Search

In this study, 10 min and 2 km high-resolution blended fog data (HRBFD) were generated using grid visibility data (GVD) and data from a GK2A (GEO-KOMPSAT-2A) fog product (GKFP) in Korea. As the blending method, the decision tree method (DTM) was used to consider the different characteristics of the two-input data (categorical data and continuity data). The blending of the two datasets was performed according to the presence or absence of the input data and considered the spatial representation of the GVD and the accuracy of the GKFP. The quality of the HRBFD was evaluated through visual comparison using GVD, GKFP, and visible images of the GK2A. The HRBFD seems to have partly solved the problem of fog detection in areas where visibility meters are rare or absent through the detection of fog occurring in the sea or mountain areas. In addition, the critical problem of the GKFP, which has limitations in detecting fog occurring under clouds, has been mostly overcome. Using the DTM, we generated 23 fog cases of 10 min and 2 km HRBFD. The results confirmed that detailed spatiotemporal characteristics of fog in Korea can be analyzed if such HRBFD is generated for a long time. [ABSTRACT FROM AUTHOR]

For airborne radar, detecting a low–slow–small (LSS) target is a hot and challenging topic, which results from the rapidly increasing number of non-cooperative flying LSS targets becoming of widespread concern, and the low signal-to-clutter ratio (SCR) of LSS targets results in the targets being particularly easily overwhelmed by the clutter. In this paper, a novel light gradient boosting machine (LightGBM)-based LSS target detection algorithm for airborne radar is proposed. The proposed method, based on the current real-time clutter environment of the range cell to be detected, firstly designs a specific real-time space-time LSS target signal repository with special dimensions and structures. Then, the proposed method creatively designs a new fast-built real-time training feature dataset specifically for the LSS target and the current clutter, together with a series of unique data transformations, sample selection, data restructuring, feature extraction, and feature processing. Finally, the proposed method develops a unique machine learning-based LSS target detection classifier model for the designed training dataset, by fully excavating and utilizing the advantages of the ensemble decision trees-based LightGBM. Consequently, the pre-processed data in the range cell of interest are classified using the proposed algorithm, which achieves LSS target detection by evaluating the output results of the designed classifier. Compared with the traditional classical target detection methods, the proposed algorithm is capable of providing markedly superior performance for LSS target detection. With an appropriate computational time, the proposed algorithm attains the highest probability of detecting LSS targets under the low SCR. The simulation outcomes and detection results with the experimental data are employed to validate the effectiveness and merits of the proposed algorithm. [ABSTRACT FROM AUTHOR]

The detection of subsurface archaeological remains using a range of remote sensing methods poses several challenges. Recent studies regarding the detection of archaeological proxies like those of cropmarks highlight the complexity of the phenomenon. In this work, we present three different methods, and associated indices, for identifying stressed reflectance signatures indicating buried archaeological remains, based on a dataset of measured ground spectroradiometric reflectance. Several spectral profiles between the visible and near-infrared parts of the spectrum were taken in a controlled environment in Cyprus during 2011–2012 and are re-used in this study. The first two (spectral) methods are based on a suitable analysis of the spectral signatures in (1) the visible part of the spectrum, in particular in the neighborhood of 570 nm, and (2) the red edge part of the spectrum, in the neighborhood of 730 nm. Machine learning (decision trees) allows for the deduction of suitable wavelengths to focus on in order to formulate the proposed indices and the associated classification criteria (decision boundaries) that can enhance the detection probability of stressed vegetation. Noise in the signal is taken into account by simulating reflectance signatures perturbed by white noise. Applying decision tree classification on the ensemble of simulations and basic statistical analysis, we refine the formulation of the indices and criteria for the noisy signatures. The success rate of the proposed methods is over 90%. The third method rests on the estimation of vegetation/canopy reflectance parameters through inversion of the physical-based PROSAIL reflectance model and the associated classification through machine learning methods. The obtained results provide further insights into the formation of stress vegetation that occurred due to the presence of shallow buried archaeological remains, which are well aligned with physical-based models and existing empirical knowledge. To the best of the authors' knowledge, this is the first study demonstrating the usefulness of radiative transfer models such as PROSAIL for understanding the formation of cropmarks. Similar studies can support future research directions towards the development of regional remote sensing methods and algorithms if systematic observations are adequately dispersed in space and time. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) can accurately estimate the location and probability of landslides. An effective approach for precise LSM is crucial for minimizing casualties and damage. The existing LSM methods primarily rely on static indicators, such as geomorphology and hydrology, which are closely associated with geo-environmental conditions. However, landslide hazards are often characterized by significant surface deformation. The Small Baseline Subset-Interferometric Synthetic Aperture Radar (SBAS-InSAR) technology plays a pivotal role in detecting and characterizing surface deformation. This work endeavors to assess the accuracy of SBAS-InSAR coupled with ensemble learning for LSM. Within this research, the study area was Shiyan City, and 12 static evaluation factors were selected as input variables for the ensemble learning models to compute landslide susceptibility. The Random Forest (RF) model demonstrates superior accuracy compared to other ensemble learning models, including eXtreme Gradient Boosting, Logistic Regression, Gradient Boosting Decision Tree, and K-Nearest Neighbor. Furthermore, SBAS-InSAR was utilized to obtain surface deformation rates both in the vertical direction and along the line of sight of the satellite. The former is used as a dynamic characteristic factor, while the latter is combined with the evaluation results of the RF model to create a landslide susceptibility optimization matrix. Comparing the precision of two methods for refining LSM results, it was found that the method integrating static and dynamic factors produced a more rational and accurate landslide susceptibility map. [ABSTRACT FROM AUTHOR]

Current and upcoming Sun-Induced chlorophyll Fluorescence (SIF) satellite products (e.g., GOME, TROPOMI, OCO, FLEX) have medium-to-coarse spatial resolutions (i.e., 0.3–80 km) and integrate radiances from different sources into a single ground surface unit (i.e., pixel). However, intrapixel heterogeneity, i.e., different soil and vegetation fractional cover and/or different chlorophyll content or vegetation structure in a fluorescence pixel, increases the challenge in retrieving and quantifying SIF. High spatial resolution Sentinel-2 (S2) data (20 m) can be used to better characterize the intrapixel heterogeneity of SIF and potentially extend the application of satellite-derived SIF to heterogeneous areas. In the context of the COST Action Optical synergies for spatiotemporal SENsing of Scalable ECOphysiological traits (SENSECO), in which this study was conducted, we proposed direct (i.e., spatial heterogeneity coefficient, standard deviation, normalized entropy, ensemble decision trees) and patch mosaic (i.e., local Moran's I) approaches to characterize the spatial heterogeneity of SIF collected at 760 and 687 nm (SIF760 and SIF687, respectively) and to correlate it with the spatial heterogeneity of selected S2 derivatives. We used HyPlant airborne imagery acquired over an agricultural area in Braccagni (Italy) to emulate S2-like top-of-the-canopy reflectance and SIF imagery at different spatial resolutions (i.e., 300, 20, and 5 m). The ensemble decision trees method characterized FLEX intrapixel heterogeneity best (R2 > 0.9 for all predictors with respect to SIF760 and SIF687). Nevertheless, the standard deviation and spatial heterogeneity coefficient using k-means clustering scene classification also provided acceptable results. In particular, the near-infrared reflectance of terrestrial vegetation (NIRv) index accounted for most of the spatial heterogeneity of SIF760 in all applied methods (R2 = 0.76 with the standard deviation method; R2 = 0.63 with the spatial heterogeneity coefficient method using a scene classification map with 15 classes). The models developed for SIF687 did not perform as well as those for SIF760, possibly due to the uncertainties in fluorescence retrieval at 687 nm and the low signal-to-noise ratio in the red spectral region. Our study shows the potential of the proposed methods to be implemented as part of the FLEX ground segment processing chain to quantify the intrapixel heterogeneity of a FLEX pixel and/or as a quality flag to determine the reliability of the retrieved fluorescence. [ABSTRACT FROM AUTHOR]

Lithology classification is important in mineral resource exploration, engineering geological exploration, and disaster monitoring. Traditional laboratory methods for the qualitative analysis of rocks are limited by sampling conditions and analytical techniques, resulting in high costs, low efficiency, and the inability to quickly obtain large-scale geological information. Hyperspectral remote sensing technology can classify and identify lithology using the spectral characteristics of rock, and is characterized by fast detection, large coverage area, and environmental friendliness, which provide the application potential for lithological mapping at a large regional scale. In this study, ZY1-02D hyperspectral images were used as data sources to construct a new two-layer extreme gradient boosting (XGBoost) lithology classification model based on the XGBoost decision tree and an improved greedy search algorithm. A total of 153 spectral bands of the preprocessed hyperspectral images were input into the first layer of the XGBoost model. Based on the tree traversal structural characteristics of the leaf nodes in the XGBoost model, three built-in XGBoost importance indexes were split and combined. The improved greedy search algorithm was used to extract the spectral band variables, which were imported into the second layer of the XGBoost model, and the bat algorithm was used to optimize the modeling parameters of XGBoost. The extraction model of rock classification information was constructed, and the classification map of regional surface rock types was drawn. Field verification was performed for the two-layer XGBoost rock classification model, and its accuracy and reliability were evaluated based on four indexes, namely, accuracy, precision, recall, and F1 score. The results showed that the two-layer XGBoost model had a good lithological classification effect, robustness, and adaptability to small sample datasets. Compared with the traditional machine learning model, the two-layer XGBoost model shows superior performance. The accuracy, precision, recall, and F1 score of the verification set were 0.8343, 0.8406, 0.8350, and 0.8157, respectively. The variable extraction ability of the constructed two-layer XGBoost model was significantly improved. Compared with traditional feature selection methods, the GREED-GFC method, when applied to the two-layer XGBoost model, contributes to more stable rock classification performance and higher lithology prediction accuracy, and the smallest number of extracted features. The lithological distribution information identified by the model was in good agreement with the lithology information verified in the field. [ABSTRACT FROM AUTHOR]

Exploring the complex effects of landscape patterns on ecosystem services (ESs) has become increasingly important in offering scientific support for effective spatial planning and ecosystem management. However, there is a particular lack of research on the nonlinear effects of landscape patterns on ESs and scale dependence. Taking Huainan City (in China) as a case study, this study adopted the InVEST model to estimate four key ESs including carbon storage (CS), habitat quality (HQ), nitrogen export (NE), and water yield (WY). Then, we calculated the selected landscape metrics at multiple spatial scales. Furthermore, the gradient boosting decision tree (GBDT) model was developed to investigate the relative importance of landscape metrics in explaining ESs and their nonlinear interrelation. The results indicated that most of the selected landscape metrics were significantly correlated with ESs. The GBDT model, which can explore nonlinear relationships, performed better than the linear regression model in explaining the variations in ESs. The landscape-level metrics of the Shannon's diversity index (SHDI) and the contagion index (CONTAG) and the class-level metrics of the aggregation index (AI) and edge density (ED) were the most important variables that influenced ESs. The landscape metrics affected ESs within a certain range, and the nonlinear effects varied with scale. [ABSTRACT FROM AUTHOR]

Remote sensing image scene classification (RSISC), which aims to classify scene categories for remote sensing imagery, has broad applications in various fields. Recent deep learning (DL) successes have led to a new wave of RSISC applications; however, they lack explainability and trustworthiness. Here, we propose a bidirectional flow decision tree (BFDT) module to create a reliable RS scene classification framework. Our algorithm combines BFDT and Convolutional Neural Networks (CNNs) to make the decision process easily interpretable. First, we extract multilevel feature information from the pretrained CNN model, which provides the basis for constructing the subsequent hierarchical structure. Then the model uses the discriminative nature of scene features at different levels to gradually refine similar subsets and learn the interclass hierarchy. Meanwhile, the last fully connected layer embeds decision rules for the decision tree from the bottom up. Finally, the cascading softmax loss is used to train and learn the depth features based on the hierarchical structure formed by the tree structure that contains rich remote sensing information. We also discovered that superclass results can be obtained well for unseen classes due to its unique tree structure hierarchical property, which results in our model having a good generalization effect. The experimental results align with theoretical predictions using three popular datasets. Our proposed framework provides explainable results, leading to correctable and trustworthy approaches. [ABSTRACT FROM AUTHOR]

This study presents a new ensemble framework to predict landslide susceptibility by integrating decision trees (DTs) with the rotation forest (RF) ensemble technique. The proposed framework mainly includes four steps. First, training and validation sets are randomly selected according to historical landslide locations. Then, landslide conditioning factors are selected and screened by the gain ratio method. Next, several training subsets are produced from the training set and a series of trained DTs are obtained by using a DT as a base classifier couple with different training subsets. Finally, the resultant landslide susceptibility map is produced by combining all the DT classification results using the RF ensemble technique. Experimental results demonstrate that the performance of all the DTs can be effectively improved by integrating them with the RF ensemble technique. Specifically, the proposed ensemble methods achieved the predictive values of 0.012–0.121 higher than the DTs in terms of area under the curve (AUC). Furthermore, the proposed ensemble methods are better than the most popular ensemble methods with the predictive values of 0.005–0.083 in terms of AUC. Therefore, the proposed ensemble framework is effective to further improve the spatial prediction of landslides. [ABSTRACT FROM AUTHOR]

Forest fires are major disturbances in forest ecosystems. The rapid detection of the spatial and temporal characteristics of fires is essential for formulating targeted post-fire vegetation restoration measures and assessing fire-induced carbon emissions. We propose an accurate and efficient framework for extracting the spatiotemporal characteristics of fires using vegetation change tracker (VCT) products and the Google Earth Engine (GEE) platform. The VCT was used to extract areas of persistent forest and forest disturbance patches from Landsat images of Xichang and Muli, Liangshan prefecture, Sichuan province in southwestern China and Huma, Heilongjiang province, in northeastern China. All available Landsat images in the GEE platform in a year were normalized using the VCT-derived persisting forest mask to derive three standardized vegetation indices (normalized burn ratio (NBRr), normalized difference moisture index (NDMIr), and normalized difference vegetation index (NDVIr)). Historical forest disturbance events in Xichang were used to train two decision trees using the C4.5 data mining tool. The differenced NBRr, NDMIr, and NDVIr (dNBRr, dNDMIr, and dNDVIr) were obtained by calculating the difference in the index values between two temporally adjacent images. The occurrence time of disturbance events were extracted using the thresholds identified by decision tree 1. The use of all available images in GEE narrowed the disturbance occurrence time down to 16 days. This period was extended if images were not available or had cloud cover. Fire disturbances were distinguished from other disturbances by comparing the dNBRr, dNDMIr, and dNDVIr values with the thresholds identified by decision tree 2. The results showed that the proposed framework performed well in three study areas. The temporal accuracy for detecting disturbances in the three areas was 94.33%, 90.33%, and 89.67%, the classification accuracy of fire and non-fire disturbances was 85.33%, 89.67%, and 83.67%, and the Kappa coefficients were 0.71, 0.74, and 0.67, respectively. The proposed framework enables the efficient and rapid extraction of the spatiotemporal characteristics of forest fire disturbances using frequent Landsat time-series data, GEE, and VCT products. The results can be used in forest fire disturbance databases and to implement targeted post-disturbance vegetation restoration practices. [ABSTRACT FROM AUTHOR]

Landslides are a common geodynamic phenomenon that cause substantial life and property damage worldwide. In the present study, we developed models to evaluate landslide susceptibility in forest-covered areas in Lin'an, southeastern China using logistic regression (LR), decision tree (DT), and random forest (RF) techniques. In addition to conventional landslide-related natural and human disturbance factors, factors describing forest cover, including forest type (two plantations (hickory and bamboo) and four natural forests (conifer, hardwood, shrub, and moso bamboo) and understory vegetation conditions, were included as predictors. Model performance was evaluated based on true-positive rate, Kappa value, and area under the ROC curve using a 10-fold cross-validation method. All models exhibited good performance with measures of ≥0.70, although the LR model was relatively inferior. The key predictors were forest type, understory vegetation height (UVH), normalized differential vegetation index (NDVI) in summer, distance to road (DTRD), and maximum daily rainfall (MDR). Hickory plantations yielded the highest landslide probability, while conifer and hardwood forests had the lowest values. Bamboo plantations had probability results comparable to those of natural forests. Using the RF model, areas with a shorter UVH (<1.2 m), a lower NDVI (<0.70), a heavier MDR (>115 mm), or a shorter DTRD (<500 m) were predicted to be landslide-prone. Information on forest cover is essential for predicting landslides in areas with rich forest cover, and conversion from natural forests to plantations could increase landslide risk. Across the study areas, the northwestern part was the most landslide-prone. In terms of landslide prevention, the RF model-based map produced the most accurate predictions for the "very high" category of landslide. These results will help us better understand landslide occurrences in forest-covered areas and provide valuable information for governments in designing disaster mitigation. [ABSTRACT FROM AUTHOR]

Landslides, the second largest geological hazard after earthquakes, result in significant loss of life and property. Extracting landslide information quickly and accurately is the basis of landslide disaster prevention. Fengjie County, Chongqing, China, is a typical landslide-prone area in the Three Gorges Reservoir Area. In this study, we newly integrate Shapley Additive Explanation (SHAP) and Optuna (OPT) hyperparameter tuning into four basic machine learning algorithms: Gradient Boosting Decision Tree (GBDT), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Additive Boosting (AdaBoost). We construct four new models (SHAP-OPT-GBDT, SHAP-OPT-XGBoost, SHAP-OPT-LightGBM, and SHAP-OPT-AdaBoost) and apply the four new models to landslide extraction for the first time. Firstly, high-resolution remote sensing images were preprocessed, landslide and non-landslide samples were constructed, and an initial feature set with 48 features was built. Secondly, SHAP was used to select features with significant contributions, and the important features were selected. Finally, Optuna, the Bayesian optimization technique, was utilized to automatically select the basic models' best hyperparameters. The experimental results show that the accuracy (ACC) of these four SHAP-OPT models was above 92% and the training time was less than 1.3 s using mediocre computational hardware. Furthermore, SHAP-OPT-XGBoost achieved the highest accuracy (96.26%). Landslide distribution information in Fengjie County from 2013 to 2020 can be extracted by SHAP-OPT-XGBoost accurately and quickly. [ABSTRACT FROM AUTHOR]

Root-zone soil moisture exerts a fundamental control on vegetation, energy balance, and the carbon cycle in Arctic ecosystems, but it is still not well understood in vast, remote, and understudied regions of discontinuous permafrost. The root-zone soil moisture product (30 m resolution) used in this analysis was retrieved from a time-series P-Band (420–440 MHz) synthetic aperture radar (SAR) backscatter observations (August 2017 & October 2017). While similar approaches have been taken to retrieve surface (0 cm to 5 cm) soil moisture from L-Band (1.2 GHz) SAR backscatter, this is one of the first known attempts at reaching the root-zone in permafrost regions. Here, we analyze secondary factors (excluding primary factors, such as precipitation) controlling summer (August) soil moisture at depths of 6 cm, 12 cm, and 20 cm over a 4500 km2 area on the Seward Peninsula of Alaska. Using a random forest model, we quantify the impact of topography, vegetation, and meteorological factors on soil moisture distributions. In developing the random forest model, we explore a variety of feature scales (30 m, 60 m, 90 m, 120 m, 180 m, and 240 m), tune hyperparameters (the structure of individual decision trees making up the ensemble including the number and depth of trees), and perform the final feature selection using cross-validated recursive feature elimination. Results suggest that root-zone soil moisture on the Seward Peninsula is primarily controlled by vegetation at 6 cm, but deeper in the soil column topography and meteorological factors, such as predominant winter wind direction and summer insolation, play a larger role. The random forest model accounts for 40% to 60% of the variation observed (R2 = 0.44 at 6 cm, R2 = 0.52 at 12 cm, R2 = 0.58 at 20 cm). These results indicate that vegetation is the dominant control on soil moisture shallow in the soil column, but the impact of vegetation does not extend to deeper layers retrieved from P-Band SAR backscatter. [ABSTRACT FROM AUTHOR]

Open access to SAR data from the Sentinel 1 missions allows analyses of long-term ground surface changes. The current data-acquisition frequency of 12 days facilitates the continuous monitoring of phenomena such as volcanic and tectonic activity or mining-related deformations. SAR data are increasingly also used as input data in forecasting phenomena on the basis of machine learning. This article presents the possibility of using selected machine learning algorithms in forecasting the influence of underground mining activity on the ground surface. The study was performed for a mining protective area with a surface of over 500 km 2 and located in western Poland. The ground surface displacements were calculated for the period from November 2014 to July 2021, with the use of the Small Baseline Subset (SBAS) method. The forecasts were performed for a total of 22 identified subsidence troughs. Each of the troughs was provided with two profiles, with a total of more than 10,000 identified points. The selected algorithms served to prepare 180-day displacement forecasts. The best results (significantly better than the baseline) were obtained with the ARIMA and Holt models. Linear models also provided better results than the baseline and their performance was very good at up to 2 months forecasting. Tree-based models including their sophisticated ensemble versions: bagging (Random Forest, Extra Trees) and boosting (XGBoost, LightGBM, CatBoost, Gradient Boosting, Hist Gradient Boosting) cannot be used for this type of predictions since Decision Trees are not able to extrapolate and thus are not a valid stand-alone tool for forecasting in this type of problems. A combination of satellite remote sensing data and machine learning facilitated both the simultaneous quasi-permanent monitoring of ground surface displacements and their forecasting in a relatively long time period. [ABSTRACT FROM AUTHOR]

Landslides are a dangerous natural hazard that can critically harm road infrastructure in mountainous places, resulting in significant damage and fatalities. The primary purpose of this study was to assess the efficacy of three machine learning algorithms (MLAs) for landslide susceptibility mapping including random forest (RF), decision tree (DT), and support vector machine (SVM). We selected a case study region that is frequently affected by landslides, the important Kamyaran–Sarvabad road in the Kurdistan province of Iran. Altogether, 14 landslide evaluation factors were input into the MLAs including slope, aspect, elevation, river density, distance to river, distance to fault, fault density, distance to road, road density, land use, slope curvature, lithology, stream power index (SPI), and topographic wetness index (TWI). We identified 64 locations of landslides by field survey of which 70% were randomly employed for building and training the three MLAs while the remaining locations were used for validation. The area under the receiver operating characteristics (AUC) reached a value of 0.94 for the decision tree compared to 0.82 for the random forest, and 0.75 for support vector machines model. Thus, the decision tree model was most accurate in identifying the areas at risk for future landslides. The obtained results may inform geoscientists and those in decision-making roles for landslide management. [ABSTRACT FROM AUTHOR]

Geological models are essential components in various applications. To generate reliable realizations, the geostatistical method focuses on reproducing spatial structures from training images (TIs). Moreover, uncertainty plays an important role in Earth systems. It is beneficial for creating an ensemble of stochastic realizations with high diversity. In this work, we applied a pattern classification distribution (PCD) method to quantitatively evaluate geostatistical modeling. First, we proposed a correlation-driven template method to capture geological patterns. According to the spatial dependency of the TI, region growing and elbow-point detection were launched to create an adaptive template. Second, a combination of clustering and classification was suggested to characterize geological realizations. Aiming at simplifying parameter specification, the program employed hierarchical clustering and decision tree to categorize geological structures. Third, we designed a stacking framework to develop the multi-grid analysis. The contribution of each grid was calculated based on the morphological characteristics of TI. Our program was extensively examined by a channel model, a 2D nonstationary flume system, 2D subglacial bed topographic models in Antarctica, and 3D sandstone models. We activated various geostatistical programs to produce realizations. The experimental results indicated that PCD is capable of addressing multiple geological categories, continuous variables, and high-dimensional structures. [ABSTRACT FROM AUTHOR]

Woody plant encroachment in grassy ecosystems is a widely reported phenomenon associated with negative impacts on ecosystem functions. Most studies of this phenomenon have been carried out in arid and semi-arid grasslands. Therefore, studies in tropical regions, particularly savannas, which are composed of grassland and woodland mosaics, are needed. Our objective was to evaluate the accuracy of woody encroachment classification in the Brazilian Cerrado, a tropical savanna. We acquired dry and wet season unmanned aerial vehicle (UAV) images using RGB and multispectral cameras that were processed by the support vector machine (SVM), decision tree (DT), and random forest (RF) classifiers. We also compared two validation methods: the orthomosaic and in situ methods. We targeted two native woody species: Baccharis retusa and Trembleya parviflora. Identification of these two species was statistically (p < 0.05) most accurate in the wet season RGB images classified by the RF algorithm, with an overall accuracy (OA) of 92.7%. Relating to validation assessments, the in situ method was more susceptible to underfitting scenarios, especially using an RF classifier. The OA was higher in grassland than in woodland formations. Our results show that woody encroachment classification in a tropical savanna is possible using UAV images and field surveys and is suggested to be conducted during the wet season. It is challenging to classify UAV images in highly diverse ecosystems such as the Cerrado; therefore, whenever possible, researchers should use multiple accuracy assessment methods. In the case of using in situ accuracy assessment, we suggest a minimum of 40 training samples per class and to use multiple classifiers (e.g., RF and DT). Our findings contribute to the generation of tools that optimize time and cost for the monitoring and management of woody encroachment in tropical savannas. [ABSTRACT FROM AUTHOR]

Landslides can cause severe damage to both the environment and society, and many statistical, index-based, and inventory-based methods have been developed to assess landslide susceptibility; however, it is still challenging to choose the most effective method and properly identify major driving factors for specific regions. Here, we applied four machine learning algorithms, adaptive boosting (AdaBoost), gradient-boosting decision tree (GBDT), multilayer perceptron (MLP), and random forest (RF), to predict the landslide susceptibility at 30 m spatial scale based on thirteen landslide conditioning factors (LCFs) in a landslide-vulnerable region. Based on inventory landslide points, the classification results were evaluated, and indicated that the performance of the RF (F1-score: 0.85, AUC: 0.92), AdaBoost (F1-score: 0.83, AUC: 0.91), and GBDT (F1-score: 0.83, AUC: 0.88) methods were significantly better than the MLP (F1-score: 0.76, AUC: 0.79) method. The results further indicated that the areas with high and very high landslide risk (susceptibility greater than 0.5) accounted for about 40% of the study region. All four models matched well and predicted similar spatial distribution patterns in landslide susceptibility, with the very high risk areas mostly distributed in the western and southeastern regions. Daoshi, Qingliangfeng, Jinnan, and Linglong towns have the highest landslide risk, with mean susceptibility levels greater than 0.5. The leading contributing factors to landslide susceptibility were slightly different for the four models; however, population density, distance to road, and relief amplitude were generally among the top leading factors for most towns. Our study provided significant information on the highly landslide-prone areas and the major contributing factors for decision-makers and policy planners, and suggested that different areas should take unique precautions to mitigate or avoid severe damage from landslide events. [ABSTRACT FROM AUTHOR]

This study aimed to explore and compare the application of current state-of-the-art machine learning techniques, including bagging (Bag) and rotation forest (RF), to assess landslide susceptibility with the base classifier best-first decision tree (BFT). The proposed two novel ensemble frameworks, BagBFT and RFBFT, and the base model BFT, were used to model landslide susceptibility in Zhashui County (China), which suffers from landslides. Firstly, we identified 169 landslides through field surveys and image interpretation. Then, a landslide inventory map was built. These 169 historical landslides were randomly classified into two groups: 70% for training data and 30% for validation data. Then, 15 landslide conditioning factors were considered for mapping landslide susceptibility. The three ensemble outputs were estimated with a receiver operating characteristic (ROC) curve and statistical tests, as well as a new approach, the improved frequency ratio accuracy. The areas under the ROC curve (AUCs) for the training data (success rate) of the three algorithms were 0.722 for BFT, 0.869 for BagBFT, and 0.895 for RFBFT. The AUCs for the validating groups (prediction rates) were 0.718, 0.834, and 0.872, respectively. The frequency ratio accuracy of the three models was 0.76163 for the BFT model, 0.92220 for the BagBFT model, and 0.92224 for the RFBFT model. Both BagBFT and RFBFT ensembles can improve the accuracy of the BFT base model, and RFBFT was relatively better. Therefore, the RFBFT model is the most effective approach for the accurate modeling of landslide susceptibility mapping (LSM). All three models can improve the identification of landslide-prone areas, enhance risk management ability, and afford more detailed information for land-use planning and policy setting. [ABSTRACT FROM AUTHOR]

Sea ice type classification is of great significance for the exploration of waterways, fisheries, and offshore operations in the Arctic. However, to date, there is no multiple remote sensing method to detect sea ice type in the Arctic. This study develops a multiple sea ice type algorithm using the HaiYang-2B Scatterometer (HY-2B SCA). First, the parameters most applicable to classify sea ice type are selected through feature extraction, and a stacking model is established for the first time, which integrates decision tree and image segmentation algorithms. Finally, multiple sea ice types are classified in the Arctic, comprising Nilas, Young Ice, First Year Ice, Old Ice, and Fast Ice. Comparing the results with the Ocean and Sea Ice Satellite Application Facility (OSI-SAF) Sea Ice Type dataset (SIT) indicates that the sea ice type classified by HY-2B SCA (Stacking-HY2B) is similar to OSI-SAF SIT with regard to the changing trends in extent of sea ice. We use the Copernicus Marine Environment Monitoring Service (CMEMS) high-resolution sea ice type data and EM-Bird ice thickness data to validate the result, and accuracies of 87% and 88% are obtained, respectively. This indicates that the algorithm in this work is comparable with the performance of OSI-SAF dataset, while providing information of multiple sea ice types. [ABSTRACT FROM AUTHOR]

Hurricanes, rapidly increasing in complexity and strength in a warmer world, are one of the worst natural disasters in the 21st century. Further studies integrating the changing hurricane features are thus crucial to aid in the prediction of major hurricanes. With this in mind, we present a new framework based on automated decision tree analysis, which has the capability to identify the most important cloud structural parameters from GOES imagery as predictors for hurricane intensification potential in the Atlantic and Pacific oceans. The proposed framework has been proved effective for predicting major hurricanes with an overall accuracy of 73% from 6 to 54 h in advance (both regions combined). [ABSTRACT FROM AUTHOR]

Soil salinization is a widespread and important environmental problem. We propose a high-precision remote sensing identification method for saline-alkaline areas using multi-source data, a method which is of some significance for improving ecological and environmental problems on a global scale which have been caused by soil salinization. Its principle is to identify saline-alkaline areas from remote sensing imagery by a decision tree model combining four spectral indices named NDSI34 (Normalized Difference Spectral Index of Band 3 and Band 4), NDSI25 (Normalized Difference Spectral Index of Band 2 and Band 5), NDSI237 (Normalized Difference Spectral Index of Band 3 and Band 4) and NDSInew (New Normalized Difference Salt Index) that can distinguish saline-alkaline areas from other features. In this method, the complementary information within the multi-source data is used to improve classification accuracy. The main steps of the method include multi-source data acquisition, adaptive feature fusion of multi-source data, feature identification and integrated expression of the saline-alkaline area from multi-source data, fine classification of the saline-alkaline area, and accuracy verification. Taking Minqin County, Gansu Province, China as the study area, we use the method to identify saline-alkaline areas based on GF-2, GF-6/WFV and DEM data. The results show that the overall accuracy of the method is 88.11%, which is 7.69% higher than that of the traditional methods, indicating that it could effectively identify the distribution of saline-alkaline areas, and thus provide a scientific technique for the quick identification of saline-alkaline areas in large regions. [ABSTRACT FROM AUTHOR]

Polar-orbiting satellites have been widely used for detecting sea fog because of their wide coverage and high spatial and spectral resolution. FengYun-3D (FY-3D) is a Chinese satellite that provides global sea fog observation. From January 2021 to October 2022, the backscatter and virtual file manager products from CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation) were used to label samples of different atmospheric conditions in FY-3D images, including clear sky, sea fog, low stratus, fog below low stratus, mid–high-level clouds, and fog below the mid–high-level clouds. A 13-dimensional feature matrix was constructed after extracting and analyzing the spectral and texture features of these samples. In order to detect daytime sea fog using a 13-dimensional feature matrix and CALIPSO sample labels, four supervised classification models were developed, including Decision Tree (DT), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Neural Network. The accuracy of each model was evaluated and compared using a 10-fold cross-validation procedure. The study found that the SVM, KNN, and Neural Network performed equally well in identifying low stratus, with 85% to 86% probability of detection (POD). As well as identifying the basic components of sea fog, the SVM model demonstrated the highest POD (93.8%), while the KNN had the lowest POD (92.4%). The study concludes that the SVM, KNN, and Neural Network can effectively distinguish sea fog from low stratus. The models, however, were less effective at detecting sub-cloud fog, with only 11.6% POD for fog below low stratus, and 57.4% POD for fog below mid–high-level clouds. In light of this, future research should focus on improving sub-cloud fog detection by considering cloud layers. [ABSTRACT FROM AUTHOR]

Weather radar research has produced numerous radar-based rainfall estimators based on climate, rainfall intensity, a variety of ground-truthing instruments and sensors (e.g., rain gauges, disdrometers), and techniques. Although each research direction gives improvement, their collective application in an operational sense still yields uncertainty in rainfall estimation at times. This study aims to explore the concept of implementing Machine Learning (ML) models in optimizing the radar-based rainfall estimations at the bin level from a group of estimator. The Canadian King City C-Band radar was used with a GEONOR T-200B rain gauge (a total of 263 sample points) to establish a group of polarimetric-based rainfall estimators (R(Z), R(Z, ZDR), R(KDP)). The estimators were used to train three ML models, namely Decision Tree, Random Forest, and Gradient Boost, to choose the optimal rainfall estimators based on radar variables (Z, ZDR, KDP). Data from the Canadian Exeter C-Band radar and a Texas Electronics TE525 tipping bucket gauge at a different location were used to verify the ML models and compare their results to the most commonly used Z-R relations. The verification process shows promising results for the ML models, specifically the Gradient Boost model. These encouraging results need to be further explored with more sample points to further refine the ML models. [ABSTRACT FROM AUTHOR]

Seagrass beds provide a range of ecosystem services but are at risk from anthropogenic pressures. While recent progress has been made, the distribution and condition of South Pacific seagrass is relatively poorly known and selecting an appropriate approach for mapping it is challenging. A variety of remote sensing tools are available for this purpose and here we develop a mapping toolbox and associated decision tree tailored to the South Pacific context. The decision tree considers the scale at which data are needed, the reason that monitoring is required, the finances available, technical skills of the monitoring team, data resolution, site safety/accessibility and whether seagrass is predominantly intertidal or subtidal. Satellite mapping is recommended for monitoring at the national and regional scale, with associated ground-reference data where possible but without if time and funds are limiting. At the local scale, satellite, remotely piloted aircraft (RPA), kites, underwater camera systems and in situ surveys are all recommended. In the special cases of community-based initiatives and emergency response monitoring, in situ or satellite/RPA are recommended, respectively. For other types of monitoring the primary driver is funding, with in situ, kite and satellite recommended when finances are limited and satellite, underwater camera, RPA or kites otherwise, dependent on specific circumstances. The tools can be used individually or in combination, though caution is recommended when combining tools due to data comparability. [ABSTRACT FROM AUTHOR]

Aquaculture plays a key role in achieving Sustainable Development Goals (SDGs), while it is difficult to accurately extract single-object aquaculture ponds (SOAPs) from medium-resolution remote sensing images (Mr-RSIs). Due to the limited spatial resolutions of Mr-RSIs, most studies have aimed to obtain aquaculture areas rather than SOAPs. This study proposed an object-oriented method for extracting SOAPs. We developed an iterative algorithm combining grayscale morphology and edge detection to segment water bodies and proposed a segmentation degree detection approach to select and edit potential SOAPs. Then a classification decision tree combining aquaculture knowledge about morphological, spectral, and spatial characteristics of SOAPs was constructed for object filter. We selected a 707.26 km2 study region in Sri Lanka and realized our method on Google Earth Engine (GEE). A 25.11 km 2 plot was chosen for verification, where 433 SOAPs were manually labeled from 0.5 m high-resolution RSIs. The results showed that our method could extract SOAPs with high accuracy. The relative error of total areas between extracted result and the labeled dataset was 1.13%. The MIoU of the proposed method was 0.6965, representing an improvement of between 0.1925 and 0.3268 over the comparative segmentation algorithms provided by GEE. The proposed method provides an available solution for extracting SOAPs over a large region and shows high spatiotemporal transferability and potential for identifying other objects. [ABSTRACT FROM AUTHOR]

Landslide is a natural disaster that seriously affects human life and social development. In this study, the characteristics and effectiveness of convolutional neural network (CNN) and conventional machine learning (ML) methods in a landslide susceptibility assessment (LSA) are compared. Six ML methods used in this study are Adaboost, multilayer perceptron neural network (MLP-NN), random forest (RF), naive Bayes, decision tree (DT), and gradient boosting decision tree (GBDT). First, the basic knowledge and structures of the CNN and ML methods, and the steps of the LSA are introduced. Then, 11 conditioning factors in three categories in the Hongxi River Basin, Pingwu County, Mianyang City, Sichuan Province are chosen to build the train, validation, and test samples. The CNN and ML models are constructed based on these samples. For comparison, indicator methods, statistical methods, and landslide susceptibility maps (LSMs) are used. The result shows that the CNN can obtain the highest accuracy (86.41%) and the highest AUC (0.9249) in the LSA. The statistical methods represented by the mean and variance of TP and TN perform more firmly on the possibility of landslide occurrence. Furthermore, the LSMs show that all models can successfully identify most of the landslide points, but for areas with a low frequency of landslides, some models are insufficient. The CNN model demonstrates better results in the recognition of the landslides' cluster region, this is also related to the convolution operation that takes the surrounding environment information into account. The higher accuracy and more concentrative possibility of CNN in LSA is of great significance for disaster prevention and mitigation, which can help the efficient use of human and material resources. Although CNN performs better than other methods, there are still some limitations, the identification of low-cluster landside areas can be enhanced by improving the CNN model. [ABSTRACT FROM AUTHOR]

An important task in the Electronic Support Measures (ESM) field is analyzing and recognizing radar signals. Feature extraction is one of the primary key elements of radar emitter recognition algorithms. Current research mainly finds statistical features such as the mean and variance of parameters from pluses as the input features of the classifier. However, data noise in intercepted pulse signals greatly interferes with the accuracy of the extracted statistical features and seriously affects the recognition rate of radar emitters. In this paper, we proposed a method of radar emitter recognition. We first clustered parameter sets to establish a set of frequent items and their corresponding clustering centers. Next, we concatenated the clustering centers of each frequent item into a feature vector associated with the data volume dimensions. Then, we built a decision tree classification model based on the feature vector, and finally we used the learned model for the recognition of unknown radar pulse trains. The simulation results show that the proposed method has better robustness when applied to a variety of data volumes and data noise scenarios compared with long short-term memory (LSTM) and support vector machine (SVM) methods. [ABSTRACT FROM AUTHOR]

Space weather describes varying conditions between the Sun and Earth that can degrade Global Navigation Satellite Systems (GNSS) operations. Thus, these effects should be precisely and timely corrected for accurate and reliable GNSS applications. That can be modeled with the Vertical Total Electron Content (VTEC) in the Earth's ionosphere. This study investigates different learning algorithms to approximate nonlinear space weather processes and forecast VTEC for 1 h and 24 h in the future for low-, mid- and high-latitude ionospheric grid points along the same longitude. VTEC models are developed using learning algorithms of Decision Tree and ensemble learning of Random Forest, Adaptive Boosting (AdaBoost), and eXtreme Gradient Boosting (XGBoost). Furthermore, ensemble models are combined into a single meta-model Voting Regressor. Models were trained, optimized, and validated with the time series cross-validation technique. Moreover, the relative importance of input variables to the VTEC forecast is estimated. The results show that the developed models perform well in both quiet and storm conditions, where multi-tree ensemble learning outperforms the single Decision Tree. In particular, the meta-estimator Voting Regressor provides mostly the lowest RMSE and the highest correlation coefficients as it averages predictions from different well-performing models. Furthermore, expanding the input dataset with time derivatives, moving averages, and daily differences, as well as modifying data, such as differencing, enhances the learning of space weather features, especially over a longer forecast horizon. [ABSTRACT FROM AUTHOR]

This study investigates the effectiveness of gradient boosting decision trees techniques in estimating mangrove above-ground biomass (AGB) at the Can Gio biosphere reserve (Vietnam). For this purpose, we employed a novel gradient-boosting regression technique called the extreme gradient boosting regression (XGBR) algorithm implemented and verified a mangrove AGB model using data from a field survey of 121 sampling plots conducted during the dry season. The dataset fuses the data of the Sentinel-2 multispectral instrument (MSI) and the dual polarimetric (HH, HV) data of ALOS-2 PALSAR-2. The performance standards of the proposed model (root-mean-square error (RMSE) and coefficient of determination (R2)) were compared with those of other machine learning techniques, namely gradient boosting regression (GBR), support vector regression (SVR), Gaussian process regression (GPR), and random forests regression (RFR). The XGBR model obtained a promising result with R2 = 0.805, RMSE = 28.13 Mg ha−1, and the model yielded the highest predictive performance among the five machine learning models. In the XGBR model, the estimated mangrove AGB ranged from 11 to 293 Mg ha−1 (average = 106.93 Mg ha−1). This work demonstrates that XGBR with the combined Sentinel-2 and ALOS-2 PALSAR-2 data can accurately estimate the mangrove AGB in the Can Gio biosphere reserve. The general applicability of the XGBR model combined with multiple sourced optical and SAR data should be further tested and compared in a large-scale study of forest AGBs in different geographical and climatic ecosystems. [ABSTRACT FROM AUTHOR]

A quantum machine is a human-made device whose collective motion follows the laws of quantum mechanics. Quantum machine learning (QML) is machine learning for quantum computers. The availability of quantum processors has led to practical applications of QML algorithms in the remote sensing field. Quantum machines can learn from fewer data than non-quantum machines, but because of their low processing speed, quantum machines cannot be applied to an image that has hundreds of thousands of pixels. Researchers around the world are exploring applications for QML and in this work, it is applied for pseudo-labelling of samples. Here, a PRISMA (PRecursore IperSpettrale della Missione Applicativa) hyperspectral dataset is prepared by quantum-based pseudo-labelling and 11 different machine learning algorithms viz., support vector machine (SVM), K-nearest neighbour (KNN), random forest (RF), light gradient boosting machine (LGBM), XGBoost, support vector classifier (SVC) + decision tree (DT), RF + SVC, RF + DT, XGBoost + SVC, XGBoost + DT, and XGBoost + RF with this dataset are evaluated. An accuracy of 86% was obtained for the classification of pine trees using the hybrid XGBoost + decision tree technique. [ABSTRACT FROM AUTHOR]

The detection of waste plastics in the marine and terrestrial environment using satellite Earth Observation data offers the possibility of large-scale mapping and reducing on-the-ground manual investigation. In addition, costs are kept to a minimum by utilizing free-to-access Copernicus data. A Machine Learning-based classifier was developed to run on Sentinel-1 and -2 data. In support of the training and validation, a dataset was created with terrestrial and aquatic cases by manually digitizing varying landcover classes alongside plastics under the sub-categories of greenhouses, plastic, tyres and waste sites. The trained classifier, including an Artificial Neural Network and post-processing decision tree, was verified using five locations encompassing these different forms of plastic. Although exact matchups are challenging to digitize, the performance has generated high accuracy statistics, and the resulting land cover classifications have been used to map the occurrence of plastic waste in aquatic and terrestrial environments. [ABSTRACT FROM AUTHOR]

Crop yield prediction before the harvest is crucial for food security, grain trade, and policy making. Previously, several machine learning methods have been applied to predict crop yield using different types of variables. In this study, we propose using the Geographically Weighted Random Forest Regression (GWRFR) approach to improve crop yield prediction at the county level in the US Corn Belt. We trained the GWRFR and five other popular machine learning algorithms (Multiple Linear Regression (MLR), Partial Least Square Regression (PLSR), Support Vector Regression (SVR), Decision Tree Regression (DTR), and Random Forest Regression (RFR)) with the following different sets of features: (1) full length features; (2) vegetation indices; (3) gross primary production (GPP); (4) climate data; and (5) soil data. We compared the results of the GWRFR with those of the other five models. The results show that the GWRFR with full length features (R2 = 0.90 and RMSE = 0.764 MT/ha) outperforms other machine learning algorithms. For individual categories of features such as GPP, vegetation indices, climate, and soil features, the GWRFR also outperforms other models. The Moran's I value of the residuals generated by GWRFR is smaller than that of other models, which shows that GWRFR can better address the spatial non-stationarity issue. The proposed method in this article can also be potentially used to improve yield prediction for other types of crops in other regions. [ABSTRACT FROM AUTHOR]

Gridded precipitation data with a high spatiotemporal resolution are of great importance for studies in hydrology, meteorology, and agronomy. Observational data from meteorological stations cannot accurately reflect the spatiotemporal distribution and variations of precipitation over a large area. Meanwhile, radar-derived precipitation data are restricted by low accuracy in areas of complex terrain and satellite-based precipitation data by low spatial resolution. Therefore, hourly precipitation models were employed to merge data from meteorological stations, Radar, and satellites; the models used five machine learning algorithms (XGBoost, gradient boosting decision tree, random forests (RF), LightGBM, and multiple linear regression (MLR)), as well as the CoKriging method. In the north of Guangdong Province, data of four heavy rainfall events in 2018 were processed with geographic data to obtain merged hourly precipitation data. The CoKriging method secured the best prediction of spatial distribution of accumulated precipitation, followed by the tree-based machine learning (ML) algorithms, and significantly, the prediction of MLR deviated from the actual pattern. All machine learning methods showed poor performances for timepoints with little precipitation during the heavy rainfall events. The tree-based ML method showed poor performance at some timepoints when precipitation was over-related to latitude, longitude, and distance from the coast. [ABSTRACT FROM AUTHOR]

An accurate classification of the precipitation type is important for forecasters, particularly in the winter season. We explored the capability of three supervised machine learning (ML) methods (decision tree, random forest, and support vector machine) to determine ground precipitation types (no precipitation, rain, mixed, and snow) for winter precipitation. We provided information on the particle characteristics within a radar sampling volume and the environmental condition to the ML model with the simultaneous use of polarimetric radar variables and thermodynamic variables. The ML algorithms were optimized using predictor selection and hyperparameter tuning in order to maximize the computational efficiency and accuracy. The random forest (RF) had the highest skill scores in all precipitation types and outperformed the operational scheme. The spatial distribution of the precipitation type from the RF model showed a good agreement with the surface observation. As a result, RF is recommended for the real-time precipitation type classification due to its easy implementation, computational efficiency, and satisfactory accuracy. In addition to the validation, this study confirmed the strong dependence of precipitation type on wet-bulb temperature and a 1000–850 hPa layer thickness. The results also suggested that the base heights of the radar echo are useful in discriminating non-precipitating area. [ABSTRACT FROM AUTHOR]

Because transmitting polarization can be an arbitrary elliptical wave, and theoretically, there are numerous possibilities of hybrid dual-pol modes, therefore, it is necessary to explore the feature recognition and classification ability of compact polarimetric (CP) parameters under different transmitting and receiving modes to different ground objects. In this paper, we first simulated, extracted, and analyzed the scattering intensity of two types of rice of six temporal CP synthetic aperture radar (SAR) data under three transmitting modes. Then, during different phenology stages, the optimal parameters for distinguishing transplanting hybrid rice (T–H) and direct-sown japonica rice (D–J) were acquired. Finally, a decision tree classification model was established based on the optimal parameters to carry out the fine classification of the two types of rice and to verify the results. The results showed that this strategy can obtain a high classification accuracy for the two types of rice with an overall classification accuracy of more than 95% and a kappa coefficient of more than 0.94. In addition, and importantly, we found that the CP parameters in the 1103 period (harvest stage) were the best CP parameters to distinguish the two types of rice, followed by the 0730 (seedling–elongation stage), 0612 (seedling stage), and 0916 (heading–flowering stage) periods. [ABSTRACT FROM AUTHOR]

Sea ice type classification is of great significance for the exploration of waterways, fisheries, and offshore operations in the Arctic. However, to date, there is no multiple remote sensing method to detect sea ice type in the Arctic. This study develops a multiple sea ice type algorithm using the HaiYang-2B Scatterometer (HY-2B SCA). First, the parameters most applicable to classify sea ice type are selected through feature extraction, and a stacking model is established for the first time, which integrates decision tree and image segmentation algorithms. Finally, multiple sea ice types are classified in the Arctic, comprising Nilas, Young Ice, First Year Ice, Old Ice, and Fast Ice. Comparing the results with the Ocean and Sea Ice Satellite Application Facility (OSI-SAF) Sea Ice Type dataset (SIT) indicates that the sea ice type classified by HY-2B SCA (Stacking-HY2B) is similar to OSI-SAF SIT with regard to the changing trends in extent of sea ice. We use the Copernicus Marine Environment Monitoring Service (CMEMS) high-resolution sea ice type data and EM-Bird ice thickness data to validate the result, and accuracies of 87% and 88% are obtained, respectively. This indicates that the algorithm in this work is comparable with the performance of OSI-SAF dataset, while providing information of multiple sea ice types.

Because traditional decision tree (DT) induction methods cannot efficiently take advantage of geospatial knowledge in the classification of remotely sensed imagery, several researchers have presented a co-location decision tree (CL-DT) method that combines the co-location technique with the traditional DT method. However, the CL-DT method only considers the Euclidean distance of neighborhood events, which cannot truly reflect the co-location relationship between instances for which there is a nonlinear distribution in a high-dimensional space. For this reason, this paper develops the theory and method for a maximum variance unfolding (MVU)-based CL-DT method (known as MVU-based CL-DT), which includes unfolding input data, unfolded distance calculations, MVU-based co-location rule generation, and MVU-based CL-DT generation. The proposed method has been validated by classifying remotely sensed imagery and is compared with four other types of methods, i.e., CL-DT, classification and regression tree (CART), random forests (RFs), and stacked auto-encoders (SAE), whose classification results are taken as “true values.” The experimental results demonstrate that: (1) the relative classification accuracies of the proposed method in three test areas are higher than CL-DT and CART, and are at the same level compared to RFs; and (2) the total number of nodes, the number of leaf nodes, and the number of levels are significantly decreased by the proposed method. The time taken for the data processing, decision tree generation, drawing of the tree, and generation of the rules are also shortened by the proposed method compared to CL-DT, CART, and RFs. [ABSTRACT FROM AUTHOR]

Phenology-based multi-index with the random forest (RF) algorithm can be used to overcome the shortcomings of traditional deforestation mapping that involves pixel-based classification, such as ISODATA or decision trees, and single images. The purpose of this study was to investigate methods to identify specific types of deforestation in North Korea, and to increase the accuracy of classification, using phenological characteristics extracted with multi-index and random forest algorithms. The mapping of deforestation area based on RF was carried out by merging phenology-based multi-indices (i.e., normalized difference vegetation index (NDVI), normalized difference water index (NDWI), and normalized difference soil index (NDSI)) derived from MODIS (Moderate Resolution Imaging Spectroradiometer) products and topographical variables. Our results showed overall classification accuracy of 89.38%, with corresponding kappa coefficients of 0.87. In particular, for forest and farm land categories with similar phenological characteristic (e.g., paddy, plateau vegetation, unstocked forest, hillside field), this approach improved the classification accuracy in comparison with pixel-based methods and other classes. The deforestation types were identified by incorporating point data from high-resolution imagery, outcomes of image classification, and slope data. Our study demonstrated that the proposed methodology could be used for deciding on the restoration priority and monitoring the expansion of deforestation areas. [ABSTRACT FROM AUTHOR]

Extracting 3D tree models based on terrestrial laser scanning (TLS) point clouds is a challenging task as trees are complex objects. Current TLS devices acquire high-density data that allow a detailed reconstruction of the tree topology. However, in dense forests a fully automatic reconstruction of trees is often limited by occlusion, wind influences and co-registration issues. In this paper, a semi-automatic method for extracting branching and stem structure based on equirectangular projections (range and intensity maps) is presented. The digitization of branches and stems is based on 2D maps, which enables simple navigation and raster processing. The modeling is performed for each viewpoint individually instead of using a registered point cloud. Previously reconstructed 2D-skeletons are transformed between the maps. Therefore, wind influences, orientation imperfections of scans and data gaps can be overcome. The method is applied to a TLS dataset acquired in a forest in Germany. In total 34 scans were carried out within a managed forest to measure approximately 90 spruce trees with minimal occlusions. The results demonstrate the feasibility of the presented approach to extract tree models with a high completeness and correctness and provide an excellent input for further modeling applications. [ABSTRACT FROM AUTHOR]

Hydrological signatures, i.e., statistical features of streamflow time series, are used to characterize the hydrology of a region. A relevant problem is the prediction of hydrological signatures in ungauged regions using the attributes obtained from remote sensing measurements at ungauged and gauged regions together with estimated hydrological signatures from gauged regions. The relevant framework is formulated as a regression problem, where the attributes are the predictor variables and the hydrological signatures are the dependent variables. Here we aim to provide probabilistic predictions of hydrological signatures using statistical boosting in a regression setting. We predict 12 hydrological signatures using 28 attributes in 667 basins in the contiguous US. We provide formal assessment of probabilistic predictions using quantile scores. We also exploit the statistical boosting properties with respect to the interpretability of derived models. It is shown that probabilistic predictions at quantile levels 2.5% and 97.5% using linear models as base learners exhibit better performance compared to more flexible boosting models that use both linear models and stumps (i.e., one-level decision trees). On the contrary, boosting models that use both linear models and stumps perform better than boosting with linear models when used for point predictions. Moreover, it is shown that climatic indices and topographic characteristics are the most important attributes for predicting hydrological signatures. [ABSTRACT FROM AUTHOR]

Multi-feature, especially multi-temporal, remote-sensing data have the potential to improve land cover classification accuracy. However, sometimes it is difficult to utilize all the features efficiently. To enhance classification performance based on multi-feature imagery, an improved rotation forest, combining Principal Component Analysis (PCA) and a boosting naïve Bayesian tree (NBTree), is proposed. First, feature extraction was carried out with PCA. The feature set was randomly split into several disjoint subsets; then, PCA was applied to each subset, and new training data for linear extracted features based on original training data were obtained. These steps were repeated several times. Second, based on the new training data, a boosting naïve Bayesian tree was constructed as the base classifier, which aims to achieve lower prediction error than a decision tree in the original rotation forest. At the classification phase, the improved rotation forest has two-layer voting. It first obtains several predictions through weighted voting in a boosting naïve Bayesian tree; then, the first-layer vote predicts by majority to obtain the final result. To examine the classification performance, the improved rotation forest was applied to multi-feature remote-sensing images, including MODIS Enhanced Vegetation Index (EVI) imagery time series, MODIS Surface Reflectance products and ancillary data in Shandong Province for 2013. The EVI imagery time series was preprocessed using harmonic analysis of time series (HANTS) to reduce the noise effects. The overall accuracy of the final classification result was 89.17%, and the Kappa coefficient was 0.71, which outperforms the original rotation forest and other classifier ensemble results, as well as the NASA land cover product. However, this new algorithm requires more computational time, meaning the efficiency needs to be further improved. Generally, the improved rotation forest has a potential advantage in remote-sensing classification. [ABSTRACT FROM AUTHOR]

In recent years, several powerful machine learning (ML) algorithms have been developed for image classification, especially those based on ensemble learning (EL). In particular, Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) methods have attracted researchers' attention in data science due to their superior results compared to other commonly used ML algorithms. Despite their popularity within the computer science community, they have not yet been well examined in detail in the field of Earth Observation (EO) for satellite image classification. As such, this study investigates the capability of different EL algorithms, generally known as bagging and boosting algorithms, including Adaptive Boosting (AdaBoost), Gradient Boosting Machine (GBM), XGBoost, LightGBM, and Random Forest (RF), for the classification of Remote Sensing (RS) data. In particular, different classification scenarios were designed to compare the performance of these algorithms on three different types of RS data, namely high-resolution multispectral, hyperspectral, and Polarimetric Synthetic Aperture Radar (PolSAR) data. Moreover, the Decision Tree (DT) single classifier, as a base classifier, is considered to evaluate the classification's accuracy. The experimental results demonstrated that the RF and XGBoost methods for the multispectral image, the LightGBM and XGBoost methods for hyperspectral data, and the XGBoost and RF algorithms for PolSAR data produced higher classification accuracies compared to other ML techniques. This demonstrates the great capability of the XGBoost method for the classification of different types of RS data. [ABSTRACT FROM AUTHOR]

This study investigated the usability of hyperspectral remote sensing for characterizing vegetation at hazardous waste sites. The specific objectives of this study were to: (1) estimate leaf-area-index (LAI) of the vegetation using three different methods (i.e., vegetation indices, red-edge positioning (REP), and machine learning regression trees), and (2) map the vegetation cover using machine learning decision trees based on either the scaled reflectance data or mixture tuned matched filtering (MTMF)-derived metrics and vegetation indices. HyMap airborne data (126 bands at 2.3 × 2.3 m spatial resolution), collected over the U.S. Department of Energy uranium processing sites near Monticello, Utah and Monument Valley, Arizona, were used. Grass and shrub species were mixed on an engineered disposal cell cover at the Monticello site while shrub species were dominant in the phytoremediation plantings at the Monument Valley site. Regression trees resulted in the best calibration performance of LAI estimation (R2 > 0.80. The use of REPs failed to accurately predict LAI (R2 < 0.2). The use of the MTMF-derived metrics (matched filter scores and infeasibility) and a range of vegetation indices in decision trees improved the vegetation mapping when compared to the decision tree classification using just the scaled reflectance. Results suggest that hyperspectral imagery are useful for characterizing biophysical characteristics (LAI) and vegetation cover on capped hazardous waste sites. However, it is believed that the vegetation mapping would benefit from the use of higher spatial resolution hyperspectral data due to the small size of many of the vegetation patches ( < 1 m) found on the sites.

Unprecedented rapid urbanization in China during the past several decades has been accompanied by extensive urban landscape renewal, which has increased the urban thermal environmental risk. However, landscape change is a sufficient but not necessary condition for land surface temperature (LST) variation. Many studies have merely highlighted the correlation between landscape pattern and LST, while neglecting to comprehensively present the spatiotemporal diversification of LST change under urban landscape renewal. Taking the main city of Shenzhen as a case study area, this study tracked the landscape renewal and LST variation for the period 1987-2015 using 49 Landsat images. A decision tree algorithm suitable for fast landscape type interpretation was developed to map the landscape renewal. Analytical tools that identified hot-cold spots, the gravity center, and transect of LST movement were adopted to identify LST changes. The results showed that the spatial variation of LST was not completely consistent with landscape change. The transformation from Green landscape to Grey landscape usually increased the LST within a median of 0.2 °C, while the reverse transformation did not obviously decrease the LST (the median was nearly 0 °C). The median of LST change from Blue landscape to Grey landscape was 1.0 °C, corresponding to 0.5 °C in the reverse transformation. The imbalance of LST change between the loss and gain of Green or Blue landscape indicates the importance of protecting natural space, where the benefits in terms of temperature mitigation cannot be completely substituted by reverse transformation. [ABSTRACT FROM AUTHOR]

The main requirement for preserving European hay meadows in good condition is through prerequisite cut management. However, monitoring these practices on a larger scale is very difficult. Our study analyses the use of MODIS vegetation indices products, namely EVI and NDVI, to discriminate cut and uncut meadows in Slovakia. We tested the added value of simple transformations of raw data series (seasonal statistics, first difference series), compared EVI and NDVI, and analyzed optimal periods, the number of scenes and the effect of smoothing on classification performance. The first difference series transformation saw substantial improvement in classification results. The best case NDVI series classification yielded overall accuracy of 85% with balanced rates of producer’s and user’s accuracies for both classes. EVI yielded slightly lower values, though not significantly different, although user accuracy of cut meadows achieved only 67%. Optimal periods for discriminating cut and uncut meadows lay between 16 May and 4 August, meaning only seven consecutive images are enough to accurately detect cutting in hay meadows. More importantly, the 16-day compositing period seemed to be enough for detection of cutting, which would be the time span that might be hopefully achieved by upcoming on-board HR sensors (e.g., Sentinel-2).

High-spatial-resolution precipitation data are of great significance in many applications, such as ecology, hydrology, and meteorology. Acquiring high-precision and high-resolution precipitation data in a large area is still a great challenge. In this study, a downscaling–merging scheme based on random forest and cokriging is presented to solve this problem. First, the enhanced decision tree model, which is based on random forest from machine learning algorithms, is used to reduce the spatial resolution of satellite daily precipitation data to 0.01°. The downscaled satellite-based daily precipitation is then merged with gauge observations using the cokriging method. The scheme is applied to downscale the Global Precipitation Measurement Mission (GPM) daily precipitation product over the upstream part of the Hanjiang Basin. The experimental results indicate that (1) the downscaling model based on random forest can correctly spatially downscale the GPM daily precipitation data, which retains the accuracy of the original GPM data and greatly improves their spatial details; (2) the GPM precipitation data can be downscaled on the seasonal scale; and (3) the merging method based on cokriging greatly improves the accuracy of the downscaled GPM daily precipitation data. This study provides an efficient scheme for generating high-resolution and high-quality daily precipitation data in a large area. [ABSTRACT FROM AUTHOR]

Decision tree classification is one of the most efficient methods for obtaining land use/land cover (LULC) information from remotely sensed imageries. However, traditional decision tree classification methods cannot effectively eliminate the influence of mixed pixels. This study aimed to integrate pixel unmixing and decision tree to improve LULC classification by removing mixed pixel influence. The abundance and minimum noise fraction (MNF) results that were obtained from mixed pixel decomposition were added to decision tree multi-features using a three-dimensional (3D) Terrain model, which was created using an image fusion digital elevation model (DEM), to select training samples (ROIs), and improve ROI separability. A Landsat-8 OLI image of the Yunlong Reservoir Basin in Kunming was used to test this proposed method. Study results showed that the Kappa coefficient and the overall accuracy of integrated pixel unmixing and decision tree method increased by 0.093% and 10%, respectively, as compared with the original decision tree method. This proposed method could effectively eliminate the influence of mixed pixels and improve the accuracy in complex LULC classifications. [ABSTRACT FROM AUTHOR]

Recently, studies have focused more attention on surface feature extraction using thermal infrared remote sensing (TIRS) as supplementary materials. Innovatively, in this paper, using three-date (winter, early spring, and end of spring) TIRS Band 10 images of Landsat-8, we proposed an empirical normalized difference of a seasonal brightness temperature index (NDSTI) for enhancing a built-up area based on the contrast heat emission seasonal response of a built-up area to solar radiation, and adopted a decision tree classification method for the rapidly accurate extraction of the built-up area. Four study areas, including one major experimental study area (Tangshan) and three verification areas (Minqin, Laizhou, and Yugan) in different climate zones, respectively, were used to empirically establish the overall strategy system, then we specified constrained conditions of this strategy. Moreover, we compared the NDSTI to the current built-up indices, respectively, for extracting the built-up area. The results showed that (1) the new index (NDSTI) exploited the seasonal thermal characteristic variation between the built-up area and other covers in the time series analysis, helping achieve more accurate built-up area extraction than other spectral indices; (2) this strategy could effectively realize rapid built-up area extraction with generally satisfied overall accuracy (over 80%), and was especially excellent in Tangshan and Laizhou; however, (3) it may be constrained by climate patterns and other surface characteristics, which need to be improved from the view of the results of Minqin and Yugan. In summary, the method developed in this study has the potential and advantage to extract the built-up area rapidly from the multi-seasonal thermal infrared remote sensing data. It could be an operative tool for long-term monitoring of built-up areas efficiently and for more applications of thermal infrared images in the future. [ABSTRACT FROM AUTHOR]

This paper intends to find a more cost-effective way for training oil spill classification systems by introducing active learning (AL) and exploring its potential, so that satisfying classifiers could be learned with reduced number of labeled samples. The dataset used has 143 oil spills and 124 look-alikes from 198 RADARSAT images covering the east and west coasts of Canada from 2004 to 2013. Six uncertainty-based active sample selecting (ACS) methods are designed to choose the most informative samples. A method for reducing information redundancy amongst the selected samples and a method with varying sample preference are considered. Four classifiers (k-nearest neighbor (KNN), support vector machine (SVM), linear discriminant analysis (LDA) and decision tree (DT)) are coupled with ACS methods to explore the interaction and possible preference between classifiers and ACS methods. Three kinds of measures are adopted to highlight different aspect of classification performance of these AL-boosted classifiers. Overall, AL proves its strong potential with 4% to 78% reduction on training samples in different settings. The SVM classifier shows to be the best one for using in the AL frame, with perfect performance evolving curves in different kinds of measures. The exploration and exploitation criterion can further improve the performance of the AL-boosted SVM classifier but not of the other classifiers. [ABSTRACT FROM AUTHOR]