Search

Flash-flood is considered to be one of the most destructive natural hazards in the world, which is difficult to accurately model and predict. The objective of the present research is to propose new ensembles of bivariate statistics and artificial intelligences and to introduce a comprehensive methodology for predicting flood susceptibility. The Putna river catchment of Romania is selected as a case study. In this regard, a total of six ensemble models were proposed and verified: Multilayer Perceptron neural network-Frequency Ratio (MLP-FR), Multilayer Perceptron neural network -Weights of Evidence (MLP-WOE), Rotation Forest-Frequency Ratio (RF-FR), Rotation Forest-Weights of Evidence (RF-WOE), Classification and Regression Tree-Frequency Ratio (CART-FR), and Classification and Regression Tree-Weights of Evidence (CART-WOE). In a first step, a geospatial database was created for the study area. This database includes 132 flood locations and 14 conditioning factors (lithology, slope angle, plan curvature, hydrological soil group, topographic wetness index, landuse, convergence index, elevation, distance from river, profile curvature, rainfall, aspect, stream power index, and topographic position index). In the next step, the Information Gain Ratio was used to evaluate the predictive ability of these factors. Subsequently, the database was used to train and validate the six ensemble models. The Receiver operating characteristic (ROC) curve, area under the curve (AUC), and statistical measures were used to evaluate the performance of the models. The results show that the prediction capability of the proposed ensemble models varied from 86.8% (the RF-FR model) to 93.9% (the RF-WOE model). These values indicate a high prediction performance for all the models. Therefore, we can state that the proposed ensemble models are new reliable tools which can be used for flood susceptibility modelling. Unlabelled Image • New six artificial intelligence ensemble models were proposed for food modelling. • 14 conditioning factors were considered as predictors for flood potential. • All ensemble models have high prediction performance. • MLP-WOE and RF-FR have the best prediction performance (>91%). • Slope is the most important factor for flood occurrence. [ABSTRACT FROM AUTHOR]

Abstract Forest fire danger map at regional scale is considered of utmost importance for local authority to efficiently allocate its resources to fire prevention measures and establish appropriate land use plans. This study aims at introduce a new machine learning method, named as DFP-MnBpAnn, based on Artificial Neural Network (Ann) with a novel hybrid training algorithm of Differential Flower Pollination (DFP) and mini-match backpropagation (MnBp) for spatial modeling of forest fire danger. Tropical forest of the Lam Dong province (Vietnam) was used as case study. To achieve this task, a Geographical Information System (GIS) database of the forest fire for the study area was established. Accordingly, DFP, as a metaheuristic method, is used to optimize the weights and structure of Ann to fit the GIS database at hand. Whereas, MnBp is employed periodically during the DFP-based optimization process, in which MnBp acts as a local search aiming to accelerate both the quality of the found solutions and the convergence rate. Experimental outcomes demonstrate that the proposed DFP-MnBpAnn model is superior to other benchmark methods with satisfactory prediction accuracy (Classification Accuracy Rate = 88.43%). This fact confirms that DFP-MnBpAnn is a promising alternative for the problem of large-scale forest fire danger mapping. Highlights • DFP-MnBpAnn is proposed for forest fire modeling. • DFP-MnBpAnn has high performance on the training and validation datasets. • DFP-MnBpAnn outperforms benchmark models i.e. PSO-NF, BpANN, SVM, LSSVM, and RF. [ABSTRACT FROM AUTHOR]

Abstract In many engineering projects, the soil compression coefficient is an important parameter used for estimating the settlement of soil layers. The common practice of determining the soil compression coefficient via the oedometer test is time-consuming and expensive. This study proposes a machine learning solution to replace the conventional tests used for obtaining the coefficient of soil compression. The new approach is an integration of the Multi-Layer Perceptron Neural Network (MLP Neural Nets) and Particle Swarm Optimization (PSO). These two computational intelligence methods work synergistically to establish a prediction model of soil compression coefficient. The PSO metaheuristic is employed to optimize the MLP Neural Nets model structure. To train and validate the proposed method, named as PSO-MLP Neural Nets, a dataset of 154 soil samples featuring 12 influencing factors has been collected from the geotechnical investigation process of a high-rise building project. Experimental results show that the proposed PSO-MLP Neural Nets has attained the most accurate prediction of the soil compression coefficient performance with RMSE = 0.0267, MAE = 0.0145, and R2 = 0.884. The result of the proposed model is significantly better than those obtained from other benchmark methods including the backpropagation neural network, the radial basis function neural network, the support vector regression, the random forest, and the Gaussian process. Based on the experimental results, the newly constructed PSO-MLP Neural Nets is very potential to be a new alternative to assist geotechnical engineers in design phase of civil engineering projects. [ABSTRACT FROM AUTHOR]

In this paper, a hybrid machine learning ensemble approach namely the Rotation Forest based Radial Basis Function (RFRBF) neural network is proposed for spatial prediction of landslides in part of the Himalayan area (India). The proposed approach is an integration of the Radial Basis Function (RBF) neural network classifier and Rotation Forest ensemble, which are state-of-the art machine learning algorithms for classification problems. For this purpose, a spatial database of the study area was established that consists of 930 landslide locations and fifteen influencing parameters (slope angle, road density, curvature, land use, distance to road, plan curvature, lineament density, distance to lineaments, rainfall, distance to river, profile curvature, elevation, slope aspect, river density, and soil type). Using the database, training and validation datasets were generated for constructing and validating the model. Performance of the model was assessed using the Receiver Operating Characteristic (ROC) curve, area under the ROC curve (AUC), statistical analysis methods, and the Chi square test. In addition, Logistic Regression (LR), Multi-layer Perceptron Neural Networks (MLP Neural Nets), Naïve Bayes (NB), and the hybrid model of Rotation Forest and Decision Trees (RFDT) were selected for comparison. The results show that the proposed RFRBF model has the highest prediction capability in comparison to the other models (LR, MLP Neural Nets, NB, and RFDT); therefore, the proposed RFRBF model is promising and should be used as an alternative technique for landslide susceptibility modeling. [ABSTRACT FROM AUTHOR]

The main objective of this study is to evaluate and compare the performance of landslide models using machine learning ensemble technique for landslide susceptibility assessment. This technique is a combination of ensemble methods (AdaBoost, Bagging, Dagging, MultiBoost, Rotation Forest, and Random SubSpace) and the base classifier of Multiple Perceptron Neural Networks (MLP Neural Nets). Ensemble techniques have been widely applied in other fields; however, their application is still rare in the assessment of landslide problems. Meanwhile, MLP Neural Nets, which is known as an artificial neural network, has been applied widely and efficiently in landslide problems. In the present study, landslide models of part Himalayan area (India) have been constructed and validated. For the evaluation and comparison of these models, receiver operating characteristic curve and Chi Square test methods have been applied. Overall, all landslide models performed well in landslide susuceptibility assessment but the performance of the MultiBoost model is the highest (AUC = 0.886), followed by Dagging model (AUC = 0.885), the Rotation Forest model (AUC = 0.882), the Bagging and Random SubSpace models (AUC = 0.881), and the AdaBoost model (AUC = 0.876), respectively. Moreover, machine learning ensemble models have improved significantly the performance of the base classifier of MLP Neural Nets (AUC = 0.874). Analysis of results indicates that landslide models using machine learning ensemble frameworks are promising methods which can be used as alternatives of individual base classifiers for landslide susceptibility assessment of other prone areas. [ABSTRACT FROM AUTHOR]

Abstract: This study investigates the potential application of artificial neural networks in landslide susceptibility mapping in the Hoa Binh province of Vietnam. A landslide inventory map of the study area was prepared by combining landslide locations investigated through three projects during the last 10years. Some recent landslide locations were identified based on SPOT satellite images, field surveys, and existing literature. The images have a spatial resolution of 2.5m. Ten landslide conditioning factors were utilized in the multilayer feed-forward neural network analysis: slope, aspect, relief amplitude, lithology, land use, soil type, rainfall, distance to roads, distance to rivers and distance to faults. Two back-propagation training algorithms, Levenberg–Marquardt and Bayesian regularization, were utilized to determine synoptic weights using a training dataset. Relative importance of each landslide conditioning factor was assessed using the above mentioned synoptic weights. The final connection weights obtained in the training phase were applied to the entire study area to produce landslide susceptibility indexes. The results were then imported to a GIS and landslide susceptibility maps were constructed. Landslide locations not used in the training phase were used to verify and compare the results of the landslide susceptibility maps. Finally, the two landslide susceptibility maps were validated using the prediction-rate method. Subsequently, areas under the prediction curves were assessed. The prediction accuracy of landslide susceptibility maps produced by the Bayesian regularization neural network and the Levenberg–Marquardt neural network were 90.3% and 86.1% respectively. These results indicate that the two models seem to have good predictive capability. The Bayesian regularization network model appears more robust and efficient than the Levenberg–Marquardt network model for landslide susceptibility mapping. [Copyright &y& Elsevier]

Abstract: The objective of this study is to investigate a potential application of the Adaptive Neuro-Fuzzy Inference System (ANFIS) and the Geographic Information System (GIS) as a relatively new approach for landslide susceptibility mapping in the Hoa Binh province of Vietnam. Firstly, a landslide inventory map with a total of 118 landslide locations was constructed from various sources. Then the landslide inventory was randomly split into a testing dataset 70% (82 landslide locations) for training the models and the remaining 30% (36 landslides locations) was used for validation purpose. Ten landslide conditioning factors such as slope, aspect, curvature, lithology, land use, soil type, rainfall, distance to roads, distance to rivers, and distance to faults were considered in the analysis. The hybrid learning algorithm and six different membership functions (Gaussmf, Gauss2mf, Gbellmf, Sigmf, Dsigmf, Psigmf) were applied to generate the landslide susceptibility maps. The validation dataset, which was not considered in the ANFIS modeling process, was used to validate the landslide susceptibility maps using the prediction rate method. The validation results showed that the area under the curve (AUC) for six ANFIS models vary from 0.739 to 0.848. It indicates that the prediction capability depends on the membership functions used in the ANFIS. The models with Sigmf (0.848) and Gaussmf (0.825) have shown the highest prediction capability. The results of this study show that landslide susceptibility mapping in the Hoa Binh province of Vietnam using the ANFIS approach is viable. As far as the performance of the ANFIS approach is concerned, the results appeared to be quite satisfactory, the zones determined on the map being zones of relative susceptibility. [Copyright &y& Elsevier]

Wildfire is an environmental hazard that has both local and global effects, causing economic losses and various severe environmental problems. Due to the adverse effects of climate changes and anthropogenic activities, wildfire is anticipated more frequent and extreme; therefore, new and more efficient tools for forest fire prevention and control are essential. This study proposes a new deep neural computing approach for spatial prediction of wildfire in a tropical climate area. For this purpose, deep neural computing (Deep-NC) with a structure of 3 hidden layers was proposed. The Rectified Linear Unit (ReLU) activation function was adopted to infer wildfire dangers from the input factors. To search and optimize the weights of the model, Stochastic Gradient Descent (SGD), Root Mean Square Propagation (RMSProp), Adaptive Moment Estimation (Adam), and Adadelta optimizers were employed. Also, this study has established a Geographic Information System (GIS) database for Gia Lai province (Vietnam) to train and verify the newly developed deep computing approach. The twelve ignition factors, namely, slope, aspect, elevation, curvature, land use, NVDI, NDWI, NDMI, temperature, wind speed, relative humidity, and rainfall, have been used to characterize the study area with respect to forest fire susceptibility. According to experimental results, the Adam optimized Deep-NC model delivered the highest predictive accuracy (AUC = 0.894, Kappa = 0.63). Accordingly, this model has been employed to establish a forest fire susceptibility map for Gia Lai province. The proposed Deep-NC model and the newly constructed forest fire susceptibility map can help local authorities in land use planning and hazard mitigation/prevention. [Display omitted] • The performance of Deep-NC for wildfire danger modeling was assessed • The ADAM optimized Deep-NC has the best and outperformed other models. • ADAM based Deep-NC is a tool for spatial prediction of wildfire danger. [ABSTRACT FROM AUTHOR]

Taking into account the exponential growth of the number of flash-floods events worldwide, the detection of areas prone to these natural hazards is one of the main activities taken in order to mitigate the negative effects of these risk phenomena. In the present paper, new modeling approaches, Alternating Decision Tree (ADT) integrated with IOE (ADT-IOE) and ADT integrated with AHP (ADT-AHP), were proposed for flash-flood susceptibility mapping across the Suha river catchment (Romania). Besides, two stand-alone methods, Index of Entropy (IOE) and Analytical Hierarchy Process (AHP), were also investigated. For this regard, 111 torrential points and 111 non-torrential points along with 8 flash-flood conditioning factors have been involved in the training process of the four models. The quality of the flash-flood models was checked by using the ROC Curve method, classification accuracy (CLA), and Kappa index. The result shows that the two ensemble models, the ADT-IOE (AUC = 0.972, CLC = 86.37%, Kappa statistics = 0.727) and the ADT-AHP (AUC = 0.926, CLA = 87.88%, Kappa statistics = 0.758), have high prediction performance and outperform the other models. Therefore, ADT-IOE and ADT-AHP are new and promising tools for flash-flood susceptibility modeling. Unlabelled Image • Flash-flood potential mapping was done through 2 stand-alone models and 2 ensembles. • AHP, IOE, ADT, and their ensembles were considered flash-flood susceptibility. • Torrential and non-torrential pixels were considered as dependent variables. • ADT-AHP and ADT-AHP ensemble models yield high prediction performance. • The ensemble models are new tools for flash-flood susceptibility mapping. [ABSTRACT FROM AUTHOR]

Unlabelled Image • The performance of DLNN was assessed for flood susceptibility mapping. • DLNN was compared with the MLP-NN and SVM in terms of their performance. • DLNN with ADAM optimization is robust and outperformed other models. • DLNN is a new promising tool for predicting flash flood in prone areas. This research proposes and evaluates a new approach for flash flood susceptibility mapping based on Deep Learning Neural Network (DLNN)) algorithm, with a case study at a high-frequency tropical storm area in the northwest mountainous region of Vietnam. Accordingly, a DLNN structure with 192 neurons in 3 hidden layers was proposed to construct an inference model that predicts different levels of susceptibility to flash flood. The Rectified Linear Unit (ReLU) and the sigmoid were selected as the activate function and the transfer function, respectively, whereas the Adaptive moment estimation (Adam) was used to update and optimize the weights of the DLNN. A database for the study area, which includes factors of elevation, slope, curvature, aspect, stream density, NDVI, soil type, lithology, and rainfall, was established to train and validate the proposed model. Feature selection was carried out for these factors using the Information gain ratio. The results show that the DLNN attains a good prediction accuracy with Classification Accuracy Rate = 92.05%, Positive Predictive Value = 94.55% and Negative Predictive Value = 89.55%. Compared to benchmarks, Multilayer Perceptron Neural Network and Support Vector Machine, the DLNN performs better; therefore, it could be concluded that the proposed hybridization of GIS and deep learning can be a promising tool to assist the government authorities and involving parties in flash flood mitigation and land-use planning. [ABSTRACT FROM AUTHOR]

• Proposed a hybrid method for landslide susceptibility mapping. • LSSVC is employed for pattern recognition. • Bat Algorithm is used to optimize the model performance. • A GIS database in Lang Son province (Vietnam) is employed. • The hybrid model has a good prediction performance (CAR = 90.44%). This study develops a machine learning method that hybridizes the Least Squares Support Vector Classification (LSSVC) and Bat Algorithm (BA), named as BA-LSSVC, for spatial prediction of shallow landslide. To construct and verify the hybrid method, a Geographic Information System (GIS) database for the study area of Lang Son province (Vietnam) has been employed. LSSVC is used to separate data samples in the GIS database into two categories of non-landslide (negative class) and landslide (positive class). The BA metaheuristic is employed to assist the LSSVC model selection process by fine-tuning its hyper-parameters: the regularization coefficient and the kernel function parameter. Experimental results point out that the hybrid BA-LSSVC can help to achieve a desired prediction with an accuracy rate of more than 90%. The performance of BA-LSSVC is also better than those of benchmark methods, including the Convolutional Neural Network, Relevance Vector Machine, Artificial Neural Network, and Logistic Regression. Hence, the newly developed model is a capable tool to assist local authority in landslide hazard mitigation and management. [ABSTRACT FROM AUTHOR]

Abstract Landslides represent a part of the cascade of geological hazards in a wide range of geo-environments. In this study, we aim to investigate and compare the performance of two state-of-the-art machine learning models, i.e., decision tree (DT) and random forest (RF) approaches to model the massive rainfall-triggered landslide occurrences in the Izu-Oshima Volcanic Island, Japan at a regional scale. At first, a landslide inventory map is prepared consisting of 44 landslide polygons (10,444 pixels) from aerial photo-interpretation and field surveys. To estimate the robustness of the models, we randomly adapted two different samples (S1 and S2), comprising of both positive and negative cells (70% of total landslides - 7293 pixels) for training and remaining (30%–3151 pixels) for validation. Twelve causative factors including altitude, slope angle, slope aspect, plan curvature, total curvature, compound topographic index, stream power index, distance to drainage network, drainage density, distance to geological boundaries, lithology and cumulative rainfall were selected as predictors to implement the landslide susceptibility model. The area under the receiver operating characteristics (ROC) curves (AUC) and other statistical signifiers were used to verify the model accuracies. The result shows that the DT and RF models achieved remarkable predictive performance (AUC > 0.9), producing near accurate susceptibility maps. The overall efficiency of RF (AUC = 0.956) is found significantly higher than the DT (AUC = 0.928) results. Additionally, we noticed that the performance of RF for modeling landslide susceptibility is very robust even though the training and validation samples are altered. Considering the performances, we suggest that both RF and DT models can be used in other similar non-eruption-related landslide studies in the tephra-deposited rich volcanoes, as they are capable of rapidly generating accurate and stable LSM maps for risk mitigation, management practices, and decision-making. Moreover, the RF-based model is promising and enough to be recommended as a method to map regional landslide susceptibility. Graphical abstract Unlabelled Image Highlights • Decision tree and random forest models applied to map landslide-prone areas in a volcanic Island. • Two sample set (S1, and S2) for computing the robustness of the model • LSM maps were compared using different assessment principles. • Random forest performs better on both samples with AUC > 0.9. [ABSTRACT FROM AUTHOR]

Landslide susceptibility assessment of Uttarakhand area of India has been done by applying five machine learning methods namely Support Vector Machines (SVM), Logistic Regression (LR), Fisher's Linear Discriminant Analysis (FLDA), Bayesian Network (BN), and Naïve Bayes (NB). Performance of these methods has been evaluated using the ROC curve and statistical index based methods. Analysis and comparison of the results show that all five landslide models performed well for landslide susceptibility assessment (AUC = 0.910–0.950). However, it has been observed that the SVM model (AUC = 0.950) has the best performance in comparison to other landslide models, followed by the LR model (AUC = 0.922), the FLDA model (AUC = 0.921), the BN model (AUC = 0.915), and the NB model (AUC = 0.910), respectively. [ABSTRACT FROM AUTHOR]

Marine gravity anomalies are essential data for determining coastal geoid , investigating tectonics and crustal structures, and offshore explorations. The objective of this study is to present a methodology for estimating marine gravity anomalies from CryoSat-2 and Saral/AltiKa satellite altimeter data for the Gulf of Tonkin of Vietnam with a high-resolution 2′ × 2′ grid. A total of 15,665 sea surface height (SSH) grid points, including derived from the Cryosat-2 (6842 grid points) and Saral/AltiKa (8823 grid points) satellite altimeter data were used. Then, the remove-restore technique and the crossover adjustment algorithm were used to remove the long-wavelength geoid height, the mean dynamic topography h MDT , and time-varying sea-surface topography h t . The residual geoid heights Δ N were used to determine the residual gravity anomalies δ g using the Least-Squares Collocation method, whereas the Earth Geopotential Model was employed to restore the long-wavelength gravity anomalies Δ g EIGEN . GPS/leveling and tidal gauge of 31 tidal stations were used for assessing and choosing the best Earth Geopotential Model and Mean Dynamic Topography models for the study area (EIGEN6C4 and DTU15MDT models). The accuracy of the final marine gravity anomaly result was assessed using 56,978 marine gravity points, which were distributed in the study area. The result showed that the standard deviation between the satellite-derived gravity anomalies and checked points is ± 3.36 mGal, indicating good accuracy. After improving with ship-measured gravity anomalies, the accuracy of satellite-derived marine gravity anomaly improves to ± 2.63 mGal. The results of this research are useful for geodetic and geophysical applications in the region. [ABSTRACT FROM AUTHOR]

Although estimating the uncertainty of models used for modelling nitrate contamination of groundwater is essential in groundwater management, it has been generally ignored. This issue motivates this research to explore the predictive uncertainty of machine-learning (ML) models in this field of study using two different residuals uncertainty methods: quantile regression (QR) and uncertainty estimation based on local errors and clustering (UNEEC). Prediction-interval coverage probability (PICP), the most important of the statistical measures of uncertainty, was used to evaluate uncertainty. Additionally, three state-of-the-art ML models including support vector machine (SVM), random forest (RF), and k -nearest neighbor (k NN) were selected to spatially model groundwater nitrate concentrations. The models were calibrated with nitrate concentrations from 80 wells (70% of the data) and then validated with nitrate concentrations from 34 wells (30% of the data). Both uncertainty and predictive performance criteria should be considered when comparing and selecting the best model. Results highlight that the k NN model is the best model because not only did it have the lowest uncertainty based on the PICP statistic in both the QR (0.94) and the UNEEC (in all clusters, 0.85–0.91) methods, but it also had predictive performance statistics (RMSE = 10.63, R2 = 0.71) that were relatively similar to RF (RMSE = 10.41, R2 = 0.72) and higher than SVM (RMSE = 13.28, R2 = 0.58). Determining the uncertainty of ML models used for spatially modelling groundwater-nitrate pollution enables managers to achieve better risk-based decision making and consequently increases the reliability and credibility of groundwater-nitrate predictions. Unlabelled Image • Predictive uncertainty of models was estimated using the QR and UNEEC methods. • Random Forest model had the lower uncertainty band width based on the both methods. • Groundwater nitrate (NO 3) concentrations were predicted using RF, SVM, and K NN. • Random Forest model outperformed other models in terms of predictive performance. • Hydraulic conductivity and elevation had the highest contribution to the modelling. [ABSTRACT FROM AUTHOR]

Gully erosion is considered as a severe environmental problem in many areas of the world which causes huge damages to agricultural lands and infrastructures (i.e. roads, buildings, and bridges); however, gully erosion modeling and prediction with high accuracy are still difficult due to the complex interactions of various factors. The objective of this research was to develop and introduce three new ensemble models, which were based on Complex Proportional Assessment of Alternatives (COPRAS), Logistic Regression (LR), Boosted Regression Tree (BRT), Random Forest (RF), and Frequency Ratio (FR) for spatial prediction of gully erosion with a case study at the Najafabad watershed (Iran). For this purpose, a total of 290 head-cut of gullies and 17 conditioning factors were collected and used to establish a geospatial database. Subsequently, FR was used to determine the spatial relationship between the conditioning factors and the head-cut of gullies, whereas RF, BRT, and LR were used to quantify the relative importance of these factors. In the next step, three ensemble gully erosion models, named COPRAS-FR-RF, COPRAS-FR-BRT, and COPRAS-FR-LR were developed and verified. The Success Rate Curve (SRC), and the Prediction Rate Curve (PRC) and their areas under the curves (AUC) were used to check the performance of the three proposed models. The result showed that Soil group, geomorphology, and drainage density factors played the key role on the occurrence of the gully erosion. All the three models have very high degree-of-fit and the prediction performance, the COPRAS-FR-RF model (AUC-SRC = 0.974 and AUC-PRC = 0.929), the COPRAS-FR-BRT model (AUC-SRC = 0.973 and AUC-PRC = 0.928), and the COPRAS-FR-LR model (AUC-SRC = 0.972 and AUC-PRC = 0.926); therefore, it is concluded that they are efficient and new powerful tools which could be used for predicting gully erosion in prone-areas. Unlabelled Image • RF, BRT, and LR are effective methods in determining the importance of the criteria for the COPRAS. • Three new ensemble models, COPRAS-FR-RF, COPRAS-FR-BRT, and COPRAS-FR-LR, were proposed for gully erosion modeling. • The proposed models provide >92% prediction accuracy. [ABSTRACT FROM AUTHOR]

In this study, we developed Different Artificial Intelligence (AI) models namely Artificial Neural Network (ANN), Adaptive Network based Fuzzy Inference System (ANFIS) and Support Vector Machine (SVM) for the prediction of Compression Coefficient of soil (Cc) which is one of the most important geotechnical parameters. A Monte Carlo approach was used for the sensitivity analysis of the AI models and input parameters. For the construction and validation of the models, 189 soft clayey soil samples were analyzed. In the models study, 13 input parameters: depth of sample, bulk density, plasticity index, moisture content, clay content, specific gravity, void ratio, liquid limit, dry density, porosity, plastic limit, degree of saturation, and liquidity index were used to obtain one output parameter "Cc". Validation of the models was done using statistical methods such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of determination (R2). Results of the model validation indicate that though performance of all the three models is good but SVM model is the best in the prediction of Cc. The Monte Carlo method based sensitivity analysis results show that out of the 13 input parameters considered for the models study, four parameters namely clay, degree of saturation, specific gravity and depth of sample are the most relevant in the prediction of Cc, and other parameters (bulk density, dry density, void ratio and porosity) are the most insignificant parameters for the prediction of Cc. Removal of these insignificant parameters helped to reduce the dimension of the input space and also model running time, and improved significantly the performance of the AI models. The results of this study might help in selecting the suitable AI models and input parameters for better and quick prediction of the Cc of soil. Unlabelled Image • Different Artificial Intelligence (AI) methods were developed for the prediction of Cc. • Monte Carlo approach is proposed for the sensitivity analysis of the AI models. • RMSE, MAE, and R2 were used for the validation of the models. • SVM is the best method in comparison to ANN and ANFIS methods for the prediction of Cc. [ABSTRACT FROM AUTHOR]

Land subsidence (LS) is among the most critical environmental problems, affecting both agricultural sustainability and urban infrastructure. Existing methods often use either simple regression models or complex hydraulic models to explain and predict LS. There are few studies that identify the risk factors and predict the risk of LS using machine learning models. This study compares four tree-based machine learning models for land subsidence hazard modelling at a study area in Hamadan plain (Iran). The study also analyzes the importance of six risk factors including topography (elevation, slope), geomorphology (distance from stream, drainage density), hydrology (groundwater drawdown) and lithology on LS. Thematic layers of each variable related to the LS phenomenon are prepared and utilized as the inputs to the four tree-based machine learning models, including the Rule-Based Decision Tree (RBDT), Boosted Regression Trees (BRT), Classification And Regression Tree (CART), and the Random Forest (RF) algorithms to produce a consolidated LS hazard map. The accuracy of the generated maps is then evaluated using the area under the receiver operating characteristic curve (AUC) and the True Skill Statistics (TSS). The RF approach had the lowest predictive error for mapping the LS hazard (i.e., AUC 96.7% for training, AUC 93.8% for validation, TSS 0.912 for training, TSS 0.904 for validation) followed by BRT. Groundwater drawdown was seen to be the most influential factor that contributed to land subsidence in the present study area, followed by lithology and distance from the stream network. Unlabelled Image • Four tree-based models were applied for land subsidence modelling. • RF is the best model to spatially predict land subsidence hazard. • Groundwater drawdown was the most influential factor. [ABSTRACT FROM AUTHOR]

Groundwater vulnerability assessment is a measure of potential groundwater contamination for areas of interest. The main objective of this study is to modify original DRASTIC model using four objective methods, Weights-of-Evidence (WOE), Shannon Entropy (SE), Logistic Model Tree (LMT), and Bootstrap Aggregating (BA) to create a map of groundwater vulnerability for the Sari-Behshahr plain, Iran. The study also investigated impact of addition of eight additional factors (distance to fault, fault density, distance to river, river density, land-use, soil order, geological time scale, and altitude) to improve groundwater vulnerability assessment. A total of 109 nitrate concentration data points were used for modeling and validation purposes. The efficacy of the four methods was evaluated quantitatively using the Area Under the Receiver Operating Characteristic (ROC) Curve (AUC). AUC value for original DRASTIC model without any modification of weights and rates was 0.50. Modification of weights and rates resulted in better performance with AUC values of 0.64, 0.65, 0.75, and 0.81 for BA, SE, LMT, and WOE methods, respectively. This indicates that performance of WOE is the best in assessing groundwater vulnerability for DRASTIC model with 7 factors. The results also show more improvement in predictability of the WOE model by introducing 8 additional factors to the DRASTIC as AUC value increased to 0.91. The most effective contributing factor for ground water vulnerability in the study area is the net recharge. The least effective factors are the impact of vadose zone and hydraulic conductivity. [ABSTRACT FROM AUTHOR]

Shear strength of the soil is an important engineering parameter used in the design and audit of geo-technical structures. In this research, we aim to investigate and compare the performance of four machine learning methods, Particle Swarm Optimization - Adaptive Network based Fuzzy Inference System (PANFIS), Genetic Algorithm - Adaptive Network based Fuzzy Inference System (GANFIS), Support Vector Regression (SVR), and Artificial Neural Networks (ANN), for predicting the strength of soft soils. For this purpose, case studies of 188 plastic clay soil samples collected from two major projects, Nhat Tan and Cua Dai bridges in Viet Nam have been used for generating training and testing datasets for constructing and validating the models. Validation and comparison of the models have been carried out using RMSE, and R. The results show that the PANFIS has the highest prediction capability (RMSE = 0.038 and R = 0.601), followed by the GANFIS (RMSE = 0.04 and R = 0.569), SVR (RMSE = 0.044 and R = 0.549), and ANN (RMSE = 0.059 and R = 0.49). It can be concluded that out of four models the PANFIS indicates as a promising technique for prediction of the strength of soft soils. [ABSTRACT FROM AUTHOR]

Floods are one of the most damaging natural hazards causing huge loss of property, infrastructure and lives. Prediction of occurrence of flash flood locations is very difficult due to sudden change in climatic condition and manmade factors. However, prior identification of flood susceptible areas can be done with the help of machine learning techniques for proper timely management of flood hazards. In this study, we tested four decision trees based machine learning models namely Logistic Model Trees (LMT), Reduced Error Pruning Trees (REPT), Naïve Bayes Trees (NBT), and Alternating Decision Trees (ADT) for flash flood susceptibility mapping at the Haraz Watershed in the northern part of Iran. For this, a spatial database was constructed with 201 present and past flood locations and eleven flood-influencing factors namely ground slope, altitude, curvature, Stream Power Index (SPI), Topographic Wetness Index (TWI), land use, rainfall, river density, distance from river, lithology, and Normalized Difference Vegetation Index (NDVI) . Statistical evaluation measures, the Receiver Operating Characteristic (ROC) curve, and Freidman and Wilcoxon signed-rank tests were used to validate and compare the prediction capability of the models. Results show that the ADT model has the highest prediction capability for flash flood susceptibility assessment, followed by the NBT, the LMT, and the REPT, respectively. These techniques have proven successful in quickly determining flood susceptible areas. [ABSTRACT FROM AUTHOR]

This study relates changes in social vulnerability of 20 counties on the northern coasts of the Persian Gulf (PG) and the Gulf of Oman (GO) over a 30-year period (1988–2017) to changing socio-economic conditions and environmental (climate) hazard. Social vulnerability in 2030, 2040 and 2050 is predicted based on the RCP8.5 climate change scenario that projects drought intensities and rising sea levels. Social vulnerability was based on the three dimensions of sensitivity, exposure, and adaptive capacity using 18 socio-economic and five climate indicators identified by experts. All but one indicator related very strongly to the dimension it sought to represent. Despite improvements in adaptive capacity over time, social vulnerability increased between 1988 and 2017 and rates of change accelerated after change point years that occurred between 1998 and 2002 in most counties. Extrapolating past changes of each indicator over time enabled forecasts of social vulnerability in the future. While social variability decreased between 2017 and 2030, it increased again between 2030 and 2050. The lowest future social vulnerability is expected along the eastern PG coast, the greatest along the western PG and the GO. The worsening of socio-economic indicators contributed to increased sensitivity, and increased drought intensities plus the expected rise in sea levels will lead to social vulnerabilities in 2050 comparable to present levels. Between 1.4 and 1.7 M people will live in areas that are likely submerged by water in the future. About 80% of these people live in six counties with variable social vulnerabilities. While counties with lower social variabilities might be better able to cope with the challenges posed by climate change, adaptation programs to enhance the resilience of the residents in these and the remaining counties along the PG and the GO need to be implemented soon to avoid uncontrolled mass migration of millions of people from the region. Unlabelled Image • Social vulnerability was assessed in 20 coastal Iranian counties between 1998 and 2017. • Demographic, socioeconomic, and climatic indicators indicate increased vulnerability. • Social vulnerability between 2030 and 2050 is enhanced by drought and sea level rise. • About 1.3–1.7 M people will be displaced from submerged areas between 2030 and 2050. • The six counties most affected by sea level rise differ in social vulnerability. [ABSTRACT FROM AUTHOR]

Determining the level of ecosystems exposure to multiple environmental hazards or risk factors is of paramount importance for developing, adopting, and planning management strategies to minimize the harmful effects of these hazards. We quantified the level of exposure of mangroves on the northern coasts of the Persian Gulf (PG) and the Gulf of Oman (GO) between 1986 and 2019 to eight environmental hazards, i.e., drought, maximum temperatures, rising sea levels, change of freshwater inflows to coasts, extreme storm surges, significant wave height (SWH), seaward edge retreat in the mangroves, and fishery intensity. Based on expert opinion, fuzzy weights were used to integrate these exposures into a single index (EI) for the region. Experts gave the greatest weight/importance to the risks posed by sea-level rise and seaward retreat of mangroves and the lowest risk to significant wave height and fishery intensity in coastal waters. The overall EI and six of eight individual variables (except fishery intensity and maximum temperatures) pointed to exposure levels of mangroves that increased from the coasts of the PG (EI 0.69) to the GO (EI 6.69). Since these hazards are expected to continue in the future, local/regional management responses should focus on minimizing regional anthropogenic threats and halt conversion of natural areas to agricultural and open areas to maintain freshwater inputs to coastal areas, particularly on the GO. Further, uplands that may serve as future refugia into which mangroves may expand over time as sea levels continue to rise should be protected from development. This was the first study that used an analytic framework to compute a mangrove exposure index to a suite of physical and socio-economic hazards across a region. This framework may provide insights into cost-effective resilience-based design and management of socio-ecologically coupled ecosystems in an era of increasing types and intensities of environmental hazards. Unlabelled Image • Computing mangrove exposure index to climatic, marine and socioeconomic hazards • Experts deem sea level rise and seaward retreat as greatest threats to mangroves. • Significant wave height and fishery intensity deemed least important threats. • Exposure levels increases from the coasts of the Persian Gulf to the Gulf of Oman. • Adaptation plans should be directed to minimize adverse anthropogenic influences. [ABSTRACT FROM AUTHOR]

Dust particles are transported globally. Dust storms can adversely impact both human health and the environment, but they also impact transportation infrastructure, agriculture, and industry, occasionally severely. The identification of the locations that are the primary sources of dust, especially in arid and semi-arid environments, remains a challenge as these sites are often in remote or data-scarce regions. In this study, a new method using state-of-the-art machine-learning algorithms – random forest (RF), support vector machines (SVM), and multivariate adaptive regression splines (MARS) – was evaluated for its ability to spatially model the distribution of dust-source potential in eastern Iran. To accomplish this, empirically identified dust-source locations were determined with the ozone monitoring instrument aerosol index and the Moderate-Resolution Imaging Spectroradiometer (MODIS) Deep Blue aerosol optical thickness methods. The identified areas were divided into training (70%) and validation (30%) sets. Measurements of the conditioning factors (lithology, wind speed, maximum air temperature, land use, slope angle, soil, rainfall, and land cover) were compiled for the study area and predictive models were developed. The area-under-the-receiver operating characteristics curve (AUC) and true-skill statistics (TSS) were used to validate the maps of the models' predictions. The results show that the RF algorithm performed best (AUC = 89.4% and TSS = 0.751), followed by the SVM (AUC = 87.5%, TSS = 0.73) and the MARS algorithm (AUC = 81%, TSS = 0.69). The results of the RF indicated that wind speed and land cover are the most important factors affecting dust generation. The region of highest dust-source potential that was identified by the RF is in the eastern parts of the study region. This model can be applied to other arid and semi-arid environments that experience dust storms to promote management that prevents desertification and reduces dust production. Unlabelled Image • Three machine learning models were used to identify dust-source potential areas. • Random Forest is the best model to spatially predict dust-source potential areas. • Wind speed and land cover are the most important determinants of dust storm occurrence. • Population decline is statistically-significantly spatially correlated to dust-source areas. [ABSTRACT FROM AUTHOR]

Sediment control is important for supplying clean water. Although check dams control sediment yield, site selection for check dams based on the sediment trapping efficiency (TE) is often complex and time-consuming. Currently, a multi-step trial-and-error process is used to find the optimal sediment TE for check dam construction, which limits this approach in practice. To cope with this challenge, we developed a user-friendly, cost- and time-efficient geographic information system (GIS)-based tool, the trap efficiency tool (TET), in the Python programming language. We applied the tool to two watersheds, the Hableh-Rud and the Poldokhtar, in Iran. To identify suitable sites for check dams, four scenarios (S1: TE ≥ 60%, S2: TE ≥ 70%, S3: TE ≥ 80%, S4: TE ≥ 90%) were tested. TET identified 189, 117, 96, and 77 suitable sites for building check dams in S1, S2, S3, and S4, respectively, in the Hableh-Rud watershed, and 346, 204, 156, and 60 sites in S1, S2, S3, and S4, respectively, in the Poldokhtar watershed. Evaluation of 136 existing check dams in the Hableh-Rud watershed indicated that only 10% and 5% were well-located and these were in the TE classes of 80–90% and ≥90%, respectively. In the Poldokhtar watershed, only 11% and 8% of the 207 existing check dams fell into TE classes 80–90% and ≥90%, respectively. Thus, the conventional approach for locating suitable sites at which check dams should be constructed is not effective at reaching suitable sediment control efficiency. Importantly, TET provides valuable insights for site selection of check dams and can help decision makers avoid monetary losses incurred by inefficient check-dam performance. • A trap efficiency tool (TET) was designed using the Python programming language. • TET identified suitable check dam sites in the Hableh-Rud and Poldokhtar watersheds. • Suitable sites for constructing check dams at four TE levels were identified. • More than 71% and 55% of existing check dams in the 2 watersheds were improperly situated. • TET avoids the financial waste caused by inefficient performance of constructed check dams. [ABSTRACT FROM AUTHOR]

• An integrated agricultural-meteorological drought was spatially modeled. • The hybridized drought hazard model ANFIS-BA performed best. • Hybrid models prevent overfitting and can improve ANFIS performance. • The most important factors for spatial drought hazard were determined. Widespread detrimental and long-lasting droughts are having catastrophic impacts around the globe. Researchers, organizations, and policy makers need to work together to obtain precise information, enabling timely and accurate decision making to mitigate drought impacts. In this study, a spatial modeling approach based on an adaptive neuro-fuzzy inference system (ANFIS) and several metaheuristic optimizations (ANFIS-BA, ANFIS-GA, ANFIS-ICA, ANFIS-PSO) was developed to predict the spatial occurrence of drought in a region in southeastern Queensland, Australia. In this approach, data describing the distribution of eight drought-contributing factors were prepared for input into the models to serve as independent variables. Relative departures of rainfall (RDR) and relative departures of soil moisture (RDSM) were analyzed to identify locations where drought conditions have occurred. The set of locations in the study area identified as having experienced drought conditions was randomly divided into two groups, 70% were used for training and 30% for validation. The models employed these data to generate maps that predict the locations that would be expected to experience drought. The prediction accuracy of the model-produced drought maps was scrutinized with two evaluation metrics: area under the receiver operating characteristic curve (AUC) and root mean square error (RMSE). The results demonstrate that the hybridized models (ANFIS-BA (AUC mean = 83.7%, RMSE mean = 0.236), ANFIS-GA (AUC mean = 81.62%, RMSE mean = 0.247), ANFIS-ICA (AUC mean = 82.12%, RMSE mean = 0.247), and ANFIS-PSO (AUC mean = 81.42%, RMSE mean = 0.255)) yield better predictive performance than the standalone ANFIS model (AUC mean = 71.8%, RMSE mean = 0.344). Furthermore, sensitivity analyses indicated that plant-available water capacity, the percentage of soil comprised of sand, and mean annual precipitation were the most important predictors of drought hazard. The versatility of the new approach for spatial drought modeling and the capacity of ANFIS model hybridization to improve model performance suggests great potential to assist decision makers in their formulations of drought risk, recovery, and response management, and in the development of contingency plans. [ABSTRACT FROM AUTHOR]

• DNN was verified and assessed for shallow landslide susceptibility mapping. • Three optimization algorithms, Adam, SGD, and RMSProp, were used. • Two loss functions, MSE and CE, were investigated. • DNN with Adam and MSE is the best and outperformed other models. • DNN-Adam-MSE is a new tool for shallow landslide susceptibility modeling. This research aims at investigating the capability of Keras's deep learning models with three robust optimization algorithms (stochastic gradient descent, root mean square propagation, and adaptive moment optimization) and two-loss functions for spatial modeling of landslide hazard at a regional scale. Shallow landslides at the Ha Long area (Vietnam) were selected as a case study. For this regard, set of ten influencing factors (slope, aspect, curvature, topographic wetness index, landuse, distance to road, distance to river, soil type, distance to fault, and lithology) and 193 landslide polygons were prepared to construct a Geographic Information System (GIS) database for the study area. Using the collected database, the DNN with its potential of realizing complex functional mapping hidden in the data is used to generalize a decision boundary that separates the learning space into two distinct categories: landslide (a positive class) and non-landslide (a negative class). Experimental results point out that the utilized the Keras's deep learning model with the Adam optimization and the mean squared error lost function is the best with the prediction performance of 84.0%. The performance is better than those of the employed benchmark approaches of random forest, J48 decision tree, classification tree, and logistic model tree. We conclude that the Keras's deep learning model is a new tool for shallow susceptibility mapping at landslide-prone areas. [ABSTRACT FROM AUTHOR]

• RF and ANFIS-ICA models were less sensitive to reductions of sample size. • The RF outperformed other models in terms of goodness-of-fit and predictive performance. • The RSP and lithology were the main geo-environmental spring-affecting factors. • Approximately 18.68% of the study area has high or very high groundwater potential. Machine learning models have attracted much research attention for groundwater potential mapping. However, the accuracy of models for groundwater potential mapping is significantly influenced by sample size and this is still a challenge. This study evaluates the influence of sample size on the accuracy of different individual and hybrid models, adaptive neuro-fuzzy inference system (ANFIS), ANFIS-imperial competitive algorithm (ANFIS-ICA), alternating decision tree (ADT), and random forest (RF) to model groundwater potential, considering the number of springs from 177 to 714. A well-documented inventory of springs, as a natural representative of groundwater potential, was used to designate four sample data sets: 100% (D 1), 75% (D 2), 50% (D 3), and 25% (D 4) of the entire springs inventory. Each data set was randomly split into two groups of 30% (for training) and 70% (for validation). Fifteen diverse geo-environmental factors were employed as independent variables. The area under the operating receiver characteristic curve (AUROC) and the true skill statistic (TSS) as two cutoff-independent and cutoff-dependent performance metrics were used to assess the performance of models. Results showed that the sample size influenced the performance of four machine learning algorithms, but RF had a lower sensitivity to the reduction of sample size. In addition, validation results revealed that RF (AUROC = 90.74–96.32%, TSS = 0.79–0.85) had the best performance based on all four sample data sets, followed by ANFIS-ICA (AUROC = 81.23–91.55%, TSS = 0.74–0.81), ADT (AUROC = 79.29–88.46%, TSS = 0.59–0.74), and ANFIS (AUROC = 73.11–88.43%, TSS = 0.59–0.74). Further, the relative slope position, lithology, and distance from faults were the main spring-affecting factors contributing to groundwater potential modelling. This study can provide useful guidelines and a valuable reference for selecting machine learning models when a complete spring inventory in a watershed is unavailable. [ABSTRACT FROM AUTHOR]

Dust storms are believed to play an essential role in many climatological, geochemical, and environmental processes. This atmospheric phenomenon can have a significant negative impact on public health and significantly disturb natural ecosystems. Identifying dust-source areas is thus a fundamental task to control the effects of this hazard. This study is the first attempt to identify dust source areas using hybridized machine-learning algorithms. Each hybridized model, designed as an intelligent system, consists of an adaptive neuro-fuzzy inference system (ANFIS), integrated with a combination of metaheuristic optimization algorithms: the bat algorithm (BA), cultural algorithm (CA), and differential evolution (DE). The data acquired from two key sources – the Moderate Resolution Imaging Spectroradiometer (MODIS) Deep Blue and the Ozone Monitoring Instrument (OMI) – are incorporated into the hybridized model, along with relevant data from field surveys and dust samples. Goodness-of-fit analyses are performed to evaluate the predictive capability of the hybridized models using different statistical criteria, including the true skill statistic (TSS) and the area under the receiver operating characteristic curve (AUC). The results demonstrate that the hybridized ANFIS-DE model (with AUC = 84.1%, TSS = 0.73) outperforms the other comparative hybridized models tailored for dust-storm prediction. The results provide evidence that the hybridized ANFIS-DE model should be explored as a promising, cost-effective method for efficiently identifying the dust-source areas, with benefits for both public health and natural environments where excessive dust presents significant challenges. • A new framework was developed for identification of dust-sources. • Three novel hybridized ANFIS models were developed: ANFIS-BA, ANFIS-CA, ANFIS-DE. • The hybridized ANFIS-DE model had the highest accuracy (AUC = 84.1%, TSS = 0.73). • All hybridized models outperformed the standalone ANFIS model. [ABSTRACT FROM AUTHOR]