Search

Understanding changes of soil organic carbon (SOC) in top layers of croplands and their driving factors is a vital prerequisite in decision-making for maintaining sustainable agriculture. However, high-precision estimation of SOC of croplands at regional scale is still an issue to be solved. Based on soil samples, synthetic image of bare soil and geographical data, this paper predicted SOC density of croplands using Random Forest model in the Black Soil Region of Jilin Province, China in 2005 and 2020, and analyzed its influencing factors. Results showed that random forest model that integrates bare soil composite images improve the accuracy and robustness of SOC density prediction. From 2005 to 2020, the total SOC storage in croplands decreased from 89.96 to 82.79 Tg C with an annual decrease of 0.48 Tg C yr−1. The mean value of SOC density of croplands decreased from 3.42 to 3.32 kg/m2, and high values are distributed in middle parts. Changes of SOC represented significant heterogeneity spatially. 62.14% of croplands with SOC density greater than 4.0 kg/m2 decreased significantly, and 38.60% of croplands with SOC density between 2.5 and 3.0 kg/m2 significantly increased. Climatic factors made great contributions to SOC density, however, their relative importance (RI) to SOC density decreased from 44.65% to 37.26% during the study period. Synthetic images of bare soil constituted 23.54% and 26.29% of RI in the SOC density prediction, respectively, and the contribution of each band was quite different. The RIs of topographic and vegetation factors were low but increased significantly from 2005 to 2020. This study can aid local land managers and governmental agencies in assessing carbon sequestration potential and carbon credits, thus contributing to the protection and sustainable use of black soils. [ABSTRACT FROM AUTHOR]

The accurate estimation of forest aboveground biomass is of great significance for forest management and carbon balance monitoring. Remote sensing instruments have been widely applied in forest parameters inversion with wide coverage and high spatiotemporal resolution. In this paper, the capability of different remote-sensed imagery was investigated, including multispectral images (GaoFen-6, Sentinel-2 and Landsat-8) and various SAR (Synthetic Aperture Radar) data (GaoFen-3, Sentinel-1, ALOS-2), in aboveground forest biomass estimation. In particular, based on the forest inventory data of Hangzhou in China, the Random Forest (RF), Convolutional Neural Network (CNN) and Convolutional Neural Networks Long Short-Term Memory Networks (CNN-LSTM) algorithms were deployed to construct the forest biomass estimation models, respectively. The estimate accuracies were evaluated under the different configurations of images and methods. The results show that for the SAR data, ALOS-2 has a higher biomass estimation accuracy than the GaoFen-3 and Sentinel-1. Moreover, the GaoFen-6 data is slightly worse than Sentinel-2 and Landsat-8 optical data in biomass estimation. In contrast with the single source, integrating multisource data can effectively enhance accuracy, with improvements ranging from 5% to 10%. The CNN-LSTM generally performs better than CNN and RF, regardless of the data used. The combination of CNN-LSTM and multisource data provided the best results in this case and can achieve the maximum R2 value of up to 0.74. It was found that the majority of the biomass values in the study area in 2018 ranged from 60 to 90 Mg/ha, with an average value of 64.20 Mg/ha. [ABSTRACT FROM AUTHOR]

Ponds constitute a pivotal component of aquatic ecosystems. The aquatic ecosystem of the Huai River Basin (HRB) in China was once damaged by severe pollution, and numerous ponds in the basin have not been secured. In this paper, Shenqiu County, a typical county in HRB with many ponds, is selected. Based on high-resolution images with ALOS in 2010, GF-2 in 2016, and GF-1 in 2022, we employed discriminant analysis (DA), classification and regression tree, support vector machine, and random forest to extract the ponds based on object-oriented and further analyzed the spatial-temporal variations of the ponds in this county. The results of the DA in these three years exhibited a higher kappa coefficient (>0.7), and overall accuracy (>75%), signifying superior performance when compared to the other three methods. There were 4625, 5315, and 4748 ponds in 2010, 2016, and 2022, with a total area of 12.87, 11.99, and 9.37 km2, respectively. The number of ponds had a trend of rising in the initial period (2010–2016) and falling later (2016–2022), while the total area revealed a continuous decline. Meanwhile, these ponds showed a clustering phenomenon with three main clustering areas, and the scope of the clustering areas also changed to a certain extent from 2010 to 2022. Our study offers valuable methodological support for the ecological monitoring and management of water environments in regions characterized by a dense concentration of ponds. The crucial data related to ponds in this study will help inform both environmental and social development initiatives within the area. [ABSTRACT FROM AUTHOR]

Remote-sensing images of high spatial resolution (HSR) are valuable sources of fine-grained spatial information for various applications, such as urban surveys and governance. There is continuing research on positional errors in remote-sensing images and their impacts in geoprocessing and applications. This paper explores the combined use of multi-point geostatistics (MPS), machine learning—in particular, generalized additive modeling (GAM)—and computer-image correlation for characterizing positional errors in images—in particular, HSR images. These methods are employed because of the merits of MPS in being flexible for non-parametric and joint simulation of positional errors in X and Y coordinates, the merits of GAM in being capable of handling non-stationarity in-positional errors through error de-trending, and the merits of computer-image correlation in being cost-effective in furnishing the training data (TD) required in MPS. Procedurally, image correlation is applied to identify homologous image points in reference-test image pairs to extract image displacements automatically in constructing TD. To cope with the complexity of urban scenes and the unavailability of truly orthorectified images, visual screening is performed to clean the raw displacement data to create quality-enhanced TD, while manual digitization is used to obtain reference sample data, including conditioning data (CD), for MPS and test data for performance evaluation. GAM is used to decompose CD and TD into trends and residuals. With CD and TD both de-trended, the direct sampling (DS) algorithm for MPS is applied to simulate residuals over a simulation grid (SG) at 80 m spatial resolution. With the realizations of residuals and, hence, positional errors generated in this way, the means, standard deviation, and cross correlation in bivariate positional errors at SG nodes are computed. The simulated error fields are also used to generate equal-probable realizations of vertices that define some road centerlines (RCLs), selected for this research through interpolation over the aforementioned simulated error fields, leading to error metrics for the RCLs and for the lengths of some RCL segments. The enhanced georectification of the RCLs is facilitated through error correction. A case study based in Shanghai municipality, China, was carried out, using HSR images as part of generalized point clouds that were developed. The experiment results confirmed that by using the proposed methods, spatially explicit positional-error metrics, including means, standard deviation, and cross correlation, can be quantified flexibly, with those in the selected RCLs and the lengths of some RCL segments derived easily through error propagation. The reference positions of these RCLs were obtained through error correction. The positional accuracy gains achieved by the proposed methods were found to be comparable with those achieved by conventional image georectification, in which the CD were used as image-georectification control data. The proposed methods are valuable not only for uncertainty-informed image geolocation and analysis, but also for integrated geoinformation processing. [ABSTRACT FROM AUTHOR]

The global navigation satellite system (GNSS) position time series provides essential data for geodynamic and geophysical studies. Interpolation of the GNSS position time series is necessary because missing data will produce inaccurate conclusions made from the studies. The spatio-temporal correlations between GNSS reference stations cannot be considered when using traditional interpolation methods. This paper examines the use of machine learning models to reflect the spatio-temporal correlation among GNSS reference stations. To form the machine learning problem, the time series to be interpolated are treated as output values, and the time series from the remaining GNSS reference stations are used as input data. Specifically, three machine learning algorithms (i.e., the gradient boosting decision tree (GBDT), eXtreme gradient boosting (XGBoost), and random forest (RF)) are utilized to perform interpolation with the time series data from five GNSS reference stations in North China. The results of the interpolation of discrete points indicate that the three machine learning models achieve similar interpolation precision in the Up component, which is 45% better than the traditional cubic spline interpolation precision. The results of the interpolation of continuous missing data indicate that seasonal oscillations caused by thermal expansion effects in summer significantly affect the interpolation precision. Meanwhile, we improved the interpolation precision of the three models by adding data from five stations which have high correlation with the initial five GNSS reference stations. The interpolated time series for the North, East, and Up (NEU) are examined by principal component analysis (PCA), and the results show that the GBDT and RF models perform interpolation better than the XGBoost model. [ABSTRACT FROM AUTHOR]

To improve the accuracy and reliability of precipitation estimation, numerous models based on machine learning technology have been developed for integrating data from multiple sources. However, little attention has been paid to extracting the spatiotemporal correlation patterns between satellite products and rain gauge observations during the merging process. This paper focuses on this issue by proposing an integrated framework to generate an accurate and reliable spatiotemporal estimation of precipitation. The proposed framework integrates Funk-Singular Value Decomposition (F-SVD) in the recommender system to achieve the accurate spatial distribution of precipitation based on the spatiotemporal interpolation of rain gauge observations and Convolutional Long Short-Term Memory (ConvLSTM) to merge precipitation data from interpolation results and satellite observation through exploiting the spatiotemporal correlation pattern between them. The framework (FS-ConvLSTM) is utilized to obtain hourly precipitation merging data with a resolution of 0.1° in Jianxi Basin, southeast of China, from both rain gauge data and Global Precipitation Measurement (GPM) from 2006 to 2018. The LSTM and Inverse Distance Weighting (IDW) are constructed for comparison purposes. The results demonstrate that the framework could not only provide more accurate precipitation distribution but also achieve better stability and reliability. Compared with other models, it performs better in variation process description and rainfall capture capability, and the root mean square error (RSME) and probability of detection (POD) are improved by 63.6% and 22.9% from the original GPM, respectively. In addition, the merged precipitation combines the strength of different data while mitigating their weaknesses and has good agreement with observed precipitation in terms of magnitude and spatial distribution. Consequently, the proposed framework provides a valuable tool to improve the accuracy of precipitation estimation, which can have important implications for water resource management and natural disaster preparedness. [ABSTRACT FROM AUTHOR]

Using the Gravity Recovery and Climate Experiment (GRACE) satellite to monitor groundwater storage (GWS) anomalies (GWSAs) at the local scale is difficult due to the low spatial resolution of GRACE. Many attempts have been made to downscale GRACE-based GWSAs to a finer resolution using statistical downscaling approaches. However, the time-lag effect of GWSAs relative to environmental variables and optimal model parameters is always ignored, making it challenging to achieve good spatial downscaling, especially for arid regions with longer groundwater infiltration paths. In this paper, we present a novel spatial downscaling method for constructing GRACE-based 1 km-resolution GWSAs by using the back propagation neural network (BPNN) and considering the time-lag effect and the number of hidden neurons in the model. The method was validated in Alxa League, China. The results show that a good simulation performance was achieved by adopting varying lag times (from 0 to 4 months) for the environmental variables and 14 hidden neurons for all the networks, with a mean correlation coefficient (CC) of 0.81 and a mean root-mean-square error (RMSE) of 0.70 cm for each month from April 2002 to December 2020. The downscaled GWSAs were highly consistent with the original data in terms of long-term temporal variations (the decline rate of the GWSAs was about −0.40 ± 0.01 cm/year) and spatial distribution. This study provides a feasible approach for downscaling GRACE data to 1 km resolution in arid regions, thereby assisting with the sustainable management and conservation of groundwater resources at different scales. [ABSTRACT FROM AUTHOR]

Analysis of the spatial and temporal variation patterns of surface evapotranspiration is important for understanding global climate change, promoting scientific deployment of regional water resources, and improving crop yield and water productivity. Based on Landsat 8 OIL_TIRS data and remote sensing image data of the lower Yangtze River urban cluster for the same period of 2016–2021, combined with soil and meteorological data of the study area, this paper constructed a multiple linear regression (MLR) model and an extreme learning machine (ELM) inversion model with evapotranspiration as the target and, based on the model inversion, quantitatively and qualitatively analyzed the spatial and temporal variability in surface evapotranspiration in the study area in the past six years. The results show that both models based on feature factors and spectral indices obtained a good inversion accuracy, with the fusion of feature factors effectively improving the inversion ability of the model for ET. The best model for ET in 2016, 2017, and 2021 was MLR, with an R2 greater than 0.8; the best model for ET in 2018–2019 was ELM, with an R2 of 0.83 and 0.62, respectively. The inter-annual ET in the study area showed a "double-peak" dynamic variation, with peaks in 2018 and 2020; the intra-annual ET showed a single-peak cycle, with peaks in July–August. Seasonal differences were obvious, and spatially high-ET areas were mainly found in rural areas north of the Yangtze River and central and western China where agricultural land is concentrated. The net solar radiation, soil heat flux, soil temperature and humidity, and fractional vegetation cover all had significant positive effects on ET, with correlation coefficients ranging from 0.39 to 0.94. This study can provide methodological and scientific support for the quantitative and qualitative estimation of regional ET. [ABSTRACT FROM AUTHOR]

The technology of remote sensing-assisted tree species classification is increasingly developing, but the rapid refinement of tree species classification on a large scale is still challenging. As one of the treasures of ecological resources in China, Arxan has 80% forest cover, and tree species classification surveys guarantee ecological environment management and sustainable development. In this study, we identified tree species in three samples within the Arxan Duraer Forestry Zone based on the spectral, textural, and topographic features of unmanned aerial vehicle (UAV) multispectral remote sensing imagery and light detection and ranging (LiDAR) point cloud data as classification variables to distinguish among birch, larch, and nonforest areas. The best extracted classification variables were combined to compare the accuracy of the random forest (RF), support vector machine (SVM), and classification and regression tree (CART) methodologies for classifying species into three sample strips in the Arxan Duraer Forestry Zone. Furthermore, the effect on the overall classification results of adding a canopy height model (CHM) was investigated based on spectral and texture feature classification combined with field measurement data to improve the accuracy. The results showed that the overall accuracy of the RF was 79%, and the kappa coefficient was 0.63. After adding the CHM extracted from the point cloud data, the overall accuracy was improved by 7%, and the kappa coefficient increased to 0.75. The overall accuracy of the CART model was 78%, and the kappa coefficient was 0.63; the overall accuracy of the SVM was 81%, and the kappa coefficient was 0.67; and the overall accuracy of the RF was 86%, and the kappa coefficient was 0.75. To verify whether the above results can be applied to a large area, Google Earth Engine was used to write code to extract the features required for classification from Sentinel-2 multispectral and radar topographic data (create equivalent conditions), and six tree species and one nonforest in the study area were classified using RF, with an overall accuracy of 0.98, and a kappa coefficient of 0.97. In this paper, we mainly integrate active and passive remote sensing data for forest surveying and add vertical data to a two-dimensional image to form a three-dimensional scene. The main goal of the research is not only to find schemes to improve the accuracy of tree species classification, but also to apply the results to large-scale areas. This is necessary to improve the time-consuming and labor-intensive traditional forest survey methods and to ensure the accuracy and reliability of survey data. [ABSTRACT FROM AUTHOR]

Soil salinization is a critical and global environmental problem. Effectively mapping and monitoring the spatial distribution of soil salinity is essential. The main aim of this work was to map soil salinity in Shandong Province located on the Yellow River Delta of China using Sentinel-1/2 remote sensing data and digital elevation model (DEM) data, coupled with soil sampling data, and combined with four regression models: support vector regression (SVR), stepwise multi-regression (SMR), partial least squares regression (PLSR) and random forest regression (RFR). For these purposes, 60 soil samples were collected during the field survey conducted from 9 to 14 October 2019, corresponding to the Sentinel-1/2 and DEM data. Then we established a soil salinity and feature dataset based on the sampled data and the features extracted from Sentinel-1/2 and DEM data. This study adopted the feature importance of the RF model to screen all features. The results showed that the CRSI index made the greatest contribution in retrieving soil salinity in this region. In this paper, 18 sampling points were used to validate and compare the performance of the four models. The results reveal that, compared with the other regression models, the PLSR model has the best performance (R2 = 0.66, and RMSE = 1.30). Finally, the PLSR method was used to predict the spatial distribution of soil salinity in the Yellow River Delta. We concluded that the model can be used effectively for the quantitative estimation of soil salinity and provides a useful tool for ecological construction. [ABSTRACT FROM AUTHOR]

Combinations of multi-sensor remote sensing images and machine learning have attracted much attention in recent years due to the spectral similarity of wetland plant canopy. However, the integration of hyperspectral and quad-polarization synthetic aperture radar (SAR) images for classifying marsh vegetation has still been faced with the challenges of using machine learning algorithms. To resolve this issue, this study proposed an approach to classifying marsh vegetation in the Honghe National Nature Reserve, northeast China, by combining backscattering coefficient and polarimetric decomposition parameters of C-band and L-band quad-polarization SAR data with hyperspectral images. We further developed an ensemble learning model by stacking Random Forest (RF), CatBoost and XGBoost algorithms for marsh vegetation mapping and evaluated its classification performance of marsh vegetation between combinations of hyperspectral and full-polarization SAR data and any of the lone sensor images. Finally, this paper explored the effect of different polarimetric decomposition methods and wavelengths of radar on classifying wetland vegetation. We found that a combination of ZH-1 hyperspectral images, C-band GF-3, and L-band ALOS-2 quad-polarization SAR images achieved the highest overall classification accuracy (93.13%), which was 5.58–9.01% higher than that only using C-band or L-band quad-polarization SAR images. This study confirmed that stacking ensemble learning provided better performance than a single machine learning model using multi-source images in most of the classification schemes, with the overall accuracy ranging from 77.02% to 92.27%. The CatBoost algorithm was capable of identifying forests and deep-water marsh vegetation. We further found that L-band ALOS-2 SAR images achieved higher classification accuracy when compared to C-band GF-3 polarimetric SAR data. ALOS-2 was more sensitive to deep-water marsh vegetation classification, while GF-3 was more sensitive to shallow-water marsh vegetation mapping. Finally, scattering model-based decomposition provided important polarimetric parameters from ALOS-2 SAR images for marsh vegetation classification, while eigenvector/eigenvalue-based and two-component decompositions produced a great contribution when using GF-3 SAR images. [ABSTRACT FROM AUTHOR]

The parameters of hydrological models should be determined before applying those models to estimate or predict hydrological processes. The Xin'anjiang (XAJ) hydrological model is widely used throughout China. Since the prediction in ungauged basins (PUB) era, the regionalization of the XAJ model parameters has been a subject of intense focus; nevertheless, while many efforts have targeted parameters related to runoff yield using in-site data sets, classic regression has predominantly been applied. In this paper, we employed remotely sensed underlying surface data and a machine learning approach to establish models for estimating the runoff routing parameter, namely, CS, of the XAJ model. The study was conducted on 114 catchments from the Catchment Attributes and MEteorology for Large-sample Studies (CAMELS) data set, and the relationships between CS and various underlying surface characteristics were explored by a gradient-boosted regression tree (GBRT). The results showed that the drainage density, stream source density and area of the catchment were the three major factors with the most significant impact on CS. The best correlation coefficient (r), root mean square error (RMSE) and mean absolute error (MAE) between the GBRT-estimated and calibrated CS were 0.96, 0.06 and 0.04, respectively, verifying the good performance of GBRT in estimating CS. Although bias was noted between the GBRT-estimated and calibrated CS, runoff simulations using the GBRT-estimated CS could still achieve results comparable to those using the calibrated CS. Further validations based on two catchments in China confirmed the overall robustness and accuracy of simulating runoff processes using the GBRT-estimated CS. Our results confirm the following hypotheses: (1) with the help of large sample of catchments and associated remote sensing data, the ML-based approach can capture the nonstationary and nonlinear relationships between CS and the underlying surface characteristics and (2) CS estimated by ML from large samples has a robustness that can guarantee the overall performance of the XAJ mode. This study advances the methodology for quantitatively estimating the XAJ model parameters and can be extended to other parameters or other models. [ABSTRACT FROM AUTHOR]

Previous wildfire risk assessments have problems such as subjectivity of weight allocation and the linearization of statistical models, resulting in generally low robustness and low generalization ability of fire risk assessment models. Therefore, in this paper, we explored the potential of integration machine learning algorithms to build wildfire risk assessment models. Based on analyzing fire data's spatial and temporal distribution, we selected 10 triggering factors of topography, meteorology, vegetation, and human activities, using frequency ratio (FR) to provide uniform data representation of triggering factors. Next, we used the Bayesian optimization (BO) algorithm to perform hyperparametric optimization solutions for various machine learning models: support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost). Finally, we constructed an integration machine learning algorithm to acquire a fire risk grading map and the importance evaluation corresponding to each triggering factor. For validation purposes, we selected Liangshan Prefecture in Sichuan Province as the specific study area and obtained MCD64A1 burned area product to extract the extent of burned areas in Liangshan Prefecture from 2011 to 2020. The accuracy, kappa coefficient, and area under curve (AUC) were then applied to assess the predictive power and consistency of the fire risk classification maps. The experimental analysis showed that among the three models, FR-BO-XGBoost had the best performance in wildfire risk assessment in the Liangshan region (AUC = 0.887), followed by FR-BO-RF (AUC = 0.876) and FR-BO-SVM (AUC = 0.820). The feature importance result indicated that the study area's most significant effects on wildfires were precipitation, NDVI, land cover, and maximum temperature. The proposed method avoided the subjective weighting and model linearization problems. Compared with the previous methods, it automatically acquired the importance of the triggering factors to the wildfire, which had certain advantages in wildfire risk assessment, and was worthy of further promotion. [ABSTRACT FROM AUTHOR]

A forest fire susceptibility map generated with the fire susceptibility model is the basis of fire prevention resource allocation. A more reliable susceptibility map helps improve the effectiveness of resource allocation. Thus, further improving the prediction accuracy is always the goal of fire susceptibility modeling. This paper developed a forest fire susceptibility model based on an ensemble learning method, namely light gradient boosting machine (LightGBM), to produce an accurate fire susceptibility map. In the modeling, a subtropical national forest park in the Jiangsu province of China was used as the case study area. We collected and selected eight variables from the fire occurrence driving factors for modeling based on correlation analysis. These variables are from topographic factors, climatic factors, human activity factors, and vegetation factors. For comparative analysis, another two popular modeling methods, namely logistic regression (LR) and random forest (RF) were also applied to construct the fire susceptibility models. The results show that temperature was the main driving factor of fire in the area. In the produced fire susceptibility map, the extremely high and high susceptibility areas that were classified by LR, RF, and LightGBM were 5.82%, 18.61%, and 19%, respectively. The F1-score of the LightGBM model is higher than the LR and RF models. The accuracy of the model of LightGBM, RF, and LR is 88.8%, 84.8%, and 82.6%, respectively. The area under the curve (AUC) of them is 0.935, 0.918, and 0.868, respectively. The introduced ensemble learning method shows better ability on performance evaluation metrics. [ABSTRACT FROM AUTHOR]

After the "5·12" Wenchuan earthquake in 2008, collapses and landslides have occurred continuously, resulting in the accumulation of a large quantity of loose sediment on slopes or in gullies, providing rich material source reserves for the occurrence of debris flow and flash flood disasters. Therefore, it is of great significance to build a collapse and landslide susceptibility evaluation model in Wenchuan County for local disaster prevention and mitigation. Taking Wenchuan County as the research object and according to the data of 1081 historical collapse and landslide disaster points, as well as the natural environment, this paper first selects six categories of environmental factors (13 environmental factors in total) including topography (slope, aspect, curvature, terrain relief, TWI), geological structure (lithology, soil type, distance to fault), meteorology and hydrology (rainfall, distance to river), seismic impact (PGA), ecological impact (NDVI), and impact of human activity (land use). It then builds three single models (LR, SVM, RF) and three CF-based hybrid models (CF-LR, CF-SVM, CF-RF), and makes a comparative analysis of the accuracy and reliability of the models, thereby obtaining the optimal model in the research area. Finally, this study discusses the contribution of environmental factors to the collapse and the landslide susceptibility prediction of the optimal model. The research results show that (1) the areas prone to extremely high collapse and landslide predicted by the six models (LR, CF-LR, SVM, CF-SVM, RF and CF-RF) have an area of 730.595 km2, 377.521 km2, 361.772 km2, 372.979 km2, 318.631 km2, and 306.51 km2, respectively, and the frequency ratio precision of collapses and landslides is 0.916, 0.938, 0.955, 0.956, 0.972, and 0.984, respectively; (2) the ranking of the comprehensive index based on the confusion matrix is CF-RF>RF>CF-SVM>CF-LR>SVM>LR and the ranking of the AUC value is CF-RF>RF>CF-SVM>CF-LR>SVM>LR. To a certain extent, the coupling models can improve precision more over the single models. The CF-RF model ranks the highest in all indexes, with a POA value of 257.046 and an AUC value of 0.946; (3) rainfall, soil type, and distance to river are the three most important environmental factors, accounting for 24.216%, 22.309%, and 11.41%, respectively. Therefore, it is necessary to strengthen the monitoring of mountains and rock masses close to rivers in case of rainstorms in Wenchuan county and other similar areas prone to post-earthquake landslides. [ABSTRACT FROM AUTHOR]

Landslide is a devastating natural disaster, causing loss of life and property. It is likely to occur more frequently due to increasing urbanization, deforestation, and climate change. Landslide susceptibility mapping is vital to safeguard life and property. This article surveys machine learning (ML) models used for landslide susceptibility mapping to understand the current trend by analyzing published articles based on the ML models, landslide causative factors (LCFs), study location, datasets, evaluation methods, and model performance. Existing literature considered in this comprehensive survey is systematically selected using the ROSES protocol. The trend indicates a growing interest in the field. The choice of LCFs depends on data availability and case study location; China is the most studied location, and area under the receiver operating characteristic curve (AUC) is considered the best evaluation metric. Many ML models have achieved an AUC value > 0.90, indicating high reliability of the susceptibility map generated. This paper also discusses the recently developed hybrid, ensemble, and deep learning (DL) models in landslide susceptibility mapping. Generally, hybrid, ensemble, and DL models outperform conventional ML models. Based on the survey, a few recommendations and future works which may help the new researchers in the field are also presented. [ABSTRACT FROM AUTHOR]

Extraterrestrial solar radiation (ESR) is the essential basic background for solar radiation, which determines the occurrence of the weather and atmospheric phenomena. Since the influence of ESR variation on actual rugged terrain is a diverse, complex, and dynamic process, simulating ESR over a large spatial-temporal span, especially with a high-resolution digital elevation model (DEM), is a significant challenge. In this paper, we developed a new scheme for simulating ESR over the entirety of China using a DEM with a resolution of 30 m. To fully consider regional terrain status, the feature variables used were elevation, slope, and aspects of the located grid and the surrounding four grids to reveal the topography. In addition, latitude was used as a feature variable to consider the geographical location, and the month number was used to consider the duration. On the basis of different geographical locations, the training dataset was established from 20,000 grids. With the feature variable composition and training dataset, a backpropagation artificial neural network (BP ANN) was found to have the best performance compared with the other three machine learning methods in simulating ESR for a DEM. In terms of the proposed scheme and BP ANN, we drew an ESR map of China with a resolution of 30 m. The determination coefficient of the simulation result achieved 0.99 and the root-mean-square error was less than 50 MJ/m2 in all sample areas, confirming its remarkable accuracy. In terms of efficiency, the time consumption of ESR simulated using the proposed scheme shrinks over 150 times in all sample areas compared to that simulated via the theoretical model. Simultaneously, the developed scheme was also used to simulate an ESR for a DEM with a resolution of 90 m to verify the universality and robustness of the developed scheme. In addition, we used the proposed scheme to derive the direct solar radiation and global solar radiation, thereby further proving the reliability and applicability of our study. Overall, our work convincingly proved that the proposed scheme is a potential and effective approach for quickly simulating ESR with high accuracy. This study provides the basis for different solar radiation inversions of long time series and large spatial scales, offering additional insights for simulating ESR on a large spatial-temporal span. [ABSTRACT FROM AUTHOR]

Due to the country's rapid economic growth, the problem of air pollution in China is becoming increasingly serious. In order to achieve a win-win situation for the environment and urban development, the government has issued many policies to strengthen environmental protection. PM2.5 is the primary particulate matter in air pollution, so an accurate estimation of PM2.5 distribution is of great significance. Although previous studies have attempted to retrieve PM2.5 using geostatistical or aerosol remote sensing retrieval methods, the current rough resolution and accuracy remain as limitations of such methods. This paper proposes a fine-grained spatiotemporal PM2.5 retrieval method that comprehensively considers various datasets, such as Landsat 8 satellite images, ground monitoring station data, and socio-economic data, to explore the applicability of different machine learning algorithms in PM2.5 retrieval. Six typical algorithms were used to train the multi-dimensional elements in a series of experiments. The characteristics of retrieval accuracy in different scenarios were clarified mainly according to the validation index, R2. The random forest algorithm was shown to have the best numerical and PM2.5-based air-quality-category accuracy, with a cross-validated R2 of 0.86 and a category retrieval accuracy of 0.83, while both maintained excellent retrieval accuracy and achieved a high spatiotemporal resolution. Based on this retrieval model, we evaluated the PM2.5 distribution characteristics and hourly variation in the sample area, as well as the functions of different input variables in the model. The PM2.5 retrieval method proposed in this paper provides a new model for fine-grained PM2.5 concentration estimation to determine the distribution laws of air pollutants and thereby specify more effective measures to realize the high-quality development of the city. [ABSTRACT FROM AUTHOR]

Avalanche disasters are extremely destructive and catastrophic, often causing serious casualties, economic losses and surface erosion. However, far too little attention has been paid to utilizing remote sensing mapping avalanches quickly and automatically to mitigate calamity. Such endeavors are limited by formidable natural conditions, human subjective judgement and insufficient understanding of avalanches, so they have been incomplete and inaccurate. This paper presents an objective and widely serviceable method for regional auto-detection using the scattering and interference characteristics of avalanches extracted from Sentinel-1 SLC images. Six indices are established to distinguish avalanches from surrounding undisturbed snow. The active avalanche belts in Kizilkeya and Aktep of the Western TianShan Mountains in China lend urgency to this research. Implementation found that smaller avalanches can be consistently identified more accurately in descending images. Specifically, 281 and 311 avalanches were detected in the ascending and descending of Kizilkeya, respectively. The corresponding numbers on Aktep are 104 and 114, respectively. The resolution area of single avalanche detection can reach 0.09 km2. The performance of the model was excellent in all cases (areas under the curve are 0.831 and 0.940 in descending and ascending of Kizilkeya, respectively; and 0.807 and 0.938 of Aktep, respectively). Overall, the evaluation of statistical indices are POD > 0.75, FAR < 0.34, FOM < 0.13 and TSS > 0.75. The results indicate that the performance of the innovation proposed in this paper, which employs multivariate comprehensive descriptions of avalanche characteristics to actualize regional automatic detection, can be more objective, accurate, applicable and robust to a certain extent. The latest and more complete avalanche inventory generated by this design can effectively assist in addressing the increasingly severe avalanche disasters and improving public awareness of avalanches in alpine areas. [ABSTRACT FROM AUTHOR]

Precipitation nowcasting by radar echo extrapolation using machine learning algorithms is a field worthy of further study, since rainfall prediction is essential in work and life. Current methods of predicting the radar echo images need further improvement in prediction accuracy as well as in presenting the predicted details of the radar echo images. In this paper, we propose a two-stage spatiotemporal context refinement network (2S-STRef) to predict future pixel-level radar echo maps (deterministic output) more accurately and with more distinct details. The first stage is an efficient and concise spatiotemporal prediction network, which uses the spatiotemporal RNN module embedded in an encoder and decoder structure to give a first-stage prediction. The second stage is a proposed detail refinement net, which can preserve the high-frequency detailed feature of the radar echo images by using the multi-scale feature extraction and fusion residual block. We used a real-world radar echo map dataset of South China to evaluate the proposed 2S-STRef model. The experiments showed that compared with the PredRNN++ and ConvLSTM method, our 2S-STRef model performs better on the precipitation nowcasting, as well as at the image quality evaluating index and the forecasting indices. At a given 45 dBZ echo threshold (heavy precipitation) and with a 2 h lead time, the widely used CSI, HSS, and SSIM indices of the proposed 2S-STRef model are found equal to 0.195, 0.312, and 0.665, respectively. In this case, the proposed model outperforms the OpticalFlow method and PredRNN++ model. [ABSTRACT FROM AUTHOR]

The accurate and timely estimation of regional crop biomass at different growth stages is of great importance in guiding crop management decision making. The recent availability of long time series of remote sensing data offers opportunities for crop monitoring. In this paper, four machine learning models, namely random forest (RF), support vector machine (SVM), artificial neural network (ANN), and extreme gradient boosting (XGBoost) were adopted to estimate the seasonal corn biomass based on field observation data and moderate resolution imaging spectroradiometer (MODIS) reflectance data from 2012 to 2019 in the middle reaches of the Heihe River basin, China. Nine variables were selected with the forward feature selection approach from among twenty-seven variables potentially influencing corn biomass: soil-adjusted total vegetation index (SATVI), green ratio vegetation index (GRVI), Nadir_B7 (2105–2155 nm), Nadir_B6 (1628–1652 nm), land surface water index (LSWI), normalized difference vegetation index (NDVI), Nadir_B4 (545–565 nm), and Nadir_B3 (459–479 nm). The results indicated that the corn biomass was suitably estimated (the coefficient of determination (R2) was between 0.72 and 0.78) with the four machine learning models. The XGBoost model performed better than the other three models (R2 = 0.78, root mean squared error (RMSE) = 2.86 t/ha and mean absolute error (MAE) = 1.86 t/ha). Moreover, the RF model was an effective method (R2 = 0.77, RMSE = 2.91 t/ha and MAE = 1.91 t/ha), with a performance comparable to that of the XGBoost model. This study provides a reference for estimating crop biomass from MOD43A4 datasets. In addition, the research demonstrates the potential of machine learning techniques to achieve a relatively accurate estimation of daily corn biomass at a large scale. [ABSTRACT FROM AUTHOR]

Explicit spatial information about crop types on smallholder farms is important for the development of local precision agriculture. However, due to highly fragmented and heterogeneous cropland landscapes, fine-scale mapping of smallholder crops, based on low- and medium-resolution satellite images and relying on a single machine learning (ML) classifier, generally fails to achieve satisfactory performance. This paper develops an ensemble ML-based framework to improve the accuracy of parcel-level smallholder crop mapping from very high spatial resolution (VHSR) images. A typical smallholder agricultural area in central China covered by WorldView-2 images is selected to demonstrate our approach. This approach involves the task of distinguishing eight crop-level agricultural land use types. To this end, six widely used individual ML classifiers are evaluated. We further improved their performance by independently implementing bagging and stacking ensemble learning (EL) techniques. The results show that the bagging models improved the performance of unstable classifiers, but these improvements are limited. In contrast, the stacking models perform better, and the Stacking #2 model (overall accuracy = 83.91%, kappa = 0.812), which integrates the three best-performing individual classifiers, performs the best of all of the built models and improves the classwise accuracy of almost all of the land use types. Since classification performance can be significantly improved without adding costly data collection, stacking-ensemble mapping approaches are valuable for the spatial management of complex agricultural areas. We also demonstrate that using geometric and textural features extracted from VHSR images can improve the accuracy of parcel-level smallholder crop mapping. The proposed framework shows the great potential of combining EL technology with VHSR imagery for accurate mapping of smallholder crops, which could facilitate the development of parcel-level crop identification systems in countries dominated by smallholder agriculture. [ABSTRACT FROM AUTHOR]

Precise vegetation maps of mountainous areas are of great significance to grasp the situation of an ecological environment and forest resources. In this paper, while multi-source geospatial data can generally be quickly obtained at present, to realize effective vegetation mapping in mountainous areas when samples are difficult to collect due to their perilous terrain and inaccessible deep forest, we propose a novel and intelligent method of sample collection for machine-learning (ML)-based vegetation mapping. First, we employ geo-objects (i.e., polygons) from topographic partitioning and constrained segmentation as basic mapping units and formalize the problem as a supervised classification process using ML algorithms. Second, a previously available vegetation map with rough-scale label information is overlaid on the geo-object-level polygons, and candidate geo-object-based samples can be identified when all the grids' labels of vegetation types within the geo-objects are the same. Third, various kinds of geo-object-level features are extracted according to high-spatial-resolution remote sensing (HSR-RS) images and multi-source geospatial data. Some unreliable geo-object-based samples are rejected in the candidate set by comparing their features and the rules based on local expert knowledge. Finally, based on these automatically collected samples, we train the model using a random forest (RF)-based algorithm and classify all the geo-objects with labels of vegetation types. A case experiment of Taibai Mountain in China shows that the methodology has the ability to achieve good vegetation mapping results with the rapid and convenient sample collection scheme. The map with a finer geographic distribution pattern of vegetation could clearly promote the vegetation resources investigation and monitoring of the study area; thus, the methodological framework is worth popularizing in the mapping areas such as mountainous regions where the field survey sampling is difficult to implement. [ABSTRACT FROM AUTHOR]

Remote sensing image segmentation with samples imbalance is always one of the most important issues. Typically, a high-resolution remote sensing image has the characteristics of high spatial resolution and low spectral resolution, complex large-scale land covers, small class differences for some land covers, vague foreground, and imbalanced distribution of samples. However, traditional machine learning algorithms have limitations in deep image feature extraction and dealing with sample imbalance issue. In the paper, we proposed an improved full-convolution neural network, called DeepLab V3+, with loss function based solution of samples imbalance. In addition, we select Sentinel-2 remote sensing images covering the Yuli County, Bayingolin Mongol Autonomous Prefecture, Xinjiang Uygur Autonomous Region, China as data sources, then a typical region image dataset is built by data augmentation. The experimental results show that the improved DeepLab V3+ model can not only utilize the spectral information of high-resolution remote sensing images, but also consider its rich spatial information. The classification accuracy of the proposed method on the test dataset reaches 97.97%. The mean Intersection-over-Union reaches 87.74%, and the Kappa coefficient 0.9587. The work provides methodological guidance to sample imbalance correction, and the established data resource can be a reference to further study in the future. [ABSTRACT FROM AUTHOR]

Texture information allows characterizing the regions of interest in a scene. It refers to the spatial organization of the fundamental microstructures in natural images. Texture extraction has been a challenging problem in the field of image processing for decades. In this paper, different techniques based on the classic Bag of Words (BoW) approach for solving the texture extraction problem in the case of hyperspectral images of the Earth surface are proposed. In all cases the texture extraction is performed inside regions of the scene called superpixels and the algorithms profit from the information available in all the bands of the image. The main contribution is the use of superpixel segmentation to obtain irregular patches from the images prior to texture extraction. Texture descriptors are extracted from each superpixel. Three schemes for texture extraction are proposed: codebook-based, descriptor-based, and spectral-enhanced descriptor-based. The first one is based on a codebook generator algorithm, while the other two include additional stages of keypoint detection and description. The evaluation is performed by analyzing the results of a supervised classification using Support Vector Machines (SVM), Random Forest (RF), and Extreme Learning Machines (ELM) after the texture extraction. The results show that the extraction of textures inside superpixels increases the accuracy of the obtained classification map. The proposed techniques are analyzed over different multi and hyperspectral datasets focusing on vegetation species identification. The best classification results for each image in terms of Overall Accuracy (OA) range from 81.07% to 93.77% for images taken at a river area in Galicia (Spain), and from 79.63% to 95.79% for a vast rural region in China with reasonable computation times. [ABSTRACT FROM AUTHOR]

Gully erosion is a widespread natural hazard. Gully mapping is critical to erosion monitoring and the control of degraded areas. The analysis of high-resolution remote sensing images (HRI) and terrain data mixed with developed object-based methods and field verification has been certified as a good solution for automatic gully mapping. Considering the availability of data, we used only open-source optical images (Google Earth images) to identify gully erosion through image feature modeling based on OBIA (Object-Based Image Analysis) in this paper. A two-end extrusion method using the optimal machine learning algorithm (Light Gradient Boosting Machine (LightGBM)) and eCognition software was applied for the automatic extraction of gullies at a regional scale in the black soil region of Northeast China. Due to the characteristics of optical images and the design of the method, unmanaged gullies and gullies harnessed in non-forest areas were the objects of extraction. Moderate success was achieved in the absence of terrain data. According to independent validation, the true overestimation ranged from 20% to 30% and was mainly caused by land use types with high erosion risks, such as bare land and farm lanes being falsely classified as gullies. An underestimation of less than 40% was adjacent to the correctly extracted gullied areas. The results of extraction in regions with geographical object categories of a low complexity were usually more satisfactory. The overall performance demonstrates that the present method is feasible for gully mapping at a regional scale, with high automation, low cost, and acceptable accuracy. [ABSTRACT FROM AUTHOR]

Salt-affected soil is a prominent ecological and environmental problem in dry farming areas throughout the world. China has nearly 9.9 million km 2 of salt-affected land. The identification, monitoring, and utilization of soil salinization have become important research topics for promoting sustainable progress. In this paper, using field-measured spectral data and soil salinity parameter data, through analysis and transformation of spectral data, five machine learning models, namely, random forest regression (RFR), support vector regression (SVR), gradient-boosted regression tree (GBRT), multilayer perceptron regression (MLPR), and least angle regression (Lars) are compared. The following performance measures of each model were evaluated: the collinear problems, handling data noise, stability, and the accuracy. In terms of these four aspects, the performance of each model on estimating soil salinity is evaluated. The results demonstrate that among the five models, RFR has the best performance in dealing with collinearity, RFR and MLPR have the best performance in dealing with data noise, and the SVR model is the most stable. The Lars model has the highest accuracy, with a determination coefficient ( R 2 ) of 0.87, ratio of performance to deviation (RPD) of 2.67, root mean square error (RMSE) of 0.18, and mean absolute percentage error (MAPE) of 0.11. Then, the comprehensive comparison and analysis of the five models are carried out, and it is found that the comprehensive performance of RFR model is the best; hence, this method is most suitable for estimating soil salinity using hyperspectral data. This study can provide a reference for the selection of regression methods in subsequent studies on estimating soil salinity using hyperspectral data. [ABSTRACT FROM AUTHOR]