Search

In the current context of carbon neutrality, afforestation is an effective means of absorbing carbon dioxide. Stock can be used not only as an economic value index of forest wood resources but also as an important index of biomass and carbon storage estimation in forest emission reduction project evaluation. In this paper, we propose a data-driven machine learning framework and method for predicting plantation stock based on airborne LiDAR + satellite remote sensing, and carried out experimental verification at the site of the National Forest emission reduction project in Southern China. We used step-up regression and random forest (RF) to screen LiDAR and Landsat 8 OLI multispectral indicators suitable for the prediction of plantation stock, and constructed a plantation stock model based on machine learning (support vector machine regression, RF regression). Our method is compared with traditional statistical methods (stepwise regression and partial least squares regression). Through the verification of 57 plantation field survey data, the accuracy of the stand estimation model constructed using the RF method is generally better (ΔR2 = 0.01~0.27, ΔRMSE = 1.88~13.77 m3·hm−2, ΔMAE = 1.17~13.57 m3·hm−2). The model evaluation accuracy based on machine learning is higher than that of the traditional statistical method, and the fitting R2 is greater than 0.91, while the fitting R2 of the traditional statistical method is 0.85. The best fitting models were all support vector regression models. The combination of UAV point clouds and satellite multi-spectral images has the best modeling effect, followed by LiDAR point clouds and Landsat 8. At present, this method is only applicable to artificial forests; further verification is needed for natural forests. In the future, the density and quality of higher clouds could be increased. The validity and accuracy of the method were further verified. This paper provides a method for predicting the accumulation of typical Chinese plantations at the forest farm scale based on the "airborne LiDAR + satellite remote sensing" data-driven machine learning modeling, which has potential application value for the current carbon neutrality goal of the southern plantation forest emission reduction project. [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]

Air-quality monitoring and forecasting are crucial for atmosphere pollution control and management. We propose an innovative data-driven framework for air quality index (AQI) prediction by integrating GNSS radio occultation (GNSS-RO) observation and weather modeling. Empowered by the state-of-the-art machine learning approach, our method can effectively predict regional AQI with a comparable accuracy much more quickly than the traditional numerical modeling and simulation approach. In a real case study using a representative region of China, our data-driven approach achieves a 2000 times speedup; meanwhile, the prediction error measured by rRMSE is only 2.4%. We investigate further the effects of different models, hyperparameters, and meteorological factors on the performance of our AQI prediction framework, and reveal that wind field and atmospheric boundary-layer height are important influencing factors of AQI. This paper showcases a direct application of GNSS-RO observation in assisting in forecasting regional AQI. From a machine learning point of view, it provides a new way to leverage the unique merits of GNSS atmospheric remote sensing technology with the help of the more traditional weather forecasting modeling approach. [ABSTRACT FROM AUTHOR]

The rapid and precise acquisition of the agricultural plastic greenhouse (PG) spatial distribution is essential in understanding PG usage and degradation, ensuring agricultural production, and protecting the ecological environment and human health. It is of great practical significance to realize the effective utilization of remote sensing images in the agricultural field and improve the extraction accuracy of PG remote sensing data. In this study, Landsat operational land imager (OLI) remote sensing images were used as data sources, and Shandong Province, which has the largest PG distribution in China, was selected as the study area. PGs in the study area were identified by means of contour recognition, feature set construction of the spatial structure, and machine learning. The results were as follows. (1) Through an optimal segmentation parameter approach, it was determined that the optimal segmentation scale for size, shape, and compactness should be set at 20, 0.8, and 0.5, respectively, which significantly improved PG contour recognition. (2) Among the 72 feature variables for PG spatial recognition, the number of features and classification accuracy showed a trend of first gradually increasing and then decreasing. Among them, fifteen feature variables, including the mean of bands 2 and 5; six index features (NDWI, GNDVI, SWIR1_NIR, NDVI, and PMLI); two shape features, the density and shape index; and two texture features, the contrast and standard deviation, played an important role. (3) According to the recall rate, accuracy rate, and F-value of three machine learning methods, random forest (RDF), CART decision tree (CART), and support vector machine (SVM), SVM had the best classification effect. The classification method described in this paper can accurately extract continuous plastic greenhouses through remote sensing images and provide a reference for the application of facility agriculture and non-point-source pollution control. [ABSTRACT FROM AUTHOR]

Going beyond previous work, this paper presents a systematic literature review that explores the deployment of satellites, drones, and ground-based sensors for yield prediction in agriculture. It covers multiple aspects of the topic, including crop types, key sensor platforms, data analysis techniques, and performance in estimating yield. To this end, datasets from Scopus and Web of Science were analyzed, resulting in the full review of 269 out of 1429 retrieved publications. Our study revealed that China (93 articles, >1800 citations) and the USA (58 articles, >1600 citations) are prominent contributors in this field; while satellites were the primary remote sensing platform (62%), followed by airborne (30%) and proximal sensors (27%). Additionally, statistical methods were used in 157 articles, and model-based approaches were utilized in 60 articles, while machine learning and deep learning were employed in 142 articles and 62 articles, respectively. When comparing methods, machine learning and deep learning methods exhibited high accuracy in crop yield prediction, while other techniques also demonstrated success, contingent on the specific crop platform and method employed. The findings of this study serve as a comprehensive roadmap for researchers and farmers, enabling them to make data-driven decisions and optimize agricultural practices, paving the way towards a fully digitized yield prediction. [ABSTRACT FROM AUTHOR]

Rice is an important food crop in China, and diseases and pests are the main factors threatening its safety, ecology, and efficient production. The development of remote sensing technology provides an important means for non-destructive and rapid monitoring of diseases and pests that threaten rice crops. This paper aims to provide insights into current and future trends in remote sensing for rice crop monitoring. First, we expound the mechanism of remote sensing monitoring of rice diseases and pests and introduce the applications of different commonly data sources (hyperspectral data, multispectral data, thermal infrared data, fluorescence, and multi-source data fusion) in remote sensing monitoring of rice diseases and pests. Secondly, we summarize current methods for monitoring rice diseases and pests, including statistical discriminant type, machine learning, and deep learning algorithm. Finally, we provide a general framework to facilitate the monitoring of rice diseases or pests, which provides ideas and technical guidance for remote sensing monitoring of unknown diseases and pests, and we point out the challenges and future development directions of rice disease and pest remote sensing monitoring. This work provides new ideas and references for the subsequent monitoring of rice diseases and pests using remote sensing. [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 water content is an important indicator used to maintain the ecological balance of farmland. The efficient spatial prediction of soil water content is crucial for ensuring crop growth and food production. To this end, 104 farmland soil samples were collected in the Yellow River Delta (YRD) in China, and the soil water content was determined using the drying method. A gradient boosting decision tree (GBDT) model based on a tree-structured Parzen estimator (TPE) hyperparametric optimization was developed, and then the soil water content was predicted and mapped based on the soil texture and vegetation index from Sentinel-2 remote sensing images. The results of statistical analysis showed that the soil water content had a high coefficient of variation (55.30%), a non-normal distribution, and complex spatial variability. Compared with other models, the TPE-GBDT model had the highest prediction accuracy (RMSE = 6.02% and R2 = 0.71), and its mapping results showed that the areas with high soil water content were distributed on both sides of the river and near the estuary. Furthermore, the results of Shapley additive explanation (SHAP) analysis showed that the soil texture (PC2 and PC5), modified normalized difference vegetation index (MNDVI), and Sentinel-2 red edge position (S2REP) index provided important contributions to the spatial prediction of soil water content. We found that the hydraulic physical properties of soil texture and the vegetation characteristics (such as vegetation coverage, root action, and transpiration) are the key factors affecting the spatial migration and heterogeneity of the soil water content in the study area. The above results show that the TPE algorithm can quickly capture the hyperparameters that are most suitable for the GBDT model, so that the GBDT model can ensure prediction accuracy, reduce the loss function with less training data, and accurately learn of the nonlinear relationship between soil water content and environmental factors. This paper proposes a machine learning method for hyperparameter optimization that shows considerable potential to predict the spatial heterogeneity of soil water content, which can effectively support regional farmland soil and water conservation and high-quality agricultural development. [ABSTRACT FROM AUTHOR]

Simple Summary: Grasshoppers are among the most dangerous agricultural pests of China. However, the monitoring, prediction and control of grasshoppers are complex and difficult. Therefore, it is crucial to detect the key factors affecting the spatial distribution of grasshopper occurrence, understand the role of the environmental factors in grasshopper occurrence, and study whether different laws exist between different grass types. Here we conduct a species–environmental matching model integrated by Maxent model and remote sensing data to identify the spatial distribution of grasshopper occurrence in Inner Mongolia of China, analyze the related environmental variables and detect the most relevant environmental factors for grasshopper occurrence both in typical steppe and meadow steppe. Grasshoppers mainly threaten natural grassland vegetation and crops. Therefore, it is of great significance to understand the relationship between environmental factors and grasshopper occurrence. This paper studies the spatial distribution and key factors of grasshopper occurrence in two grass types by integrating a machine learning model (Maxent) and remote sensing data within the major grasshopper occurrence areas of Inner Mongolia, China. The modelling results demonstrate that the typical steppe has larger suitable area and more proportion for grasshopper living than meadow steppe. The soil type, above biomass, altitude and temperature mainly determine the grasshopper occurrence in typical steppe and meadow steppe. However, the contribution of these factors in the two grass types is significantly different. In addition, related vegetation and meteorological factors affect the different growing stages of grasshoppers between the two grass types. This study clearly defines the different effects of key environmental factors (meteorology, vegetation, soil and topography) for grasshopper occurrence in typical steppe and meadow steppe. It also provides a methodology to guide early warning and precautions for grasshopper pest prevention. The findings of this study will be helpful for future management measures, to ensure grass ecological environment security and the sustainable development of grassland. [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]

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]

The Yellow River Delta (YRD), known for its vast and diverse wetland ecosystem, is the largest estuarine delta in China. However, human activities and climate change have significantly degraded the wetland ecosystem in recent decades in the YRD. Therefore, an understanding of the land use modifications is essential for the efficient management and preservation of ecosystems in this region. This study utilized time series of remote sensing data and the extreme gradient boosting method to generate land use maps of the YRD from 2000 to 2020. Several methods, including transition matrix, land use dynamic degree, and standard deviation ellipse, were employed to explore the characteristics of land use transitions. The results underscore significant spatial variations in land use over the past two decades. The most rapid increase was observed in built-up area, followed by terrestrial water and tidal flats, while unutilized land experienced the fastest decrease, followed by forest–grassland. The spatial distribution patterns of agricultural land, built-up area, terrestrial water, and forest–grassland demonstrated stronger directionality compared to other land use types. The wetlands have expanded in size and improved in structure. Unutilized land has been converted into artificial wetlands comprising ponds, reservoirs, salt ponds, shrimp and crab ponds, and natural wetlands featuring mudflats and forest–grassland. The wetland conservation efforts after 2008 have proven very effective, playing a positive role in ecological and environmental preservation, as well as in regional sustainable development. [ABSTRACT FROM AUTHOR]

The frequent occurrence of landslides poses a serious threat to people's lives and property. In order to evaluate disaster hazards based on remote sensing images via machine learning means, it is essential to establish an image database with manually labeled boundaries of landslides. However, the existing datasets do not cover diverse types of mountainous landslides. To address this issue, we propose a high-resolution (1 m) diverse mountainous landslide remote sensing dataset (DMLD), including 990 landslide instances across different terrain in southwestern China. To evaluate the performance of the DMLD, seven state-of-the-art deep learning models with different loss functions were implemented on it. The experiment results demonstrate not only that all of these deep learning methods with different characteristics can adapt well to the DMLD, but also that the DMLD has potential adaptability to other geographical regions. [ABSTRACT FROM AUTHOR]

China's northern Hulun Lake is a significant body of water internationally. The issue of eutrophication has gained prominence in recent years. The achievement of precise chlorophyll-a (Chl-a) monitoring is crucial for safeguarding Hulun Lake's ecosystem. The machine learning-based remote sensing inversion method has been shown to be effective in capturing the intricate relationship between independent and dependent variables; however, it lacks a priori knowledge and is limited by the quality of remote sensing data sources. The relationship between independent and dependent variables can be more accurately simulated with the use of suitable auxiliary variables. Therefore, three machine learning models—random forest (RF), adaptive boosting (AdaBoost), and extreme gradient boosting (XGBoost)—were established in this study using meteorological observation parameters as auxiliary variables combined with Sentinel-2 satellite image remote sensing band combinations as independent variables and measured Chl-a data as dependent variables. The estimation effects before and after the fusion of meteorological ground observation data were compared, and the best model was used to estimate the spatial–temporal variation trend of Chl-a in the regional water body. The results show that (1) the addition of meteorological parameters as auxiliary variables improved the precision of the three machine models; the decision coefficient (R2) rose by 7.25%, 5.71%, and 7.20%, respectively, to 0.76, 0.66, and 0.73. (2) The concentration of Chl-a in the lake region was projected from June to October 2019 to October 2021 using the RF optimal estimating model of meteorological fusion. The northeast, southwest, and south of the lake were where the comparatively high concentration values of Chl-a were located, whereas the lake's center had a generally low concentration of the substance. Chromatically, Chl-a typically peaked in August after initially increasing and then declining. (3) The three rivers that feed into the river have varying levels of water pollution, with chemical oxygen demand (COD) and total nitrogen (TN) pollution being the most severe. This is what primarily caused the higher levels of Chl-a in the northeast, southwest, and south. This study is crucial for the preservation and restoration of Hulun Lake's natural ecosystem and offers some technical support for the monitoring of the lake's concentration of Chl-a. [ABSTRACT FROM AUTHOR]

A rapid development of marine infrastructures can be observed along the global coasts. Offshore wind farms, oil and gas platforms, artificial islands, aquaculture, and more, are being constructed without a proper quantification of these human activities. Therefore, effective monitoring is required to maintain transparency towards environmental standards, marine resource management, inventorying objects, and global security. This study reviews remote sensing-based approaches to offshore infrastructure detection over the past 12 years. We analyzed 89 studies from over 30 scientific journals, highlighting spatial and temporal trends, methodological approaches, and regional and thematic research foci. Our results show a significant increase in research interest, especially since 2019. Asia, and especially China, is the predominant focus region in terms of first authorship, funding, and areas of investigation. Aquaculture is the most studied infrastructure, followed by platforms, offshore wind farms, and artificial islands. Gaofen, Sentinel, and Landsat are the most used satellite sensors for detection. The apparent shift towards automated detection methods, especially Deep Learning algorithms, reflects advances in computer vision. This study highlights the key role of earth observation in the field of off-shore infrastructure detection, which can contribute towards outlining effective monitoring practices for marine activities, as well as highlighting important knowledge gaps. [ABSTRACT FROM AUTHOR]

Since industrialization, global carbon dioxide (CO2) emissions have been rising substantially, playing an increasingly important role in global warming and climate change. As the largest CO2 emitter, China has proposed an ambitious reduction plan of peaking before 2030 and achieving carbon neutrality by 2060. Calculation of CO2 emissions inventories at regional scales (e.g., city and county) has great significance in terms of China's regional carbon policies as well as in achieving the national targets. However, most of the existing emissions data were calculated based on fossil fuel consumptions and were thus limited to the provinces in China, making it challenging to compare and analyze the CO2 emissions of different cities and counties within a province. Machine learning methods provided a promising alternative but were still suffering from the lack of availability of training samples at city or county scales. Accordingly, this study proposed to use the energy consumption per unit GDP (ECpGDP) and GDP to calculate the effective CO2 emissions, which are the CO2 emissions if all consumed energy was generated by standard coal. Random forest models were then trained to establish relationships between the remote sensing night-light data and effective CO2 emissions. A total of eight predictor variables were used, including the night-light data, the urbanization ratio, the population density, the type of sensors and administrative divisions, latitude, longitude, and the area of each city or county. Meanwhile, the mean value of the five-fold cross-validation model was used as the estimated effective CO2 emissions in order to avoid overfitting. The evaluation showed a root mean square error (RMSE) of 10.972 million tons and an overall Pearson's correlation coefficient (R) of 0.952, with satisfactory spatial and temporal consistency. The effective CO2 emissions of 349 cities and 2843 counties in China during 1992–2021 were obtained, providing a promising dataset for CO2-emission-related applications. [ABSTRACT FROM AUTHOR]

The China–Nepal Transportation Corridor is vital to the country's efforts to build a land trade route in South Asia and promote the Ring-Himalayan Economic Cooperation Belt. Due to the complex geological structure and topographical environment of the Qinghai–Tibet Plateau, coupled with the impact of climate change, the frequent occurrence of geological disasters has increased the operational difficulty of the China–Nepal Highway and the construction difficulty of the China–Nepal Railway. However, to date, there has been no systematic study of the spatial distribution of landslides along the entire route within the area, the factors influencing landslides at different scales, or the causes of landslides under different topographic backgrounds. There is an even greater lack of research on areas threatened by potential landslides. This study comprehensively applies remote sensing, mathematical statistics, and machine learning methods to map landslides along the China–Nepal transportation corridor, explore the influencing factors and causes of different types of landslides, and investigate the distribution characteristics of potential landslides. A total of 609 historic landslides have been interpreted in the study area and were found to be distributed along faults and locally concentrated. The strata from which landslides develop are relatively weak and are mainly distributed within 2 km of a fault with a slope between 20° and 30°. The direction of slope for the majority of landslides is south to south-west, and their elevation is between 4000 and 5000 m. In addition, we discovered a power law relationship between landslide area and volume (VL = 2.722 × AL1.134) and determined that there were 47 super-large landslides, 213 large landslides, and 349 small and medium-sized landslides in the area, respectively. Slope is the most significant influencing factor for the development of landslides in the area. Apart from slope, faults and strata significantly influence the development of large and medium-small landslides, respectively. We have identified 223 potential landslides in the region, 15 of which directly threaten major transport routes, mainly in the Renbu Gorge section of the China–Nepal Highway and the proposed China–Nepal Railway section from Peikucuo to Gyirong County. In addition, we also discussed the causes of landslides within three geomorphic units in the region. First, the combined effects of faulting, elevation, and relatively weak strata contribute to the development of super-large and large landslides in the Gyirong basin and gorge. Second, the relatively weak strata and the cumulative damaging effects of earthquakes promote the development of small and medium-sized landslides in the Xainza-Dinggye rift basin. Third, under the combined effect of the hanging wall effect of thrust faults and the relatively weak material composition, landslides of various types have developed in the Nagarzê mountain. It is worth noting that potential landslides have developed in all three geomorphic units mentioned above. This study provides data and theory to assist in the accurate mitigation and control of landslide hazards in the corridor. [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]

The massive increase in the amount of greenhouse gases in the atmosphere, especially carbon dioxide (CO2), has had a significant impact on the global climate. Research has revealed that lakes play an important role in the global carbon cycle and that they can shift between the roles of carbon sources and sinks. This study used Landsat satellite images to analyze the spatiotemporal characteristics and factors influencing the CO2 changes in Chagan Lake in China. We conducted six water sampling campaigns at Chagan Lake in 2020–2021 and determined the partial pressure of carbon dioxide (pCO2) from 110 water samples. Landsat surface reflectance was matched with water sampling events within ±7 days of satellite overpasses, yielding 75 matched pairs. A regression analysis indicated strong associations between pCO2 and both the band difference model of the near-infrared band and green band (Band 5-Band 3, R2 = 0.83, RMSE = 27.55 μatm) and the exponential model [((exp(b3) − exp(b5))2/(exp(b3) + exp(b5))2, R2 = 0.82, RMSE = 27.99 μatm]. A comparison between the performances of a linear regression model and a machine learning model found that the XGBoost model had the highest fitting accuracy (R2 = 0.94, RMSE = 16.86 μatm). We used Landsat/OLI images acquired mainly in 2021 to map pCO2 in Chagan Lake during the ice-free period. The pCO2 in the surface water of Chagan Lake showed considerable spatiotemporal variability within a range of 0–200 μatm. pCO2 also showed significant seasonal variations, with the lowest and highest mean values in autumn (30–50 μatm) and summer (120–150 μatm), respectively. Spatially, the pCO2 values in the southeast of Chagan Lake were higher than those in the northwest. The CO2 fluxes were calculated based on the pCO2 and ranged from −3.69 to −2.42 mmol/m2/d, indicating that Chagan Lake was absorbing CO2 (i.e., it was a weak carbon sink). Temperature, chlorophyll a, total suspended matter, and turbidity were found to have reinforcing effects on the overall trend of pCO2, while the Secchi disk depth was negatively correlated with pCO2. The results of this study provide valuable insights for assessing the role of lakes in the carbon cycle in the context of climate change. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is significant for landslide risk assessment. However, there remains no consensus on which method is optimal for LSM. This study implements a dynamic approach to landslide hazard mapping by integrating spatio-temporal probability analysis with time-varying ground deformation velocity derived from the MT-InSAR (Multi-Temporal InSAR) method. Reliable landslide susceptibility maps (LSMs) can inform landslide risk managers and government officials. First, sixteen factors were selected to construct a causal factor system for LSM. Next, Pearson correlation analysis, multicollinearity analysis, information gain ratio, and GeoDetector methods were applied to remove the least important factors of STI, plan curvature, TRI, and slope length. Subsequently, information quantity (IQ), logistic regression (LR), frequency ratio (FR), artificial neural network (ANN), random forest (RF), support vector machine (SVM), and convolutional neural network (CNN) methods were performed to construct the LSM. The results showed that the distance to a river, slope angle, distance from structure, and engineering geological rock group were the main factors controlling landslide development. A comprehensive set of statistical indicators was employed to evaluate these methods' effectiveness; sensitivity, F1-measure, and AUC (area under the curve) were calculated and subsequently compared to assess the performance of the methods. Machine learning methods' training and prediction accuracy were higher than those of statistical methods. The AUC values of the IQ, FR, LR, BP-ANN, RBF-ANN, RF, SVM, and CNN methods were 0.810, 0.854, 0.828, 0.895, 0.916, 0.932, 0.948, and 0.957, respectively. Although the performance order varied for other statistical indicators, overall, the CNN method was the best, while the BP-ANN and RBF-ANN method was the worst among the five examined machine methods. Hence, adopting the CNN approach in this study can enhance LSM accuracy, catering to the needs of planners and government agencies responsible for managing landslide-prone areas and preventing landslide-induced disasters. [ABSTRACT FROM AUTHOR]

Ulva prolifera and Sargassum are two common floating macroalgae in China's coastal algal bloom events. Ulva prolifera frequently emerges concomitantly with Sargassum outbreaks, thereby presenting challenges to the monitoring of algal blooms, thereby presenting challenges to the monitoring of algae. To tackle the challenge of differentiating between Ulva prolifera and Sargassum, this study employs Sentinel-2 MSI data for spectral analysis. Notably, significant disparities in the Remote Top of Atmosphere Reflectance (Rtoa) between Ulva prolifera and Sargassum are observed. This study proposes a random forest-based algorithm for discriminating between Ulva prolifera and Sargassum in the regions of the Yellow Sea and East China Sea. The algorithm introduced in this study attains remarkable accuracy in distinguishing Ulva prolifera and Sargassum within Sentinel-2 MSI data, achieving identical F1 scores of 99.1% for both. Moreover, when tested with GF-1 WFV data, the algorithm showcases outstanding performance; this demonstrates the algorithm's robustness and its ability to mitigate the uncertainty linked to threshold selection. Simultaneously, a comparative analysis of algae distribution was conducted for both 2017 and the period from January to May 2023. Experimental results indicate that the algorithm exhibits high accuracy in distinguishing between Ulva prolifera and Sargassum. This capability will significantly enhance the monitoring of large algae in maritime regions; this holds crucial theoretical significance and offers substantial practical value in the realm of marine ecological conservation. [ABSTRACT FROM AUTHOR]

The trophic state is an important factor reflecting the health state of lake ecosystems. To accurately assess the trophic state of large lakes, an integrated framework was developed by combining remote sensing data, field monitoring data, machine learning algorithms, and optimization algorithms. First, key meteorological and environmental factors from in situ monitoring were combined with remotely sensed reflectance data and statistical analysis was used to determine the main factors influencing the trophic state. Second, a trophic state index (TSI) inversion model was constructed using a machine learning algorithm, and this was then optimized using the sparrow search algorithm (SSA) based on a backpropagation neural network (BP-NN) to establish an SSA-BP-NN model. Third, a typical lake in China (Hongze Lake) was chosen as the case study. The application results show that, when the key environmental factors (pH, temperature, average wind speed, and sediment content) and the band combination data from Sentinel-2/MSI were used as input variables, the performance of the model was improved (R2 = 0.936, RMSE = 1.133, MAPE = 1.660%, MAD = 0.604). Compared with the performance prior to optimization (R2 = 0.834, RMSE = 1.790, MAPE = 2.679%, MAD = 1.030), the accuracy of the model was improved by 12.2%. It is worth noting that this framework could accurately identify water bodies in different trophic states. Finally, based on this framework, we mapped the spatial distribution of TSI in Hongze Lake in different seasons from 2019 to 2020 and analyzed its variation characteristics. The framework can combine regional special feature factors influenced by a complex environment with S-2/MSI data to achieve an assessment accuracy of over 90% for TSI in sensitive waters and has strong applicability and robustness. [ABSTRACT FROM AUTHOR]

With the global issues of extreme climate and urbanization, the ecological security patterns (ESPs) in the Qinling Mountains are facing prominent challenges. As a crucial ecological barrier in China, understanding the characteristics of ESPs in the Qinling Mountains is vital for achieving sustainable development. This study focuses on Yangxian and employs methods such as machine learning (ML), remote sensing (RS), geographic information systems (GISs), analytic hierarchy process and principal component analysis (AHP–PCA), and the minimum cumulative resistance (MCR) model to construct an ecological security network based on multi-factor ecological sensitivity (ES) and conduct quantitative spatial analysis. The results demonstrate that the AHP–PCA method based on ML overcomes the limitations of the single-weighting method. The ESPs of Yangxian were established, consisting of 21 main and secondary ecological sources with an area of 592.81 km2 (18.55%), 41 main and secondary ecological corridors with a length of 738.85 km, and 33 ecological nodes. A coupling relationship among three dimensions was observed: comprehensive ecological sensitivity, ESPs, and administrative districts (ADs). Huangjinxia Town (1.43 in C5) and Huayang Town (7.28 in C4) likely have significant areas of ecological vulnerability, while Machang Town and Maoping Town are important in the ESPs. ADs focus on protection and management. The second corridor indicated high-quality construction, necessitating the implementation of strict protection policies in the study area. The innovation lies in the utilization of quantitative analysis methods, such as ML and RS technologies, to construct an ecological spatial pattern planning model and propose a new perspective for the quantitative analysis of ecological space. This study provides a quantitative foundation for urban and rural ecological spatial planning in Yangxian and will help facilitate the sustainable development of ecological planning in the Qinling region. [ABSTRACT FROM AUTHOR]

PM2.5 refers to the total mass concentration of tiny particulates in the atmosphere near the surface, obtained by means of in situ observations and satellite remote sensing. Given the highly limited number of ground observation stations of inhomogeneous distribution and an ill-posed remote sensing approach, increasing efforts have been devoted to the application of machine-learning (ML) models to both ground and satellite data. A key satellite-derived parameter, aerosol optical thickness (AOD), has been most commonly used as a proxy of PM2.5, although their correlation is fraught with large uncertainties. A critical question that has been overlooked concerns how much AOD helps to improve the retrieval of PM2.5 relative to its uncertainty incurred concurrently. The question is addressed here by taking advantage of high-density PM2.5 stations in eastern China to evaluate the contributions of AOD, determined as the difference in the accuracy of PM2.5 retrievals with and without AOD for varying densities of PM2.5 stations, using four popular ML models (i.e., Random Forest, Extra-trees, XGBoost, and LightGBM). Our results reveal that as the density of monitoring stations decreases, both the feature importance and permutation importance of satellite AOD demonstrate a consistent upward trend (p < 0.05). Furthermore, the ML models without AOD exhibit faster declines in overall accuracy and predictive ability compared with the models with AOD assessed using the sample-based and station-based (spatial) independent cross-validation approaches. Overall, a 10% reduction in the number of stations results in an increase of 0.7–1.2% and 0.6–1.2% in uncertainty in estimated and predicted accuracies, respectively. These findings attest to the indispensable role of satellite AOD in the PM2.5 retrieval process through ML because it can significantly mitigate the negative impact of the sparse distribution of monitoring sites. This role becomes more important as the number of PM2.5 stations decreases. [ABSTRACT FROM AUTHOR]

UAV multispectral technology is used to obtain leaf area index (LAI) and aboveground biomass (AGB) information on Cinnamomum camphora (C. camphora) and to diagnose the growth condition of Cinnamomum camphora dwarf forests in a timely and rapid manner, which helps improve the precision management of Cinnamomum camphora dwarf forests. Multispectral remote sensing images provide large-area plant spectral information, which can provide a detailed quantitative assessment of LAI, AGB and other plant physicochemical parameters. They are very effective tools for assessing and analyzing plant health. In this study, the Cinnamomum camphora dwarf forest in the red soil area of south China is taken as the research object. Remote sensing images of Cinnamomum camphora dwarf forest canopy are obtained by the multispectral camera of an unmanned aerial vehicle (UAV). Extreme gradient boosting (XGBoost), gradient boosting decision tree (GBDT), random forest (RF), radial basis function neural network (RBFNN) and support vector regression (SVR) algorithms are used to study the correlation and estimation accuracy between the original band reflectance, spectral indices and LAI and AGB of Cinnamomum camphora. The results of this study showed the following: (1) The accuracy of model estimation based on RF is significantly different for different model inputs, while the other four models have small differences. (2) The accuracy of the XGBoost-based LAI model was the highest; with original band reflectance as the model input, the R2 of the model test set was 0.862, and the RMSE was 0.390. (3) The accuracy of the XGBoost-based AGB model was the highest; with spectral indices as the model input, the R2 of the model test set was 0.929, and the RMSE was 587.746 kg·hm−2. (4) The XGBoost model was the best model for the LAI and AGB estimation of Cinnamomum camphora, which was followed by GBDT, RF, RFNN, and SVR. This research result can provide a theoretical basis for monitoring a Cinnamomum camphora dwarf forest based on UAV multispectral technology and a reference for rapidly estimating Cinnamomum camphora growth parameters. [ABSTRACT FROM AUTHOR]

It is important to improve the accuracy of models estimating aboveground biomass (AGB) in large areas with complex geography and high forest heterogeneity. In this study, k-nearest neighbors (k-NN), gradient boosting machine (GBM), random forest (RF), quantile random forest (QRF), regularized random forest (RRF), and Bayesian regularization neural network (BRNN) machine learning algorithms were constructed to estimate the AGB of four forest types based on environmental factors and the variables selected by the Boruta algorithm in Yunnan Province and using integrated Landsat 8 OLI and Sentinel 2A images. The results showed that (1) DEM was the most important variable for estimating the AGB of coniferous forests, evergreen broadleaved forests, deciduous broadleaved forests, and mixed forests; while the vegetation index was the most important variable for estimating deciduous broadleaved forests, the climatic factors had a higher variable importance for estimating coniferous and mixed forests, and texture features and vegetation index had a higher variable importance for estimating evergreen broadleaved forests. (2) In terms of specific model performance for the four forest types, RRF was the best model both in estimating the AGB of coniferous forests and mixed forests; the R2 and RMSE for coniferous forests were 0.63 and 43.23 Mg ha−1, respectively, and the R2 and RMSE for mixed forests were 0.56 and 47.79 Mg ha−1, respectively. BRNN performed the best in estimating the AGB of evergreen broadleaved forests; the R2 was 0.53 and the RMSE was 68.16 Mg ha−1. QRF was the best in estimating the AGB of deciduous broadleaved forests, with R2 of 0.43 and RMSE of 45.09 Mg ha−1. (3) RRF was the best model for the four forest types according to the mean values, with R2 and RMSE of 0.503 and 52.335 Mg ha−1, respectively. In conclusion, different variables and suitable models should be considered when estimating the AGB of different forest types. This study could provide a reference for the estimation of forest AGB based on remote sensing in complex terrain areas with a high degree of forest heterogeneity. [ABSTRACT FROM AUTHOR]

Efficient and accurate landslide recognition is crucial for disaster prevention and post-disaster rescue efforts. However, compared to machine learning, deep learning approaches currently face challenges such as long model runtimes and inefficiency. To tackle these challenges, we proposed a novel knowledge distillation network based on Swin-Transformer (Distilled Swin-Transformer, DST) for landslide recognition. We created a new landslide sample database and combined nine landslide influencing factors (LIFs) with remote sensing images (RSIs) to evaluate the performance of DST. Our approach was tested in Zigui County, Hubei Province, China, and our quantitative evaluation showed that the combined RSIs with LIFs improved the performance of the landslide recognition model. Specifically, our model achieved an Overall Accuracy (OA), Precision, Recall, F1-Score (F1), and Kappa that were 0.8381%, 0.6988%, 0.9334%, 0.8301%, and 0.0125 higher, respectively, than when using only RSIs. Compared with the results of other neural networks, namely ResNet50, Swin-Transformer, and DeiT, our proposed deep learning model achieves the best OA (98.1717%), Precision (98.1672%), Recall (98.1667%), F1 (98.1615%), and Kappa (0.9766). DST has the lowest number of FLOPs, which is crucial for improving computational efficiency, especially in landslide recognition applications after geological disasters. Our model requires only 2.83 GFLOPs, which is the lowest among the four models and is 1.8242 GFLOPs, 1.741 GFLOPs, and 2.0284 GFLOPs less than ResNet, Swin, and DeiT, respectively. The proposed method has good applicability in rapid recognition scenarios after geological disasters. [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]

The issue of urban air quality in China has become increasingly significant due to industrialization and rapid urbanization. Although PM2.5 is the major air pollutant in most cities of northern China and has a direct negative impact on human health, there is a problem of under-representativeness at Chinese monitoring stations. In some cities, due to the relatively fewer national control stations and the fact that the stations are located closer to pollution sources, under the current assessment system, the monitoring data are not sufficient for the fairness of air quality assessment in different cities. In this article, the multispectral data of Landsat 8 data, air quality data, and meteorological data from ground monitoring stations have been integrated together and imported to different PM2.5-estimation models established based on the multi-layer back propagation neural network (MLBPN), support vector regression (SVR), and random forest (RF), respectively. According to the evaluation indices of R2, RMSE, and ME, the estimation model based on the MLBPN revealed the best PM2.5 estimation results and was therefore employed for the regional representativeness analysis in the study area of Xi'an, Shaanxi, China. The annual average PM2.5 concentration in the study area is depicted after error correction using Kriging interpolation, which can be further used to evaluate and analyze the representativeness of monitoring stations in Xi'an. By calculating the difference between the actual station annual average and the annual average of estimated PM2.5 concentration in the whole region, it can be found that the regional annual average value of PM2.5 in Xi'an is overestimated. To sum up, this article proposes a feasible method for the spatial positioning of the air quality monitoring stations to be established. [ABSTRACT FROM AUTHOR]

In recent years, remote sensing techniques such as satellite and drone-based imaging have been used to monitor Pine Wilt Disease (PWD), a widespread forest disease that causes the death of pine species. Researchers have explored the use of remote sensing imagery and deep learning algorithms to improve the accuracy of PWD detection at the single-tree level. This study introduces a novel framework for PWD detection that combines high-resolution RGB drone imagery with free-access Sentinel-2 satellite multi-spectral imagery. The proposed approach includes an PWD-infected tree detection model named YOLOv5-PWD and an effective data augmentation method. To evaluate the proposed framework, we collected data and created a dataset in Xianning City, China, consisting of object detection samples of infected trees at middle and late stages of PWD. Experimental results indicate that the YOLOv5-PWD detection model achieved 1.2% higher mAP compared to the original YOLOv5 model and a further improvement of 1.9% mAP was observed after applying our dataset augmentation method, which demonstrates the effectiveness and potential of the proposed framework for PWD detection. [ABSTRACT FROM AUTHOR]

Recognizing and classifying natural or artificial geo-objects under complex geo-scenes using remotely sensed data remains a significant challenge due to the heterogeneity in their spatial distribution and sampling bias. In this study, we propose a deep learning method of surface complexity analysis based on multiscale entropy. This method can be used to reduce sampling bias and preserve entropy-based invariance in learning for the semantic segmentation of land use and land cover (LULC) images. Our quantitative models effectively identified and extracted local surface complexity scores, demonstrating their broad applicability. We tested our method using the Gaofen-2 image dataset in mainland China and accurately estimated multiscale complexity. A downstream evaluation revealed that our approach achieved similar or better performance compared to several representative state-of-the-art deep learning methods. This highlights the innovative and significant contribution of our entropy-based complexity analysis and its applicability in improving LULC semantic segmentations through optimal stratified sampling and constrained optimization, which can also potentially be used to enhance semantic segmentation under complex geo-scenes using other machine learning methods. [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]

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]

Accurate and timely crop yield prediction over large spatial regions is critical to national food security and sustainable agricultural development. However, designing a robust model for crop yield prediction over a large spatial region remains challenging due to inadequate surveyed samples and an under-development of deep-learning frameworks. To tackle this issue, we integrated multi-source (remote sensing, weather, and soil properties) data into a dual-stream deep-learning neural network model for winter wheat in China's major planting regions. The model consists of two branches for robust feature learning: one for sequential data (remote sensing and weather series data) and the other for statical data (soil properties). The extracted features by both branches were aggregated through an adaptive fusion model to forecast the final wheat yield. We trained and tested the model by using official county-level statistics of historical winter wheat yields. The model achieved an average R2 of 0.79 and a root-mean-square error of 650.21 kg/ha, superior to the compared methods and outperforming traditional machine-learning methods. The dual-stream deep-learning neural network model provided decent in-season yield prediction, with an error of about 13% compared to official statistics about two months before harvest. By effectively extracting and aggregating features from multi-source datasets, the new approach provides a practical approach to predicting winter wheat yields at the county scale over large spatial regions. [ABSTRACT FROM AUTHOR]

Attempts have been made to incorporate remote sensing techniques and in situ observations for enhanced water quality assessments. Estimations of nonoptical indicators sensitive to water environment changes, however, have not been fully studied, mainly due to complex nonlinear relationships between the observed values and surface reflectance. In this study, we applied a novel deep learning approach driven by a range of spectral properties to retrieve 6-year changes in water quality variables, i.e., Chl-a, BOD, TN, CODMn, NH3-N, and TP, on a monthly basis between 2013 and 2018 at Dongping Lake, an impounded lake located in the Yellow River in China. Band arithmetic was used to compute 26 predictors from Landsat 8 OLI imagery for model inputs. The results showed generally strong agreement between in situ and ConvLSTM-derived lake variables, generating R2 of 0.92, 0.88, 0.84, 0.80, 0.83, and 0.77 for TN, NH3-N, CODMn, Chl-a, TP, and BOD, which suggest good performance of the developed model. We then used statistical analysis to identify the spatial and temporal heterogeneity. The framework established in this study has applications in effective water quality monitoring and serves as an alarming tool for water-environment management in the complex inland lake waters. [ABSTRACT FROM AUTHOR]

The accurate quantification of evapotranspiration (ET) is critical to the sustainable management of irrigated agriculture. In this study, we proposed a remote sensing data fusion method for predicting ET, coupling a surface energy balance system model with an enhanced spatial and temporal adaptive reflectance fusion model utilizing remote sensing inversion with satellite data from Landsat and MODIS. The method was tested for a case study with cotton fields under film-mulched drip irrigation (FMDI) in the Manas River Basin. Areas under FMDI were identified, and ET patterns were evaluated for a 21-year period spanning from 2000 to 2020. A field experiment, a regional survey, and data retrieved from the literature provided results demonstrating that the method allowed reliable estimation of ET distribution with simultaneously, relatively high spatial and temporal resolutions at both field and regional scales. ET was found to decline from upstream to downstream in the basin, with the difference gradually diminishing over time. Supported by the promotion of FMDI technology, the area under cotton production in the basin increased by an average of 4.9% annually. Limited irrigation quotas to farmers and, therefore, water application per area led to a decreasing ratio of relative water supply for potential ET and, thus, to a reduction in average actual ET of 7.5 mm year−1. The average ET in the basin declined to about 415.7 mm in 2020, 287.2 mm lower than its water demand. The dynamics of fused ET provide a reliable scientific basis for agricultural water resources planning and management and for the sustainable utilization of water and soil resources in the basin. The method, with simultaneously high temporal and spatial resolutions, should have both local and global practical potential. [ABSTRACT FROM AUTHOR]

Digital soil mapping has emerged as a new method to describe the spatial distribution of soils economically and efficiently. In this study, a lightweight soil organic matter (SOM) mapping method based on a deep residual network, which we call LSM-ResNet, is proposed to make accurate predictions with background covariates. ResNet not only integrates spatial background information around the observed environmental covariates, but also reduces problems such as information loss, which undermines the integrity of information and reduces prediction uncertainty. To train the model, rectified linear units, mean squared error, and adaptive momentum estimation were used as the activation function, loss/cost function, and optimizer, respectively. The method was tested with Landsat5, the meteorological data from WorldClim, and the 1602 sampling points set from Xinxiang, China. The performance of the proposed LSM-ResNet was compared to a traditional machine learning algorithm, the random forest (RF) algorithm, and a training set (80%) and a test set (20%) were created to test both models. The results showed that the LSM-ResNet (RMSE = 6.40, R2 = 0.51) model outperformed the RF model in both the roots mean square error (RMSE) and coefficient of determination (R2), and the training accuracy was significantly improved compared to RF (RMSE = 6.81, R2 = 0.46). The trained LSM-ResNet model was used for SOM prediction in Xinxiang, a district of plain terrain in China. The prediction maps can be deemed an accurate reflection of the spatial variability of the SOM distribution. [ABSTRACT FROM AUTHOR]

Meteorologically observed air temperature (Ta) is limited due to low density and uneven distribution that leads to uncertain accuracy. Therefore, remote sensing data have been widely used to estimate near-surface Ta on various temporal scales due to their spatially continuous characteristics. However, few studies have focused on instantaneous Ta when satellites overpass. This study aims to produce both daily and instantaneous Ta datasets at 1 km resolution for the Jingjinji area, China during 2018–2019, using machine learning methods based on remote sensing data, dense meteorological observation station data, and auxiliary data (such as elevation and normalized difference vegetation index). Newly released Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6 surface Downward Shortwave Radiation (DSR) was introduced to improve the accuracy of Ta estimation. Five machine learning algorithms were implemented and compared so that the optimal one could be selected. The random forest (RF) algorithm outperformed the others (such as decision tree, feedforward neural network, generalized linear model) and RF obtained the highest accuracy in model validation with a daily root mean square error (RMSE) of 1.29 °C, mean absolute error (MAE) of 0.94 °C, daytime instantaneous RMSE of 1.88 °C, MAE of 1.35 °C, nighttime instantaneous RMSE of 2.47 °C, and MAE of 1.83 °C. The corresponding R2 was 0.99 for daily average, 0.98 for daytime instantaneous, and 0.95 for nighttime instantaneous. Analysis showed that land surface temperature (LST) was the most important factor contributing to model accuracy, followed by solar declination and DSR, which implied that DSR should be prioritized when estimating Ta. Particularly, these results outperformed most models presented in previous studies. These findings suggested that RF could be used to estimate daily instantaneous Ta at unprecedented accuracy and temporal scale with proper training and very dense station data. The estimated dataset could be very useful for local climate and ecology studies, as well as for nature resources exploration. [ABSTRACT FROM AUTHOR]

Drought is an event of shortages in the water supply, whether atmospheric, surface water or ground water. Prolonged droughts have negative impacts on ecosystems, agriculture, society, and the economy. Although existing drought index products are widely utilized in drought monitoring, the coarse spatial resolution greatly limits their applications on regional or local scales. Machine learning driven by remote sensing observations offers an opportunity to monitor regional scale droughts. However, the limited time range of remote sensing observations such as vegetation index (VI) resulted in a substantial gap in generating high resolution drought index products before 2000. This study generated spatiotemporally continuous Standardized Precipitation Evapotranspiration Index (SPEI) data spanning from 1901–2018 in southwestern China by machine learning. It indicated that four Classification and Regression Tree (CART) approaches, decision trees (DT), random forest (RF), gradient boosted regression trees (GBRT) and extra trees (ET), can provide valid local drought information by downscaling the Estación Experimental de Aula Dei (EEAD) data. The in-situ SPEI dataset produced by the Penman–Monteith method was used as a benchmark to evaluate the temporal and spatial performance of the downscaled SPEI. In addition, the necessity of VI in SPEI downscaling was also assessed. The results showed that: (1) the ET-based product has the best performance (R2 = 0.889, MAE = 0.232, RMSE = 0.432); (2) the VI provides no significant improvement for SPEI re-construction; (3) topography exerts an obvious influence on the downscaling process, and (4) the downscaled SPEI shows more consistency with the in-situ SPEI compared with EEAD SPEI. The proposed method can be easily extended to other areas without in-situ data and enhance the ability of long-term drought monitoring. [ABSTRACT FROM AUTHOR]

Water is an imperative part of the Earth and an essential resource in human life and production. Under the effects of climate change and human activities, the spatial and temporal distribution of water bodies has been changing, and the shortage of water resources is becoming increasingly serious worldwide. Therefore, the monitoring of water bodies is indispensable. Remote sensing has the advantages of real time, wide coverage, and rich information and has become a brand-new technical means to quickly obtain water information. This study summarizes the current common methods of water extraction based on optical and radar images, including the threshold method, support vector machine, decision tree, object-oriented extraction, and deep learning, as well as the advantages and disadvantages of each method. These methods were applied to the Huai River Basin in China and Nam Co on the Qinghai-Tibet Plateau. The extraction results show that all the aforementioned approaches can obtain reliable results. Among them, the threshold segmentation method based on normalized difference water index is more robust than others. In the water extraction process, there are still many problems that restrict the accuracy of the results. In the future, researchers will continue to search for more automatic, extensive, and high-precision water extraction methods. [ABSTRACT FROM AUTHOR]

Focusing on water resources assessment in ungauged or sparse gauged areas, a comparative evaluation of areal precipitation was conducted by remote sensing data, limited gauged data, and a fusion of gauged data and remote sensing data based on machine learning. The artificial neural network (ANN) model was used to fuse the remote sensing precipitation and ground gauge precipitation. The correlation coefficient, root mean square deviation, relative deviation and consistency principle were used to evaluate the reliability of the remote sensing precipitation. The case study in the Qaidam Basin, northwest of China, shows that the precision of the original remote sensing precipitation product of Tropical Precipitation Measurement Satellite (TRMM)-3B42RT and TRMM-3B43 was 0.61, 72.25 mm, 36.51%, 27% and 0.70, 64.24 mm, 31.63%, 32%, respectively, comparing with gauged precipitation. The precision of corrected TRMM-3B42RT and TRMM-3B43 improved to 0.89, 37.51 mm, –0.08%, 41% and 0.91, 34.22 mm, 0.11%, 42%, respectively, which indicates that the data mining considering elevation, longitude and latitude as the main influencing factors of precipitation is efficient and effective. The evaluation of areal precipitation in the Qaidam Basin shows that the mean annual precipitation is 104.34 mm, 186.01 mm and 174.76 mm based on the gauge data, corrected TRMM-3B42RT and corrected TRMM-3B43. The results show many differences in the areal precipitation based on sparse gauge precipitation data and fusion remote sensing data. [ABSTRACT FROM AUTHOR]

Forest aboveground biomass (AGB) is of great significance since it represents large carbon storage and may reduce global climate change. However, there are still considerable uncertainties in forest AGB estimates, especially in rugged regions, due to the lack of effective algorithms to remove the effects of topography and the lack of comprehensive comparisons of methods used for estimation. Here, we systematically compare the performance of three sources of remote sensing data used in forest AGB estimation, along with three machine-learning algorithms using extensive field measurements (N = 1058) made in the Khingan Mountains of north-eastern China in 2008. The datasets used were obtained from the LiDAR-based Geoscience Laser Altimeter System onboard the Ice, Cloud, and land Elevation satellite (ICESat/GLAS), the optical-based Moderate Resolution Imaging Spectroradiometer (MODIS), and the SAR-based Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR). We show that terrain correction is effective for this mountainous study region and that the combination of terrain-corrected GLAS and PALSAR features with Random Forest regression produces the best results at the plot scale. Including further MODIS-based features added little power for prediction. Based upon the parsimonious data source combination, we created a map of AGB circa 2008 and its uncertainty, which yields a coefficient of determination (R2) of 0.82 and a root mean squared error of 16.84 Mg ha−1 when validated with field data. Forest AGB values in our study area were within the range 79.81 ± 16.00 Mg ha−1, ~25% larger than a previous, SAR-based, analysis. Our result provides a historic benchmark for regional carbon budget estimation. [ABSTRACT FROM AUTHOR]

The accurate monitoring of forest cover and its changes are essential for environmental change research, but current satellite products for forest coverage carry many uncertainties. This study used 30-m Landsat-8 data, and aggregated 1-m GaoFen-2 (GF-2) satellite images to construct the training samples and used multiple machine learning algorithms (MLAs) to estimate the fractional forest cover (FFC) in China's Three North Region (TNR). In this study, multiple MLAs were merged to construct stacked generalization (SG) models based on the idea of SG, and the performances of the MLAs in the FFC estimation were evaluated. The results of the 10-fold cross-validation showed that all non-linear algorithms had a good performance, with an R2 value of greater than 0.8 and a root-mean square error (RMSE) of less than 0.05. In the bagging ensemble, the random forest (RF) (R2 = 0.993, RMSE = 0.020) model performed the best and in the boosting ensemble, the light gradient boosted machine (LGBM) (R2 = 0.992, RMSE = 0.022) performed the best. Although the evaluation index of the RF is slightly better than that of the LGBM, the independent validation results show that the two models have similar performances. The model evaluation results of the independent datasets showed that, in the SG model, the performance of the SG(LGBM) (R2 = 0.991, RMSE = 0.034) was better than that of the single or non-ensemble model. Comparing the FFC estimates of our model with those of existing datasets showed that our model exhibited more forest spatial distribution details and higher accuracy in complex landscapes. Overall, in this study, the method of using high-resolution remote sensing (RS) images to extract samples for FFC estimation is feasible. Our results demonstrate the potential of the ensemble MLAs to map the FFC. The research results also show that among many MALs, the RF algorithm is the most suitable algorithm for estimating FFC, which provides a reference for future research. [ABSTRACT FROM AUTHOR]

China implemented an aggressive nationwide lockdown procedure immediately after the COVID-19 outbreak in January 2020. As China emerges from the impact of COVID-19 on national economic and industrial activities, it has become the site of a large-scale natural experiment to evaluate the impact of COVID-19 on regional air quality. However, ground measurements of fine particulate matters (PM2.5) concentrations do not offer comprehensive spatial coverage, especially in suburban and rural regions. In this study, we developed a machine learning method with satellite aerosol remote sensing data, meteorological fields and land use parameters as major predictor variables to estimate spatiotemporally resolved daily PM2.5 concentrations in China. Our study period consists of a reference semester (1 November 2018–30 April 2019) and a pandemic semester (1 November 2019–30 April 2020), with six modeling months in each semester. Each period was then divided into subperiod 1 (November and December), subperiod 2 (January and February) and subperiod 3 (March and April). The reference semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.79 (17.55 μg/m3) and the pandemic semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.83 (13.48 μg/m3) for daily PM2.5 predictions. Our prediction results showed high PM2.5 concentrations in the North China Plain, Yangtze River Delta, Sichuan Basin and Xinjiang Autonomous Region during the reference semester. PM2.5 levels were lowered by 4.8 μg/m3 during the pandemic semester compared to the reference semester and PM2.5 levels during subperiod 2 decreased most, by 18%. The southeast region was affected most by the COVID-19 outbreak with PM2.5 levels during subperiod 2 decreasing by 31%, followed by the Northern Yangtze River Delta (29%) and Pearl River Delta (24%). [ABSTRACT FROM AUTHOR]

Mapping new and old buildings are of great significance for understanding socio-economic development in rural areas. In recent years, deep neural networks have achieved remarkable building segmentation results in high-resolution remote sensing images. However, the scarce training data and the varying geographical environments have posed challenges for scalable building segmentation. This study proposes a novel framework based on Mask R-CNN, named Histogram Thresholding Mask Region-Based Convolutional Neural Network (HTMask R-CNN), to extract new and old rural buildings even when the label is scarce. The framework adopts the result of single-object instance segmentation from the orthodox Mask R-CNN. Further, it classifies the rural buildings into new and old ones based on a dynamic grayscale threshold inferred from the result of a two-object instance segmentation task where training data is scarce. We found that the framework can extract more buildings and achieve a much higher mean Average Precision (mAP) than the orthodox Mask R-CNN model. We tested the novel framework's performance with increasing training data and found that it converged even when the training samples were limited. This framework's main contribution is to allow scalable segmentation by using significantly fewer training samples than traditional machine learning practices. That makes mapping China's new and old rural buildings viable. [ABSTRACT FROM AUTHOR]

As an objective social phenomenon, poverty has accompanied the vicissitudes of human society, which is a chronic dilemma hindering human civilization. Remote sensing data, such as nighttime lights imagery, provides abundant poverty-related information that can be related to poverty. However, it may be insufficient to rely merely on nighttime lights data, because poverty is a comprehensive problem, and poverty identification may be affected by topography, especially in some developing countries or regions where agriculture accounts for a large proportion. Therefore, some geographical features may be necessary for supplements. With the support of the random forest machine learning method, we extracted 23 spatial features base on remote sensing including nighttime lights data and geographical data, and carried out the poverty identification in Guizhou Province, China, since 2012. Compared with the identifications using support vector machines and the artificial neural network, random forest showed a better accuracy. The results supported that nighttime lights and geographical features are better than those only by nighttime lights features. From 2012 to 2019, the identified poor counties in Guizhou Province showed obvious dynamic spatiotemporal characteristics. The number of poor counties has decreased consistently and contiguous poverty-stricken areas have fragmented; the number of poor counties in the northeast and southwest regions decreased faster than other areas. The reduction in poverty probability exhibited a pattern of spreading from the central and northern regions to the periphery parts. The poverty reduction was relatively slow in areas with large slope and large topographic relief. When poor counties are adjacent to more non-poor counties, they can get rid of poverty easier. This study provides a method for feature selection and recognition of poor counties by remote sensing images and offers new insights into poverty identification and regional sustainable development for other developing countries and areas. [ABSTRACT FROM AUTHOR]

Landslide susceptibility prediction (LSP) modeling is an important and challenging problem. Landslide features are generally uncorrelated or nonlinearly correlated, resulting in limited LSP performance when leveraging conventional machine learning models. In this study, a deep-learning-based model using the long short-term memory (LSTM) recurrent neural network and conditional random field (CRF) in cascade-parallel form was proposed for making LSPs based on remote sensing (RS) images and a geographic information system (GIS). The RS images are the main data sources of landslide-related environmental factors, and a GIS is used to analyze, store, and display spatial big data. The cascade-parallel LSTM-CRF consists of frequency ratio values of environmental factors in the input layers, cascade-parallel LSTM for feature extraction in the hidden layers, and cascade-parallel full connection for classification and CRF for landslide/non-landslide state modeling in the output layers. The cascade-parallel form of LSTM can extract features from different layers and merge them into concrete features. The CRF is used to calculate the energy relationship between two grid points, and the extracted features are further smoothed and optimized. As a case study, the cascade-parallel LSTM-CRF was applied to Shicheng County of Jiangxi Province in China. A total of 2709 landslide grid cells were recorded and 2709 non-landslide grid cells were randomly selected from the study area. The results show that, compared with existing main traditional machine learning algorithms, such as multilayer perception, logistic regression, and decision tree, the proposed cascade-parallel LSTM-CRF had a higher landslide prediction rate (positive predictive rate: 72.44%, negative predictive rate: 80%, total predictive rate: 75.67%). In conclusion, the proposed cascade-parallel LSTM-CRF is a novel data-driven deep learning model that overcomes the limitations of traditional machine learning algorithms and achieves promising results for making LSPs. [ABSTRACT FROM AUTHOR]