prominently used in the literature despite new developments in algorithms. • Retrieval models developed in Mediterranean areas do not apply to semi-arid areas. This paper provides an overview of the recent developments in remote sensing technology and machine learning algorithms for estimating important biophysical and biochemical parameters for precision farming. The objectives are (i) to provide an overview of recent advances in remotely sensed retrieval of biophysical and biochemical parameters brought by the developments in sensor technologies and robust machine learning algorithms and (ii) to identify the sources of uncertainty in retrieving biophysical and biochemical parameters and implications for precision agriculture. The review revealed that developments in crop biophysical and biochemical parameters retrieval techniques were mainly driven by announcements and the availability of new sensors. Two ground-breaking events can be identified, i.e., the availability of Sentinel-2 and the SuperDove constellation. The two provide high temporal-high spatial resolution data relevant for site-specific management and super-spectral configuration, enabling retrieval of crop growth and health parameters. The free availability of Sentinel-2 triggered the testing of its spectral configurations and upscaling of retrieval approaches using simulated data from field spectrometers and airborne hyperspectral sensors. SuperDoves will likely reduce the cost of very high-resolution data while providing unprecedented capabilities for detailed, accurate and frequent characterisation of field variability. Studies showed that the red-edge bands and hybrid models coupling Radiative Transfer Model (RTM) and machine learning regression algorithms (MLRA) are promising for operational and accurate monitoring of stress-related crop parameters to aid time-sensitive agronomic decisions. However, such models were tested in Mediterranean climates and performed poorly in African semi-arid areas and China's temperate continental semi-humid monsoon climates. Therefore, locally-calibrated RTM models incorporating crop-type maps and other spatio-temporal constraints may reduce uncertainties when adapted to data-scarce regions. Generally, permanent experimental sites and a lack of systematic calibration data on various crops are some limiting factors to using remote sensing technologies for PA in Sub-Saharan Africa. Other complexities arise from farm configurations, such as small field sizes and mixed cropping practices. Therefore, future studies should develop generic, scalable and transferable models, especially within under-studied areas. [ABSTRACT FROM AUTHOR]

Greenhouses have revolutionized farming all over the world. To estimate vegetable yields, greenhouse mapping using high spatial resolution (HSR) remote sensing imagery is very important. Although automatic greenhouse mapping methods have been proposed, they are often applied in limited small-scale areas (i.e. a parcel, a city, or a province). Large-scale greenhouse mapping (i.e. national-scale) faces the diversity of greenhouses in different areas, the difficulty of the simultaneous extraction of the number and area of greenhouses, and the dense spatial distribution of greenhouses. In this paper, to solve the problem of large-scale greenhouse mapping, a novel data-driven deep learning framework is proposed, which we refer to as the dense object dual-task deep learning (DELTA) framework. The dual-task learning module simultaneously extracts the number and area of greenhouses by adopting a greenhouse area extraction branch and a greenhouse number extraction branch. A high-density-biased sampler module is proposed to select more samples in areas with a dense distribution, so that the trained model is more effective at dense greenhouse extraction. Six regions in China were selected for evaluation, which obtained a performance increment of 1.8% in mean average precision (mAP) when compared with Faster R-CNN. Finally, the whole of China was taken as the research area, and remote sensing image tiles at a 1-m spatial resolution from all over China were obtained. All the images were captured by different sensors and downloaded from open-source sites or purchased. The experimental results indicate that more than 13 million greenhouses were extracted in China. [ABSTRACT FROM AUTHOR]

• An ecological risk assessment model of Jiaozhou bay in China is established. • The model is developed based on multi-source satellite and in-situ sample datasets. • The effects of multiple risk sources threatening the ecology are considered. • The risk sources for multiple years are detected using remote sensing images. • The ecological risk situation of Jiaozhou Bay for multiple years is assessed. Healthy coastal area have high ecological service value. However, because of the uncontrolled utilization of marine resources and the discharge of pollutants, the ecological environments of many coastal areas are faced with ecological risks, such as oil spills and Enteromorpha blooms. Therefore, it is of great significance to monitor and assess the ecological risks in coastal areas. The remote sensing technique can be used to obtain frequent and wide-coverage images of the areas of interest, and has become an effective tool to monitor the ecological risks occurring in coastal areas. In this study, taking Jiaozhou Bay in China as the study area, an innovative ecological risk assessment framework was established using multi-source remote sensing images and in-situ sample data, which fully considers the potential harm of oil spills and Enteromorpha blooms on the ecology of the bay. Specifically, the oil spill risk for Jiaozhou Bay from 2017 to 2019 was identified by a deep learning based method from synthetic aperture radar (SAR) images, and the Enteromorpha bloom risk for the corresponding years was identified by a floating algae index (FAI) threshold method from multi-source optical images. Then, using habitat information and ground observation data about the marine species in the bay, environmental vulnerability maps were generated. Finally, a disaster-causing factors and disaster-bearing factors weighted risk assessment model was developed. The risk source monitoring results show that the occurrence frequency of oil spills in Jiaozhou Bay decreased from 2017 to 2019, while the outbreak intensity of Enteromorpha blooms generally became higher. The ecological risk assessment results reveal that the high risk areas of Jiaozhou Bay are mainly around the transportation regions and the ports, with a higher risk degree in the eastern and southern parts. To support our work, the well-trained oil spill detection network model and some information about the oil spill events that have occurred in Jiaozhou Bay are provided in this paper. [ABSTRACT FROM AUTHOR]

[Display omitted] • Eco-environment assessment of 3 756 urban residential quarters in Xi'an, China. • Factors affecting the ecological environment of residential quarters have been detected. • Specific reference construction values for the eco-environment of residential quarters have been proposed. Residential quarters are a critical component of urban areas, which significantly affects people's health and quality of life. In recent decades, the high-density and crowded residential quarters have become the mainstream organizational model of construction in China, because of the rapid and massive urbanization in many cities. This results in the contradiction between the eco-environment needs and the accommodation needs. Therefore, a systematic evaluation of the eco-environment in residential quarters is urgently needed to address the drawbacks of current residential quarters' planning and design. This paper aims to explore ways to improve the eco-environment with residential quarters from the perspective of China. We first used the Remote Sensing Ecological Index (RSEI) to evaluate the eco-environmental quality of 3, 756 residential quarters in the main urban areas of Xi 'an, China. Then, we quantitatively analyzed the landscape patterns of the impervious surfaces in these residential quarters. Finally, we characterized the impacts of residential quarters layout on the ecological environment by using the vegetation coverage (FVC), the normalized difference bare soil index (NDBSI), building density (BD), the residential density (RD), and the population density (PD) as main factors. The results are as follows: (1) Only 20.5 % of residential quarters with good eco-environments were distributed between the Second and Third Ring Road, indicating the need for improvement in creating livable eco-environments. (2) The shape of the impervious surface of excellent ecological quarters is complex and broken, the average patch area is smaller, and the degree of aggregation and combination of patches is lower compared to residential quarters with poor ecological quality. Therefore, it is recommended to reduce the connectivity of the impervious surface to improve the eco-environment of residential quarters. (3) Among the factors affecting RSEI of residential quarters, FVC, and NDBSI contribute more than other factors. For the construction of excellent eco-environments in residential quarters, we suggest that vegetation coverage should be no less than 0.4, the impervious surface coverage should be no more than 0.53, the building density should be between 0.28 ∼ 0.35, residential density should be no more than 0.22, and the population density of residential quarter should be under 0.29. Overall, for open and low-density residential quarters, the living environment can be improved by enriching vegetation and reducing impervious surfaces, while compact, high-density residential quarters can create a livable eco-environment by optimizing the impervious surface landscape patterns. [ABSTRACT FROM AUTHOR]

Understanding the driving mechanism of natural conditions and human activities on wetland ecosystem service value (WESV) change is important for the sustainable development of wetland resources. The existing studies often analyze the direct effects of natural conditions and human activities on WESV, however, the interaction mechanism of them on WESV is still unclear. Therefore, based on the remote sensing image data of Liaoning Province China in 1990–2020, this paper combined Emergy with Partial least squares-structural equation model (PLS-SEM) to evaluate main WESV and explore the interaction mechanism between socio-economic development factor (F sed), climate factor (F cl), environment factor (F en)and terrain factor (F te) on WESV change. The results showed that the provisioning (WESV P) and regulating (WESV R) services value increased by 6.93 E+14 and 3.01 E+14 solar emjoules (sej), respectively, while the cultural services value (WESV C) decreased by 1.50 E+14sej. Overall, the total value of wetland services value (WESV T) increased by 8.45 E+14sej during the study period, with a growth rate of 37.88%. F sed was the most significant factor driving WESV growth, with path coefficient of 0.53, followed by F en and F te , with path coefficient of 0.23 and 0.20, respectively, while F cl had the greatest negative effect on WESV, with path coefficient of −0.18. From the perspective of the interaction between each factor, F cl weakened the positive effect of F sed and F en on WESV, at the same time, F sed weakened the positive effect of F te on WESV. Through the indirect influence of F te , the adverse effect of F cl on WESV was further enhanced. This study innovatively revealed the interaction mechanism of natural conditions and human activities on WESV change, and the results could provide scientific basis for the management and sustainable development of wetland resources. [Display omitted] • Emergy theory was applied to calculate the main wetland ecosystem services value (WESV). • The interaction mechanism between natural conditions and human activities on WESV change was revealed by PLS-SEM. • F sed played an important role in the indirect effect of WESV change, and the interaction between F cl and F en exacerbated their negative effect on WESV change. [ABSTRACT FROM AUTHOR]

Ocean satellites provide accurate and precise data across various scales, making them a vital tool for investigating the association between global change and ocean processes. However, low-quality data creates unavoidable gaps in satellite data, diminishing its usefulness and continuity. These deficiencies can be resolved by implementing machine learning techniques as valuable tools. This paper details a new satellite data prediction tool titled "SatelliteFixer". The SatelliteFixer model, utilizing a custom-built random forest structure, can generate dependable data with enhanced temporal-spatial continuity. This model has demonstrated feasibility with diverse satellite data sources and light bands, and outperforms the basic machine learning approach. The juxtaposition of model data with in-situ cruise sampling results allows for widespread analysis of the movement and dispersion of suspended sediment. The above entails the inversion of long-term events and the observation of short-term events, which enables accurate seasonal analysis using continuous data without the influence of uneven data volume distribution and outliers, and is also the first-time satellite data has tracked the entire process of pulsed artificial flooding. SatelliteFixer provides a fresh outlook for detailing the varying trends on consecutive timescales and successional spaces among ocean processes. • A data-driven model for ocean satellite missing data prediction is established based on a specially designed machine learning structure. • The feasibility and availability of model result are demonstrated with MODIS Aqua and GOCI visible light data. • As an example of application, SatelliteFixer data are used with in-situ cruise data to analyze the transportation and distribution of suspended sediment. • SatelliteFixer data enables the accurate seasonal analysis and the first tracking of the entire process on pulsed artificial flooding. [ABSTRACT FROM AUTHOR]

A catastrophic landslide occurred on 21 July 2020, 30 km from Enshi city, in Mazhe County of Hubei province, China. In this paper, we aimed to investigate the kinematic evolution and volumetric change related to this landslide using multi-source remote sensing measurements from synthetic aperture radar (SAR) and optical data. C-band Sentinel-1 and X-band TerraSAR-X SAR data are analyzed using several multi-temporal interferometry (MTI) time-series techniques including Persistent Scatterer Interferometry (PSI), Small Baseline subset (SBAS), and Combined eigenvalue maximum likelihood Phase Linking (CPL). The spatial pattern of surface deformation resulting from the interferometric analysis is then statistically analyzed to retrieve the pre-failure and post-failure displacements. Co-failure motions are analyzed using an image correlation technique applied to both the Planetscope and Sentinel-2 images. Moreover, 4 pairs of bistatic TerraSAR-X/TanDEM-X (TDX) data are utilized to generate high-precision digital elevation models (DEMs) and estimate the volumetric change related to the main slope failure. The pre-failure ground deformation analysis results suggest that the landslide was already active before the July 2020 failure, with the seasonality and hydraulic diffusivity being characteristics of a slow-moving landslide. Among the three different MTI methods applied, the CPL method results in a greater measurement points (MPs) density than the PSI and SBAS method when estimating the pre-failure movement of the Shaziba landslide. The July 2020 Shaziba disaster is divided into three main parts: (1) slightly horizontal deformation of 0.5–1.5 m within the northern part with ground and house cracks, (2) less collapse in the eastern part with horizontal motions reaching 30 m and (3) a highly eroded western part where vegetation was wholly lost in the main event, resulting in an collapse volume of approximately 4.98 million m3, out of which approximately around 3.4 million m3 was deposited and approximately 1.58 million m3 was washed away into the Qing River. After the failure, the marginal scrap of the main failure body, above crown of landslide and eastern part showed instability with rates of 20–30 mm/yr, suggesting that the failure zone may continue to expand. • Multi-sensor data integration to quantify kinematics and life cycle of the landslide. • Improvement in increasing density of scatterers using Combined eigenvalue maximum likelihood Phase Linking (CPL). • Analysis of the time lag between triggering factors and changes in background velocity. • Geodetic quantification of erosion and deposition during the main slope failure using time-series of Tandem-X data. [ABSTRACT FROM AUTHOR]

[Display omitted] • A novel framework is proposed to assess the risks of algal blooms in lakes. • Vine Copula was used to integrate the quantity-distribution of algal blooms. • The framework was used effectively in risk assessment of Chaohu Lake, China. • The framework can handle the problem of spatio-temporal dynamics in risk assessment. Algal blooms threaten water safety and quality, particularly in lakes with vulnerable environment and advanced urbanization. Due to low water exchange rate and multiple human activity influences, algal blooms in inland lakes display highly dynamic spatio-temporal distributions. This paper proposes a comprehensive framework for risk assessment of algal bloom coupling quantity and distribution based on remote sensing observations. The framework comprises algal bloom recognition, spatial zonation, quantity-distribution multivariate joint distribution construction, and risk assessment. The framework was applied to China's giant shallow and algae-prone lake, Chaohu Lake, to assess the annual algal bloom risk from 2000 to 2021. The results showed that this proposed method could achieve a more comprehensive and objective risk assessment by coupling the quantity and distribution of algal blooms. Furthermore, the internal dependence of algal bloom fluctuation in different zones was explicitly presented. This framework promotes systematic integration of algal bloom risk assessment, taking into account its highly spatio-temporal dynamic feature. It creates a quantitative approach to gain insights into the evolution of quantity and distribution in various environmental assessments. [ABSTRACT FROM AUTHOR]

Ecological restoration has been risen to a national strategic project for China and become an significant way to realize ecological civilization construction. As an important way to measure the effectiveness of ecological restoration engineering, the corresponding evaluation provides a reliable basis for ecological restoration decision-making and ecosystem management. Based on remote sensing technology with the advantages of short revisit time, large monitoring range, high spatial resolution, this paper proposes a remote sensing evaluation method of watershed ecological restoration effect, so as to monitor the implementation effect of ecological restoration engineering in the watershed and technically check whether the expected goals set before the restoration are achieved. The comprehensive ecological restoration engineering of Yongding River has been implemented since 2016, and it was the first focus of the coordinated development of Beijing-Tianjin-Hebei region to make advances in ecology. In this paper, year 2015 and 2020 were selected as the before and the after the ecological restoration engineering of Yongding River. The research object was the riparian zone from Luopoling Reservoir to Daning Reservoir in Yongding River Watershed. After obtaining and processing remote sensing images, ZY-3 images were used to obtain land cover data by means of human-computer interactive interpretation, and Landsat images were used to retrieve NDVI data. Based on these data, the changes of the eight indicators of water resources, water ecology and ecological function in the study area were calculated to evaluate the ecological restoration engineering effect. The results showed that the study area has basically realized the maintenance of ecological water quantity, formed the ecological corridor of Yongding River, and effectively enhanced the ecological function and improved the water environment. By verifying the feasibility of remote sensing evaluation method, this paper could provide relevant decision makers and planners with an effective evaluation method for the ecological restoration engineering effect based on remote sensing data and technology. • Remote sensing images and land cover data were utilized to calculate the indicators before and after ecological restoration. • Ecological restoration engineering effect was evaluated from water resources, water environment and ecological function. • Based on the proposed method, the ecological restoration effect in the study area was proved to be satisfactory overall. [ABSTRACT FROM AUTHOR]

• The landscape pattern of wetland is closely related to the population of cranes. • Landscape pattern has a time-lag effect on crane population. • Regression forecasting model based on Landscape pattern predicts crane population of each year. • Predicting crane populations can provide data support for wetland reserves. Reasonable prediction of the crane population that arrives at Poyang Lake Reserve for overwintering every year is of great significance to protect cranes and to provide data about reserve planning. In this paper, considering the time-lag effect of landscape index on crane population, the Landsat satellite images of Poyang Lake wetland in 19 winter periods from 2001 to 2019 were selected. The land-use types of the images were classified by support vector machine to acquire the landscape pattern index. In addition, the Pearson test was adopted to examine the correlation between crane population dynamics and landscape pattern index during overwintering. Finally, the regression model of the crane population was obtained by multivariate linear regression analysis. The results showed that the landscape index had a delayed impact on the crane population, which was significantly correlated with the crane population in the following five years. This effect on different crane population was disparate. The number of siberian cranes and white-naped crane populations showed a downward trend, while that of hooded cranes and common crane populations showed an upward trend. Furthermore, the landscape index affects the crane population to a certain extent, but the landscape pattern has diverse effects on different crane populations. It is suggested that corresponding protection measures be taken for different crane populations to provide a better habitat for cranes in Poyang Lake. [ABSTRACT FROM AUTHOR]

• The monitoring and estimation method of Winter Wheat Soil Moisture Based on multi-source sensing data and CNNR model is studied in this paper. • The proposed CNNR model has the best prediction accuracy. With the decrease of the SNR of radar echo, CNNR always has higher R2 and RPD, which indicates that the CNNR model can better predict the change trend close to the real value. Therefore, CNNR model has the best noise robustness among the three models. Rapid and accurate estimation of soil moisture content (SMC) is an important part of precision agriculture, and it is also one of the key problems to be solved in field real-time monitoring and precision irrigation. Most of the existing studies are limited to SMC monitoring of bare soil, which can be obtained by optical remote sensing (e.g., multispectral, hyperspectral) or Synthetic Aperture Radar (SAR). However, for the soil covered by vegetation, such as farmland, there are some theoretical defects with only one of the measuring methods. Meanwhile, in order to break through the limitations of low spatial and temporal resolutions of satellite remote sensing, it is of great significance to study SMC retrieval based on multi-source remote sensing data for the near earth UAV remote sensing systems. Based on this, this paper, taking the winter wheat planting area in Guanzhong plain of China as the research area, combines the advantages of ultra-wideband (UWB) radar, and multispectral remote sensing data, to reduce the influences of vegetation coverage on the estimation accuracy. A one-dimensional regression convolution neural network model is constructed to realize the quantitative prediction and estimation of SMC in farmland. The carried out experiments show that the proposed CNNR model has a better performance than traditional SVR and GRNN models and the R2, RMSE and RPD are 0.7453, 0.0140 cm3/cm3 and 1.9246, respectively. After introducing NDVI, MSAVI and DVI vegetation indices generated from multispectral images, the accuracy of the three models increased significantly. Among the three models, the constructed CNNR model has the best performance, and its R2, RMSE and RPD reach 0.9168, 0.0089 cm3/cm3, and 3.0201. Furthermore, after adding different levels of Gaussian noise to the original radar echoes, the CNNR model constructed in this paper still has the highest prediction accuracy and the strongest noise robustness. [ABSTRACT FROM AUTHOR]

[Display omitted] • Soil salinity can be well reflected by vegetation and salinity indicators. • The conversion of spectral indicators can effectively reduce the upscaling error. • Precise land regionalization will improve the accuracy of scale conversion. • The risk of soil salinization gradually decreases from coastal to inland. • Soil salinization experienced a change of increasing first and then decreasing. As a global problem of soil degradation, salinization has become a major obstacle to the sustainable development of the ecological environment and agriculture in coastal plains. However, the traditional process of salinity survey is too cumbersome, expensive and time-consuming to meet the mapping needs in a large scale. Remote sensing technology has become an important tool for digital soil mapping because of its rich sources, real-time and low cost. In order to meet the objective demand for rapid, accurate, and efficient acquisition and monitoring of soil salinization. This paper collected 61 soil samples from the Kenli District (experimental area) and extracted vegetation and salinity indicators from the Landsat image to construct the salinity inversion model by random forest algorithm. Then, taking the Yellow River Delta as the study area, the conversion coefficient of spectral indicators between Landsat and MODIS images was constructed in the form of the ratio of the mean value. Through optimization, the upscaling conversion method based on land use regionalization was proposed to realize the upscaling inversion and dynamic monitoring of soil salinization. The results showed that: (1) The random forest model based on NDVI , RVI , EVI , SI3 , and SI5 can better predict the soil salinity in the experimental area, with R2 = 0.821 and RMSE = 2.811 (validation accuracy). (2) The upscaling conversion method based on land use regionalization can effectively reduce the statistical error and collinearity of spectral indicators constructed by MODIS images and improve their correlation with OLI data and soil salinity. (3) From coastal to inland, soil salinization gradually decreases in the Yellow River Delta. From 2000 to 2020, soil salinization increased first and then decreased, and the salinized soil accounted for 20.35%∼35.10%. This study used multi-source remote sensing data to realize the collaborative inversion at different scales, which was significant for the quantitative estimation of soil salinity, salinization control, and sustainable agricultural development in coastal plains. [ABSTRACT FROM AUTHOR]

Evapotranspiration (ET) plays a vital role in surface energy balance and water management. Paddy rice is one of the most water-intensive crops and a staple food worldwide; thus, paddy field ET estimation is an impactful task. In this paper, an improved Surface Energy Balance Algorithm for Land (SEBALR) model is proposed to estimate the paddy field ET and is validated in different climate zones; it considers the advection effect and its applicable condition at the pixel scale and requires no ground-measured data. SEBALR contains two core modules: rice growth period detection and ET retrieval. SEBALR accurately monitored the planting and harvest dates, with an average error of less than 9 days, except for a paddy field located in the US, which had adopted dry seeding. The comparison between the estimation results and eddy covariance data shows that SEBALR can provide a more precise ET estimation than the original Surface Energy Balance Algorithm for Land, yielding a root mean squared error (RMSE) of 1.02 mm·d-1, mean relative error (MRE) of 22.97% and Pearson's correlation (R2) of 0.790. SEBALR successfully monitored paddy field ET from 2001 to 2019 in Nanchang City (NC), China. Yearly ET did not show an apparent change in trend from 2001 to 2019 in NC. SEBALR can be used to monitor the paddy rice growth period and estimate paddy field ET at a regional scale with limited data, providing helpful information for agricultural practices and water management. ● An improved SEBAL was proposed to quantify the ET advection effect of paddy fields. ● SEBALR can provide rice growth period information and more accurate ET. ● SEBALR can be applied on a large scale for its regional applicability and data independence. ● Paddy fields ET and its driving factors were analyzed from 2001 to 2019 in Nanchang, China. ● The structure and uncertainty of SEBALR were evaluated. [ABSTRACT FROM AUTHOR]

With the increasing problem of eutrophication, lacustrine ecosystem can undergo complex changes, often resulting in a shift from a clear macrophyte-dominated state to a turbid phytoplankton-dominated state. However, it's not clear how lake transitions occur at regional and global scales. This is due to a lack of long-term monitoring and an effective algorithm that can distinguish floating/emergent aquatic vegetation (FEAV), submerged aquatic vegetation (SAV) and algal bloom (AB). This paper presents a novel three-step classification algorithm based on Landsat imagery, (ie., vegetation and bloom indices (VBI) algorithm): (1) to distinguish between aquatic vegetation (AV) and non-aquatic vegetation (non-AV) extents by an aquatic vegetation index (AVI) derived from tasseled cap transformation; (2) to identify FEAV and SAV within AV extent by using the normalized difference vegetation index (NDVI), and (3) to extract AB from non-AV extent with floating algae index (FAI). The performance of the VBI algorithm was validated using 1307 field samples collected across 22 lakes in the middle and lower reaches of the Yangtze River (MLY) in China, with a pooled overall accuracy of 84.49%. Finally, the VBI algorithm was applied to Landsat data from 1985 to 2021 to quantify the temporal changes of FEAV, SAV and AB in large lakes (i.e., > 50 km2) along the MLY. Within these lakes, AV showed a significant decrease over the past 37 years, mainly due to the decrease of SAV; while AB occurred with higher frequency and in more lakes. The transition from a macrophyte-dominated state to a phytoplankton-dominated aquatic systems is still ongoing in the MLY area. This study shows outstanding innovation in the construction of a new classification index (i.e., AVI), and the method of calculating the adaptive threshold, which makes VBI algorithm have great potential in mapping long-term changes and capturing transformation between lake communities on a global scale. • A novel Landsat-based algorithm was developed for quantifying aquatic communities. • An index for aquatic vegetation (AV) was proposed to classify AV and non-AV. • Long-term algal blooms and AV (floating/emergent/submerged) in MLY lakes were mapped. • AV decreased while algal blooms increased in most MLY lakes from 1985 to 2021. • Shift from macrophyte- to phytoplankton-dominated state is ongoing in MLY lakes. [ABSTRACT FROM AUTHOR]

• Created a refined water resource allocation model. • Combining grid idea and remote sensing technology. • It reflects the change of water supply and demand in space and numerical value. • PSO algorithm is introduced to reduce the difficulty in solving the model. • It changes the extensive consideration in water supply nodes and water use nodes. The traditional optimal allocation of water resources mostly focuses on the allocation of different water supply sources to various types of water users in life, production, and ecology, and cannot reflect the differences in time and space of a certain type of water users. Therefore, this paper constructs a refined model for the optimal allocation of water resources, and further explores the optimal allocation of water resources with multiple water sources, multiple users, and multiple objectives. The main steps of the model construction are: (1) Regional gridding. The planting structure map and land use type map of the same resolution in the study area are selected and superimposed on them, the application attributes of each cell can be directly obtained. (2) Water demand gridding. On the basis of gridding in the study area, according to the application attributes of each cell, the quota method is used to predict the water demand of each cell. Then the water demand forecast result for the entire research area is obtained. (3) Water supply gridding. It is possible to reflect the types of water supply sources that a certain or a certain type of cell can accept, as well as the types of water supply sources that can be supplied by a certain type of water supply sources, by determining the types of water supply sources in the research area. The supply ranges that various water supply sources can reach, followed by the completion of the water supply grid. (4) According to the results of water demand gridding and water supply gridding, the water demand of each cell and the corresponding water supply source and water supply topological relationship are established, and the objective function is to minimize water shortage and maximize economic benefits to build a refined model for the optimal allocation of water resources.(5) The particle swarm algorithm is introduced to solve it, and the Pareto optimal solutions is obtained to determine the optimal water distribution for each cell. (6) Taking Guantao County, China as an example, the model was applied. In terms of minimum water shortage, the total water demand in Guantao County is predicted to be 118.8719 million m3 in the planning year (P = 50 %), while the total water supply is 125.2818 million m3. The water supply can meet the demand for water. In terms of financial advantages, the planned level of water Annual economic benefits will amount to 4.0865 million yuan. [ABSTRACT FROM AUTHOR]

This paper, firstly, investigates the relationship between clean energy and economic growth using satellite remote sensing data to proxy economic activities. Then, this paper constructs a spatial autoregressive model, spatial error model, and spatial Durbin model to study the spatial spillover effects of financial development and clean energy. Further, this paper adopts the difference-in-differences (DID) model and the propensity score matching-DID (PSM-DID) method to conduct quasi-natural experiments on clean energy and examine whether China's energy policy can promote clean energy, thereby promoting economic growth. The analysis results reveal a positive correlation between clean energy and economic growth; local financial development promotes clean energy in the region but negatively impacts the clean energy in surrounding areas. Moreover, the findings confirm a significant positive relationship between the "policy treatment effect" and clean energy. Finally, we put forward relevant policy recommendations based on the empirical results. • Satellite remote sensing data acquired by DMSP/OLS is used to proxy GDP. • Clean energy has a significant role in promoting economic growth. • Spatial econometric models are constructed to study the spatial spillover effects. • The DID model and PSM-DID method are used to conduct quasi-natural experiments. • China's energy policy has a significant promoting effect on clean energy and GDP. [ABSTRACT FROM AUTHOR]

• A detection approach for late-autumn shoots of litchi tree based on UAV remote sensing is provided. • A remote sensing dataset of late-autumn shoots of litchi tree is constructed. • A deep learning model based on the combination of YOLOv5 and Swin-transformer is proposed. • The detection accuracy of late-autumn shoots in remote sensing images is improved. • It will provide technical support for other researchers on late-autumn shoots and orchards manager. Litchi is one of the most common economic fruits in southern China, however, the growth of late-autumn shoots of litchi hinders flower bud differentiation and reduces yield of fruit. The early identification of the late-autumn shoots is of great significance for orchard management to control shoots and then increase fruit yield. At present, the identification of late-autumn shoots still relies on manual methods, which is not suitable for smart orchard management in a large area due to low recognition efficiency and high subjectivity. Therefore, a convenient, fast and cost-effective method is urgently needed. In response to this problem, the paper proposes a method based on the combination of unmanned aerial vehicle (UAV) remote sensing and object detection algorithm to detect late-autumn shoots. For this purpose, a remote sensing dataset of late-autumn shoots of litchi is first constructed by UAV. An improved YOLOv5 algorithm called YOLOv5-SBiC is then developed for late-autumn shoots identification. In the YOLOv5-SBiC algorithm, the transformer module is introduced to speed up the convergence of the network and improve detection accuracy, the attention mechanism module is employed to help the model extracting details, and BiFPN is used to better solve the multi-scale problem in detecting and then improve the recognition effect of small-sized objects. In addition, CIOU is selected as the loss function of bounding boxes regression to achieve high-precision localization of the boxes. The test results demonstrate that the recognition accuracy of YOLOv5-SBiC reaches a relatively high value of 79.6%, which is 4.0% higher than that (75.6%) of the original YOLOv5 algorithm and 15.9% higher than that (63.7%) of the pure transformer algorithm. It's also demonstrated that YOLOv5-SBiC is more competitive than the mainstream target detection algorithms in the dataset of late-autumn shoots. [ABSTRACT FROM AUTHOR]

Underground coal mining inevitably causes ground subsidence, especially in the coal-grain composite area with a high groundwater table (HGT) in eastern China, where mining subsidence causes waterlogging of arable land and threatens regional food security. Concurrent Mining and Reclamation (CMR) of underground coal mines emphasizes the implementation of reclamation measures in the subsidence process, so it is necessary to simulate and analyze dynamic subsidence to optimize reclamation planning. In this paper, the Guqiao coal mine, one of the HGT coal mining areas in eastern China, is used as a research area. First, the New Water Index (NWI) method and Otsu image segmentation algorithm were adopted to quantitatively analyze the spatial distribution and expansion of surface ponding caused by underground coal mining from 2008 to 2020 based on Landsat remote sensing images. Second, Fast Lagrangian Analysis of Continua in three Dimensions (FLAC3D) was used to invert the law of surface subsidence and the evolution characteristics of surface water caused by different mining schemes. Finally, each simulated mining scheme's reclamation timing, earthwork allocation, and reclamation rate were compared. The results show that: (1) the waterlogging area increases yearly with coal mining, reaching 1049 hm2 in 2020; (2) compared with the traditional reclamation (TR) rate of 32%, the reclamation rate of CMR could be increased to 65%; (3) the best mining scheme for CMR was up-warding skip mining with a reclamation duration of three years in this case. Optimization of mining schemes could control the temporal and spatial distribution of surface subsidence and determine the land reclamation plan. Meanwhile, the land use planning feedback could guide the mining scheme's design and realize sustainable land use during the mining life cycle in other similar areas. [Display omitted] • A method of mine land reclamation coupled with underground mining was proposed. • The influence of mining sequence on temporal and spatial development of surface subsidence was analyzed. • The reclamation rate of concurrent mining and reclamation (CMR) increased from 32% of traditional reclamation to 65%. • CMR promotes the sustainable use of farmland in coal-grain composite regions. [ABSTRACT FROM AUTHOR]

• LRSMS based on dynamic model library is designed and implemented. • DML with three modules is developed and used for locust monitoring. • Two cases form China are tested and well verify the advantages of the LRSMS. Locusts (insect pest) often cause serious loss to crops world-wide. Monitoring the occurrence of locust plagues through remote sensing technology has become a common method. However, with the abundance of remote sensing data resources, the corresponding locust remote sensing monitoring models are also diversified, and the selection of model parameters are more complex. How to manage these models scientifically for better reuse is an important problem that needs to be addressed. In order to solve these problems, we designed and implemented a locust remote sensing monitoring system (LRSMS) based on dynamic model library (DML), which is composed of model file library, model dictionary library, and model library management module. In terms of data to match the remote sensing monitoring models, there is more integration of locust habitat data in LRSMS, such as plantation, soil, which are closely related to the locust plague breeding environment. In addition, to improve the scalability of the system, a data interface is designed for the interaction between the DML and the Structured Query Language (SQL) server database. At the end of the paper, two application cases from China are given, and the results show that the LRSMS can well integrate multi-source data sets and carry out locust plague monitoring research through the dynamic model library. The approaches and methodologies presented in this paper can serve as a reference for those who are interested in remote sensing monitoring applications. [ABSTRACT FROM AUTHOR]

• DPSIRM model was used to evaluate the ecological security of Xilingol grassland. • Remote sensing, meteorology, socio-economic and survey data are used in this paper. • We use grid GIS method to obtain more accurate evaluation results. • We analyzed the spatial-temporal evolution characteristics of ecological security. • Changes in the center of gravity in different security levels are also explored. The quantitative evaluation of the ecological security can reflect the changes in the quality of ecological environment and provide a scientific basis for the formulation of land-use policy and for the construction of the ecological environment. In this study, we selected Xilingol League in Inner Mongolia, China, as the research area, and established the driving force-pressure-state-impact-response-management (DPSIRM) model to evaluate grassland ecological security. The results were as follows: (1) During 2005–2015, the overall trend of ecological security level in the study area increased, with the western region being relatively ecologically insecure and the eastern region being more secure. (2) The spatial distribution trend of ecological security level from west to east was " very serious alerts " → " serious alerts " → " moderate alerts " → " light alerts "→ " no alert ". (3) The " very serious alerts " and " serious alerts " are mainly concentrated in Sonid Left Banner and Sonid Right Banner, but from the perspective of spatial migration of ecological security, unsafe and less safe tend to shift to the northeast. This study provides a new idea for grassland ecological security evaluation research, has important practical significance for understanding the ecological security level of Xilingol grassland area, and will guide the reasonable development and utilization of Xilingol grassland. [ABSTRACT FROM AUTHOR]

• Both GPP and GPP max in Chinese grasslands showed increases at different space–time scales. • Trend analysis and breakpoint tests were used for the first time to monitor GPP change simultaneously in Chinese grasslands. • GPP max × CUP is a robust indicator to explain annual-scale variation in GPP. At the ecosystem level, Gross Primary Productivity (GPP) was defined as the organic compounds formed by plants that absorb atmospheric CO 2 through photosynthesis and sequester carbon in plant bodies. How to accurately assess the spatiotemporal evolution of ecosystem carbon indicators has become a critical issue to be addressed. Chinese grasslands are located in a typical arid and semi-arid climate zone and are sensitive to global changes, which will inevitably have serious impacts on the function and structure of Chinese grasslands. Based on this, our paper takes Chinese grassland ecosystems as the research area, and used multi-source GPP dataset from terrestrial ecosystem model simulations and remote sensing satellite observations to quantitatively analyze the spatiotemporal evolution patterns of GPP in Chinese grasslands over the past 40 years. We combined trend analysis and breakpoint test, and then to analyze the mechanism components of GPP space–time variability. The main results are as follows: (1) The model and remote sensing estimated GPP results showed that more than 80% of Chinese grasslands show a significant increasing trend over the past 40 years, with growth rates ranging from 0.68 to 3.13 g C/m−2 year−1. (2) GPP max also shows that more than 80% of Chinese grasslands are growing rapidly, with an overall growth rate of more than 0.1 g C/m−2 year−1 in each region. The overall long-term trends and interannual variability of multi-source GPP and GPP max are generally consistent, yet vegetation dynamics in local areas are still uncertain. (3) The breakpoint test showed that 'monotonically increasing' was the largest breakpoint type of GPP in Chinese grasslands (33.09%), and the direction of change of GPP in Chinese grasslands before and after the breakpoint was also increasing. (4) GPP max × CUP explained 91% of the temporal variability of annual-scale GPP in Chinese grasslands from a mechanistic view, and we found that the peak photosynthetic growth and the length of the phenological period synergistically controlled the interannual variability of GPP in Chinese grasslands. Under the scenario of rapid global change, our study accurately assessed the long-term dynamics of GPP and its mechanism-driven of grassland ecosystems in China, which is helpful for estimating and predicting the carbon budget of grassland ecosystems in China, and has important guiding significance for policy formulation to mitigate climate change. [ABSTRACT FROM AUTHOR]

• Our method has the potential to extract water with high accuracy on a large scale. • Our method realizes the conversion from unlabeled learning to noisy label learning. • The model performance is significantly higher than traditional methods. • Our method provides a new idea for unsupervised learning. Large-scale monitoring of surface water bodies is of great significance to the sustainable development of regional ecosystems. Remote sensing is currently the main method of global earth observation. On the one hand, the traditional water index is simple and efficient, but it relies on a fixed global threshold, which leads to low accuracy for water extraction. On the other hand, deep learning has achieved state-of-the-art results in classification of spectral data, but it consumes a substantial amount of manpower and time to label sufficient high-quality samples. Furthermore, spectral characteristics of water bodies in different areas vary greatly due to changes in atmospheric conditions and viewing geometry. In this paper, we propose a new accurate water extraction framework based on unsupervised deep learning and NDWI of multispectral images. Binarized NDWI images are used to identify potential water bodies, and deep learning training is performed using these pseudo samples and labels. This process realizes the conversion from unlabeled learning to noisy label learning. First, we proposed a simple and fast binarization algorithm to segment as many real water bodies as possible from NDWI images. Then a set of water confidence assessment rules was constructed from the four aspects of the spectrum, shape, agglomeration, and range. The water segments were scored and sorted to make the model start learning from easy samples and gradually advance to complex samples. Finally, the adjusted co-teaching learning strategy is adopted to filter errors introduced by noisy labels through peer networks with different learning capabilities during training. We tested the accuracy of our method using Gaofen Image Dataset (GID), Sentinel-2 and Landsat images of several water bodies in China. Compared with other methods, our method improved the F1 score by 18.1–40.3% and 6.8–22.2% for GID and Sentinel-2 images, respectively. In addition, our method is more stable in long-term water monitoring. The proposed method has the potential to be used for extracting water bodies with high accuracy on a large scale, especially in areas with complex environments and a lack of deep learning samples. And it provides a new idea for unsupervised learning in the current remote sensing field by fully combining remote sensing expertise and spectral information of ground objects. [ABSTRACT FROM AUTHOR]

• A UAV-based method estimating sediment deposited behind check dams was developed. • Errors of the optimal model for single and regional check dams are 12–13% and 2–3%. • This method avoids limitations of tedious field surveys and historical records. • This method can evaluate sediment retained by check dams in areas as large as 211,800 km2. The estimation of sediment deposited behind dams and reservoirs is of great significance for quantifying soil erosion rate and sediment flux. However, existing methods of sediment volume estimation often require detailed field measurements and historical topographic data, limiting the broader application of these methods. In this paper, focusing on the region that suffers the world's most severe soil erosion, we combine unmanned aerial vehicle (UAV) photogrammetry and simulate submerging analysis to propose a novel method for estimating sediment silted by check dams under complex topography. We obtained 1339 groups of topographic factors and ascertained the corresponding volumes of sediment deposited behind check dams in the loess hilly region of China (LHRC), and established five different models combined with regression analysis. Two different sets of data were used for method validation and optimal model determination. The results showed that the error of the optimal model in the volume estimation of single check dam and regional check dams is 12–13% and 2–3%, respectively. Additionally, the area-volume model has the potential to evaluate the sediment retention capacity of check dams (in the order of billions of cubic meters) in the whole LHRC, because the variables are easy to obtain and the model accuracy is relatively high. The further application of this method will help to evaluate the watershed sediment yield and regional soil and water conservation benefits, and is of great significance for explaining the future changes in water and sediment in the Yellow River basin. [ABSTRACT FROM AUTHOR]

• A more effective extraction method of time factor for soil type mapping based on remote sensing images was determined. • The method of principal component extraction result based on the tasseled cap transformation was more effective. • The fusion of multi-temporal features and texture features could improve the accuracies of soil type mapping. • The best texture feature for soil mapping was correlation parameter, which produced the largest overall accuracy of 86.81%. With the rapid development of remote sensing (RS) technology, remote sensing images provide an important data basis for soil type mapping. In remote sensing images, temporal factor is difficult to obtain directly, and the rich geometric features are not used adequately. Multi-temporal remote sensing data could effectively reflect the temporal variation of ground objects, while how to extract multi-temporal image features more effectively for soil type interpretation needs to be studied. Moreover, it is not clear whether multi-temporal features and texture features can be effectively integrated to improve mapping accuracies. Therefore, taking five soil types of Laoshan County, Shandong Peninsula, China as the subject investigated and six remote sensing images as data sources, this paper explored and compared two extraction methods of multi-temporal features from remote sensing images. The effects of the eight different texture features fused the multi-temporal features on digital soil mapping were also analyzed. The results showed that the principal component extraction result based on the tasseled cap transformation was better than based on the spectral band synthesis, increasing the overall accuracy by 3.83–11.41% and the kappa index by 0.03–0.13. The fusion of multi-temporal features and texture features can effectively improve accuracies of soil type mapping. After the addition of correlation texture feature parameter, the overall accuracy (86.81%) and Kappa index (0.81) were increased by 11.92% and 0.16, respectively. The research results showed that multi-temporal features in remote sensing images had great advantages in digital soil mapping, and the effective fusion with texture features provided a new idea for improving the accuracy of digital soil mapping. [ABSTRACT FROM AUTHOR]

Land surface temperature (LST) is a key parameter when monitoring land surface processes. However, cloud contamination and the tradeoff between the spatial and temporal resolutions greatly impede the access to high-quality thermal infrared (TIR) remote sensing data. Despite the massive efforts made to solve these dilemmas, it is still difficult to generate LST estimates with concurrent spatial completeness and a high spatio-temporal resolution. Land surface models (LSMs) can be used to simulate gapless LST with a high temporal resolution, but this usually comes with a low spatial resolution. In this paper, we present an integrated temperature fusion framework for satellite-observed and LSM-simulated LST data to map gapless LST at a 60-m spatial resolution and half-hourly temporal resolution. The global linear model (GloLM) model and the diurnal land surface temperature cycle (DTC) model are respectively performed as preprocessing steps for sensor and temporal normalization between the different LST data. The Landsat LST, Moderate Resolution Imaging Spectroradiometer (MODIS) LST, and Community Land Model Version 5.0 (CLM 5.0)-simulated LST are then fused using a filter-based spatio-temporal integrated fusion model. Evaluations were implemented in an urban-dominated region (the city of Wuhan in China) and a natural-dominated region (the Heihe River Basin in China), in terms of accuracy, spatial variability, and diurnal temporal dynamics. Results indicate that the fused LST under all-weather conditions is highly consistent with actual Landsat LST data (in situ LST measurements), in terms of a Pearson correlation coefficient of 0.94 (0.96–0.99), a mean absolute error of 0.71–0.98 K (0.82–3.34 K), and a root-mean-square error of 0.97–1.26 K (1.09–4.36 K). The generated diurnal Landsat-like LSTs under all weather conditions are able to support diurnal dynamic studies that are the most relevant to human activities, such as the study of urban heat islands (UHIs) and water resource management at the field scale. • Gapless LSTs with a high spatio-temporal resolution were generated. • The first time to integrally fuse multi-source satellite-observed and model-simulated data. • Improvement in the generalization ability of spatio-temporal LST fusion. • Highly accurate fused results under all-weather conditions based on extensive validation. [ABSTRACT FROM AUTHOR]

Cities are diverse, with their formation and expansion having various driving forces. Xinjiang Production and Construction Corps (XPCC) cities represent an emerging category in China's urban system. They have evolved into a special organisation integrating 'party, government, military, and enterprises' and are located in Xinjiang, the frontier region of north-western China. Investigation of such urban characteristics is helpful to enrich our diverse cognitions on cities. The purpose of this paper is to trace the growth history of XPCC cities and explore their evolution-related characteristics, modes, and driving mechanisms in terms of spatial morphology and landscapes in the process of urban expansion by utilising remote sensing interpretation data. This study led to four main conclusions. (1) The role of the XPCC in maintaining the long-term stability and prosperity of the frontier is increasingly valued across the nation, thus leading to a major increase in the quantity and quality of XPCC cities. (2) These cities have a unique spatial layout around two deserts and the border characterised by 'two rings and one line'; however, their distribution is relatively scattered. (3) The expansion of XPCC cities has become prominent along with the restructuring of land use, with wasteland reclaiming being a salient attribute of their built-up land expansion, while also protecting cultivated land. Most cities reveal a more complex and less compact morphology. (4) The external government's promotion and regulation, the basic shaping of the natural geographical environment, the internal boosting of economic increase and industrial restructuring, and the special reshaping of the XPCC-local relationship are key driving factors behind the urban temporal-spatial evolution. Overall, this study supplements current knowledge on frontier cities within the Chinese context and thus helps to improve understanding of the urban theoretical research of government behaviours and the multi-dynamics of urban development. [ABSTRACT FROM AUTHOR]

River confluences are key nodes in fluvial systems. Past studies on large river confluences were focused on the Paranà, Amazon, and Yangtze rivers, while the Yarlung Zangbo River (YZR), the longest plateau river in China, has not been investigated yet. This paper investigates a large multi-channel confluence between the YZR and one of its major tributaries, the anabranching Niyang River (NR). A remote-sensing study of this area demonstrated that the confluence planform is very stable compared with its upstream and downstream reaches that suffered from sandbar migration, meandering, and the development of a head-cut. A field survey using Acoustic Current Doppler Profiling (ADCP) was conducted at the confluence in June 2019, when the YZR and NR had discharges of about 1393 m3/s and 1299 m3/s in total, respectively. Although the discharge ratio of this confluence attained a value of 0.93, the existence of an estuary delta with a large area of floodplain largely weakened the impact of the NR on the dynamics of the YZR. The inflows of the multiple channels of the NR separated by the delta created an evident and long mixing interface and produced the left-side secondary cells in the YZR. In contrast, the curved riverbank played a more important role in the formation of deep scour holes, generating heterogeneity of the velocity distribution and the fast-mixing of flows. This study leads to an advancement in our current knowledge about multi-channel confluences. • Remote sensing images show that the plateau river confluence planform is very stable. • A field survey on flow and morphological features of the confluence was conducted. • The estuary delta with large floodplain largely weakened the confluence dynamics. • The curved river bank played an important role in river bed scours and flow mixing. [ABSTRACT FROM AUTHOR]

• The remote sensing environment index and landscape adaptability index were extracted. • Ecological vulnerability (EV) was evaluated using a self-organizing map model. • The factors influencing EV were analysed by GeoDetector. • EV in Badong County is high in the north and south, low in the middle. • Xinling and Dongdukou appear to be the most ecologically vulnerable townships. As a result of rapid economic development, ecological environmental problems have become increasingly serious. Ecological vulnerability (EV) assessment can scientifically analyse the causal mechanism and change laws of ecosystem vulnerability, providing a basis for decision-making for ecological environmental protection. Based on the pressure-state-response (PSR) model, this paper selects 13 indicators, including the geological hazard index, soil erosion sensitivity, population density, proportion of developed land, remote sensing ecological index, and landscape adaptability index, to construct an EV evaluation index system for Badong County, China. Furthermore, the analytic hierarchy process (AHP)-coefficient of variation (CV) method and the self-organizing map (SOM) model are used to comprehensively and quantitatively evaluate the EV of Badong County in 2020. The main factors affecting the EV are explored through GeoDetector. The results show that the EV in Badong County is at a moderate level and above; it is high in the north and south and low in between. According to the analysis of the EV results, the 12 townships in Badong County can be divided into 3 classes. Class I includes Xinling and Dongdukou towns, where the EV is high. Ecological governance measures should be strengthened in these two towns. Based on the GeoDetector results, the spatial distribution of EV in Badong County is mainly related to five factors: the geological hazard index, habitat quality index, proportion of developed land, remote sensing ecological environment index, and soil erosion sensitivity. [ABSTRACT FROM AUTHOR]

[Display omitted] • SIF data that characterizes the growth of sandy vegetation is more sensitive. • SIF can well capture the changes in surface moisture and groundwater. • Vegetation growth has a strong correspondence with SM and a relatively weak correlation with TWS. • Vegetation growth in the late period of the mutation year is generally better than that in the early period. • Similarities and differences of vegetation-moisture relationship in four sandy lands. Hulunbuir, Otindag, Horqin and Mu Us are the four greater sandy lands in China. They are typical ecologically fragile areas with limited ability to resist interference and are susceptible to changes in the external environment. With climate change in recent decades, these areas have experienced severe land desertification and decreased groundwater levels. The ecological environment is at risk of further deterioration, and continuous vegetation restoration and management should be adapted to the local climate and maintain the water balance with natural precipitation. This paper combines multi-source satellite observations to explore the relationship between vegetation growth and water supply. The results show four main points. (1) Although the vegetation growth in Otindag and Horqin had degraded before 2010, it improved after 2010. The vegetation growth and soil moisture (SM) of the four sandy lands both generally increased from 2003 to 2016. The SM of the Horqin increased significantly (84% anomaly > 1.4 × 10−2 cm3/cm3). Other than the Hulunbuir, the other sandy lands show a downward trend in terrestrial water storage (TWS), especially the Mu Us declined the most (97% anomaly < −40 mm). (2) Solar-induced chlorophyll fluorescence (SIF) has specific advantages in characterizing the sandy vegetation productivity and water composition changes compared to vegetation index (EVI) or gross primary productivity (GPP). (3) The regional pattern of the vegetation-moisture relationship is mainly affected by the precipitation-driven SM and depends on the types of underlying vegetation functions. (4) Both SM and TWS have approximately one-month persistence in the four sandy lands, especially the Mu Us has the longest persistence (0.94 months and 1.1 months). The persistence in the early period is generally longer than that in the late period, implying that the resilience of water components after the mutation year is stronger than before it. This study provides a scientific basis for the management and restoration of sandy lands. It is recommended that different vegetation restoration and protection strategies should be developed for specific sandy land moisture conditions in arid/semi-arid regions of northern China, which includes the improvement of water use efficiency to maintain soil moisture and the reduction of excessive groundwater uses. [ABSTRACT FROM AUTHOR]

• Long-term monitoring of riparian condition based on RS method were achieved. • We proposed a new multi-scale evaluation strategy for riparian restoration. • Ecological partitioning was useful to identify fragile areas and influence factors. • The riparian zone with a higher PSI index did not indicate a better restoration effect. • Indicators suitable for plain rivers need further development for wider application. Quantitative evaluation of ecological restoration effects (EREs) is essential to provide operational feedback and implement adaptive management for future riparian restoration. This manuscript aims to develop a multi-scale evaluation method of EREs based on remote sensing observations, which uses the composite index of physical structural integrity (PSI) to indicate riparian condition. The riparian EREs were quantitatively assessed by comparing the post-restoration PSI values with the ecological reference status derived from pre-restoration surveys. In addition, ecological partitioning was adopted to identify the vulnerable riparian areas and explore the influence indicators of ecological restoration. In this paper, we used the case study of the Liaohe River, the first protected area aimed to recover the riparian ecosystem in China, to evaluate the riparian EREs at three spatial scales using Landsat time series images from 1980 to 2019. Results indicated that: (1) the pre-restoration riparian ecosystem exhibited an clear degradation trend and ecological conservation was urgent; (2) riparian restoration is a continuous and heterogeneous ecological process, the project had not yet met its objectives in 2016 with a low ERE value of 49.1 (Poor), while achieved a significant success in 2019 with the ERE value of 68.8 (Very good); (3) the limited restoration partitions with a ERE value below 60 accounted for 42.8% of riparian zones, which were the fragile riparian ecosystems and mainly distributed in reach D and F; and (4) the dominant factors for riparian restoration were canopy coverage, human disturbance and natural vegetation percent. This study is expected to provide a practical and feasible approach for evaluating the riparian EREs and supporting the future river management, in particular for rivers lack of long-term monitoring data. [ABSTRACT FROM AUTHOR]

[Display omitted] • The framework coupling mining intensity and socioecological resilience was constructed. • The resilience and mining intensity were analyzed for 547 counties in China. • 12 socioecological indicators were used to evaluate resilience in spatial scale. • The zoning results enhanced our understanding of sustainable development in mining areas. Coal is an important global energy resource. However, the negative impacts of mining are important factors restricting the sustainable development of mining areas. Under the superimposed influences of different mining intensities, different natural ground conditions, and social-economic development, there are large differences in the restoration characteristics and response strategies for mining. To guide the development and utilization of coal resources and ecological restoration, the spatial distribution should be appropriately analyzed at the macroscale. Based on the web-crawler technology and multi-source remote sensing data, this paper attempted to estimate the mining intensity and the social-ecological resilience respectively at the county level, to divide the mining areas into nine regions in China. Based on the zoning results, differentiated strategies were proposed for ecological restoration and resource development and utilization in the various coal mining regions. The valuation results suggested that coal mines are widely distributed and abundant in China, mainly concentrated in the central, southwest and northwest regions. However, nearly half of the coal mining regions are with low resilience. Especially, high mining intensity-low resilience areas are concentrated in north China which account for 10.83%. These areas should receive special attention to prevent possible ecological risks and minimize economic effects. Social-ecological resilience measures system adaptation to coal mining disturbances, which can be regarded as the standard of mining intensity adjustment. In the future, local governments can try to combine ecological restoration and resource development strategies to provide support for the sustainable development of coal mining areas in China. [ABSTRACT FROM AUTHOR]

Sewage outfalls are the ideal locations for water pollution monitoring as they are the final gate that pollutants must pass before entering rivers and oceans. Strengthening the supervision and control of industrial sewage outfalls has become the most important task in water pollution control. Most rivers in China are highly polluted, especially those flowing through cities. It is difficult to investigate, monitor, trace, and remediate sewage outfalls because information regarding the distribution of sewage outfalls into rivers is incomplete and the existing methods of monitoring sewage outfalls into rivers are outdated. This paper proposes an innovative method for feasible and efficient acquisition of sewage outfall data through a combination of web crawler and remote sensing interpretation technologies, employing data from the Luan River Basin as a testing site. We crawled industrial information and spatial location data on the Internet and obtained the locations of industrial sewage outfalls by interpreting remote sensing images. We modeled the distribution of sewage inflow from major industries into the tributaries of the Luan River Basin. We verified the accuracy and reliability of the developed method by comparing its results with actual data collected during our field investigation. Sample field tests confirmed the high accuracy (89%) of the proposed method, indicating its successful application in the Luan River Basin. We believe that the proposed method can be used to verify and supervise sewage outfalls in other locations in the future, especially owing to its improved acquisition system, which ensures feasible and efficient measurements without the need for manual investigation. This study provides a scientific basis for decision-making with regard to sustainable development. ∙Strengthen supervision of industry sewage outfalls help river pollution control ∙Web crawler combined with remote sensing interpretation were innovatively employed ∙Web crawler successfully designed to collect industry sewage outfalls for water pollution monitoring ∙Verified new method's accuracy (89%) and reliability by field investigation ∙Measurement driven by new method is applicable to other rivers and countries [ABSTRACT FROM AUTHOR]

• The induced factors of the Baige landslide are comprehensively investigated. • The Baige landslide is caused by deep-seated gravitational slope deformation. • River downcutting is a prerequisite for the evolution of the Baige landslide. • Earthquake contributes to accelerating the development of the Baige landslide. On October 11 and November 3, 2018, two successive landslides happened at the same location near Baige Village in Tibet of China and dammed the Jinsha River twice. The induced factors and formation mechanism of the Baige landslide are still being questioned. In this paper, high-resolution topographic data, multi-temporal remote sensing images (including the unmanned aerial vehicle (UAV) 3D image, remote sensing, and Geographical Information Systems (GIS) techniques), rainfall data, seismic records, comprehensive field investigations, and numerical simulations are used to identify the main characteristics, the induced factors, and the formation mechanism of the event. The results show that deep-seated gravitational slope deformations play an important role in the evolution process of the Baige landslide; river downcutting is a prerequisite for the progressive evolution of the gravitational slope deformation behavior of the Baige landslide, whereas earthquake plays a catalytic role in the development of the gravitational slope deformation of the Baige landslide. The Baige landslide would be a typical case study for understanding the formation mechanism of large-scale landslides along the Jinsha River. [ABSTRACT FROM AUTHOR]