Evidential Belief Function Research Articles

Landslide susceptibility assessment along highways (SH-12 and NH-717A) in Darjeeling Himalayas

Landslides are the utmost damaging natural along with anthropogenic hazards which cause huge damage to assets and human lives in hilly environments all over the world. The key aim of this study is to explore landslide susceptibility (LS) maps along highways (SH-12 and NH-717A) in the Darjeeling Himalayas, India. Four data-driven bivariate statistical models (BSMs) were used to develop LS maps based on geospatial techniques. A total of 90 (100%) polygons with different size of landslide locations were identified to develop landslide inventory/distribution (LI/LD) map by incorporating Google Earth and satellite imageries, and filed investigation through the global positioning system (GPS). The LI/LD map is distributed into testing 30% (27 landslide locations), and training 70% (63 landslide locations) of the models. Eighteen landslide causative factors (LCFs) were selected to develop LS maps. Multicollinearity analysis showed that there is no collinearity problem in the LCFs. The receiver operating characteristics (ROC) curve and frequency ratio (FR) value techniques were applied for validation. The relative importance of each LCF for each model has been measured using Jack-knife test. The LS maps were divided into six landslide susceptibility zones (LSZs). The ROC curve showed the accuracy of the LSZ maps is 93.50% for FR, 85.90% for evidential belief functions (EBFs), 81.80% for fuzzy membership (FM) using FR, and 77.20% for FM using cosine amplitude (CA) approaches respectively. Among all models, the FR model showed the highest percentage of accuracy for LS assessment and prediction. The FR values of six LSZs rise from very low to very high LSZs which indicate a positive correlation between LSZs and FR values. The results of Jack-knife test showed that terrain ruggedness index (TRI) factor has the highest relative importance for all models in LSZ mappings. The study findings will help to the planner, and policymakers for spatial planning, implement slope management strategies, land use management in mountainous environments.

Integration of Logistic Regression and Evidential Belief Function for Flood Risk Assessment in the West Bengal Plain, India

Integration of Logistic Regression and Evidential Belief Function for Flood Risk Assessment in the West Bengal Plain, India

Improving the performance of evidential belief function model using five machine learning models in earth fissure susceptibility mapping

The frequency and occurrence of earth fissures threaten human settlements, infrastructure, and agricultural lands in many parts of the globe. This natural hazard has particularly posed great challenge in the North, Central, and Eastern parts of the Najaf Abad Basin of Iran. This paper presents a comparative analysis of machine learning algorithms for earth fissure susceptibility assessment, and involves the creation of an earth fissure database by selecting and validating 200 fissure locations. To improve the accuracy of the fissure locations, various geographic data sources, such as Google Earth images, field surveys, and GPS measurements, were employed. The locations were divided into training and validation sets based on sixteen conditioning factors for finding the earth fissure susceptible zones using EBF (Evidential belief function), EBF – Multilayer Perceptron (MLP), EBF – Function Tree (FT), EBF – Alternating decision tree (ADtree), EBF – Adaptive Boosting (AdaBoost), EBF – Random SubSpace (RSS), EBF – Naive Bayes Tree (NBTree) models. The results show that earth fissures occur in greater concentrations where groundwater levels are lowest, and their prevalence is closely associated to both agricultural activities and the presence of clay, clayey sand, and silty clay soils. Susceptibility analysis results revealed that 15–21 % agricultural dependent population falls under the very high susceptibility risk zone in and around the major cities of the Najaf Abad region. The receiver operating characteristic (ROC) curve analysis showed that the EBF-RSS algorithm has higher accuracy (>99 %) compared to other algorithms (>85 %) in training and validation phases for earth fissure susceptibility prediction. Our findings emphasise the need to prioritise these susceptible areas for sustainable land and water management strategies and implement effective measures to mitigate the impact of earth fissures. Our study also advances our knowledge on geospatial analysis and earth fissures-related hazard assessment by applying five machine learning algorithms that shows spatial similarity in susceptibility categories with splendid accuracy performance.

A robust ensemble of hybrid and bivariate statistical models for flood prediction mapping in Lower Damodar River Basin of India

This research explores flood prediction in the Lower Damodar River Basin (LDRB) using a hybrid ensemble of a Naïve Bayes Tree (NBT) and five bivariate statistical models such as Evidential Belief Function (EBF), Index of Entropy (IOE), Frequency Ratio (FR), Statistical Index (SI), and Modified Information Value (MIV). A total of 348 flood locations and 15 conditioning factors including hydrological, topographical and land cover were considered for this analysis. To ensure the precision of model predictions, a multicollinearity assessment was executed. Receiver operating characteristic (ROC) curve, area under curve (AUC), mean absolute error (MAE), mean squared error (MSE), and root mean squared error (RMSE) were performed to compare and asses each of the models all. The results reveal that all models performed well in creating flood hazard maps with AUC >0.8 and RMSE <0.4. FR and EBF models demonstrated the highest predictive accuracy (AUC=0.85), followed by the IOE, SI, and MIV models. The ensemble of NBT with bivariate models shows promising results, showcasing reduced error metrics and improved accuracy for the IOE, SI, and MIV models. This study highlights the potential of ensemble models in flood hazard prediction, offering valuable insights for global flood risk management. The successful application of these data-driven models showcases their importance in forecasting flood risks, aiding decision-makers and planners in developing more effective flood mitigation strategies.

Comprehensive landslide prediction mapping using bivariate statistical models of Mizoram state of Northeast India

ABSTRACT Landslides in the state of Mizoram result in damage to life and properties annually. The study focuses on landslide susceptibility zones by frequency ratio (FR), evidential belief function (EBF) and index of entropy (IOE) models. A total of 1,486 landslide points were used to build a relationship between 16 factors and landslide occurrences. The results reveal 14.44%, 19.64% and 3.55% of the area as very high susceptible zones in FR, EBF and IOE models, respectively. The AUC results support the adoption of the EBF model in land use planning and decision-making processes to enhance natural resource management and mitigate landslide risks in Mizoram.

Determining land subsidence potential using the evidential belief function model: A case study for the Bardaskan Aquifer, Iran

AbstractIn this study, we employed the evidential belief function model (EBF) to evaluate the potential for land subsidence in the primary aquifer of Bardaskan. Through field visits, we recorded GPS coordinates for 174 land subsidence points. Factors considered in assessing land subsidence potential included well density, groundwater extraction rate, geological characteristics, proximity to faults, vegetation cover, distance from the river, slope, and land use. To develop and validate the model, 70% of the recorded points were randomly selected for training and implementation, while the remaining 30% were reserved for model validation. The number and percentage of land subsidence points in the different classes of the corresponding layers were determined by integrating the training points with influential variables maps such as distance from the river, distance from the fault, land use, and extraction volume. The EBF model rate was calculated for different layer classes. For modeling, all rates of the EBF model in each cell were summated, and the ‎potential of land subsidence was calculated.‎ Finally, the map of land subsidence potential based on the EBF model was determined with GIS. The results showed that most of the subsidence points were located in alluvial sediment of the Holocene period, in areas with high groundwater harvesting, a distance of at least 3000 m from a river, a distance of at least 6000 m from a fault, low‐density rangelands, slopes of at least 0%–2%, and farmlands and gardens. A receiver operating characteristic curve analysis of the EBF model showed that it could accurately predict land subsidence in 87.5% of cases using 30% of the validation data. This suggests that the model can be used for practical applications.

Application of GIS-based data-driven bivariate statistical models for landslide prediction: a case study of highly affected landslide prone areas of Teesta River basin

Predicting landslides has become a critical global challenge for promoting sustainable development in mountainous regions. This study conducts a comparative analysis of landslide susceptibility maps (L.S.M.s) generated using two GIS-based data-driven bivariate statistical models: (a) Frequency Ratio (F.R.) and (b) Evidential Belief Function (E.B.F). These models are applied and evaluated in the high landslide-prone upper and middle Teesta basin of the Darjeeling-Sikkim Himalaya, leveraging geographic information system (GIS) and remote sensing techniques. We compile a comprehensive landslide inventory map containing 2387 regional landslide points. We use approximately 70% of this dataset for model training and reserve the remaining 30% for validation. In the construction of the Landslide Susceptibility maps (LSMs), a comprehensive set of twenty-one landslide-triggering parameters has been considered. These parameters encompass factors such as elevation, distance from drainage, distance from lineament, distance from roads, geology, geomorphology, lithology, land use, and land cover, normalized difference vegetation index, profile curvature, rainfall, relief amplitude, roughness, slope, slope aspect, slope classes, stream power index, sediment transport index, topographic position index, topographic ruggedness index, and topographic wetness index. An examination of multicollinearity statistics reveals no collinearity issues among the twenty-one causative factors utilized in this research. The final L.S.M.s demonstrate that the combined application of the F.R. and E.B.F. models yields the highest training accuracy at 98.10%. The insights derived from this study hold significant promise as valuable tools for assessing environmental hazards and land use planning.

An Ensemble Modeling of Frequency Ratio (FR) with Evidence Belief Function (EBF) for GIS-Based Landslide Susceptibility Mapping: A Case Study of the Coastal Cliff of Safi, Morocco

An Ensemble Modeling of Frequency Ratio (FR) with Evidence Belief Function (EBF) for GIS-Based Landslide Susceptibility Mapping: A Case Study of the Coastal Cliff of Safi, Morocco

Investigation of Landslide Susceptibility Decision Mechanisms in Different Ensemble-Based Machine Learning Models with Various Types of Factor Data

Machine learning (ML)-based methods of landslide susceptibility assessment primarily focus on two dimensions: accuracy and complexity. The complexity is not only influenced by specific model frameworks but also by the type and complexity of the modeling data. Therefore, considering the impact of factor data types on the model’s decision-making mechanism holds significant importance in assessing regional landslide characteristics and conducting landslide risk warnings given the achievement of good predictive performance for landslide susceptibility using excellent ML methods. The decision-making mechanism of landslide susceptibility models coupled with different types of factor data in machine learning methods was explained in this study by utilizing the Shapley Additive exPlanations (SHAP) method. Furthermore, a comparative analysis was carried out to examine the differential effects of diverse data types for identical factors on model predictions. The study area selected was Cenxi, Guangxi, where a geographic spatial database was constructed by combining 23 landslide conditioning factors with 214 landslide samples from the region. Initially, the factors were standardized using five conditional probability models, frequency ratio (FR), information value (IV), certainty factor (CF), evidential belief function (EBF), and weights of evidence (WOE), based on the spatial arrangement of landslides. This led to the formation of six types of factor databases using the initial data. Subsequently, two ensemble-based ML methods, random forest (RF) and XGBoost, were utilized to build models for predicting landslide susceptibility. Various evaluation metrics were employed to compare the predictive capabilities of different models and determined the optimal model. Simultaneously, the analysis was conducted using the interpretable SHAP method for intrinsic decision-making mechanisms of different ensemble-based ML models, with a specific focus on explaining and comparing the differential impacts of different types of factor data on prediction results. The results of the study illustrated that the XGBoost-CF model constructed with CF values of factors not only exhibited the best predictive accuracy and stability but also yielded more reasonable results for landslide susceptibility zoning, and was thus identified as the optimal model. The global interpretation results revealed that slope was the most crucial factor influencing landslides, and its interaction with other factors in the study area collectively contributed to landslide occurrences. The differences in the internal decision-making mechanisms of models based on different data types for the same factors primarily manifested in the extent of influence on prediction results and the dependency of factors, providing an explanation for the performance of standardized data in ML models and the reasons behind the higher predictive performance of coupled models based on conditional probability models and ML methods. Through comprehensive analysis of the local interpretation results from different models analyzing the same sample with different sample characteristics, the reasons for model prediction errors can be summarized, thereby providing a reference framework for constructing more accurate and rational landslide susceptibility models and facilitating landslide warning and management.

Optimizing Rotation Forest-Based Decision Tree Algorithms for Groundwater Potential Mapping

Groundwater potential mapping is an important prerequisite for evaluating the exploitation, utilization, and recharge of groundwater. The study uses BFT (best-first decision tree classifier), CART (classification and regression tree), FT (functional trees), EBF (evidential belief function) benchmark models, and RF-BFTree, RF-CART, and RF-FT ensemble models to map the groundwater potential of Wuqi County, China. Firstly, select sixteen groundwater spring-related variables, such as altitude, plan curvature, profile curvature, curvature, slope angle, slope aspect, stream power index, topographic wetness index, stream sediment transport index, normalized difference vegetation index, land use, soil, lithology, distance to roads, distance to rivers, and rainfall, and make a correlation analysis of these sixteen groundwater spring-related variables. Secondly, optimize the parameters of the seven models and select the optimal parameters for groundwater modeling in Wuqi County. The predictive performance of each model was evaluated by estimating the area under the receiver operating characteristic (ROC) curve (AUC) and statistical index (accuracy, sensitivity, and specificity). The results show that the seven models have good predictive capabilities, and the ensemble model has a larger AUC value. Among them, the RF-BFT model has the highest success rate (AUC = 0.911), followed by RF-FT (0.898), RF-CART (0.894), FT (0.852), EBF (0.824), CART (0.801), and BFtree (0.784), respectively. Groundwater potential maps of these 7 models were obtained, and four different classification methods (geometric interval, natural breaks, quantile, and equal interval) were used to reclassify the obtained GPM into 5 categories: very low (VLC), low (LC), moderate (MC), high (HC), and very high (VHC). The results show that the natural breaks method has the best classification performance, and the RF-BFT model is the most reliable. The study highlights that the proposed ensemble model has more efficient and accurate performance for groundwater potential mapping.

An ensemble of evidence belief function (EBF) with frequency ratio (FR) using geospatial data for landslide prediction in Darjeeling Himalayan region of India

Landslide is a natural hazards that can be extremely difficult to control without continuous monitoring. The research area is in the Himalayan Mountains region, characterized by complicated geological structures. Landslide occurs in this area and damages property, life, infrastructure, and others. So it is essential to study the susceptible areas for planning and decision-makers. Primarily, 1582 landslides were identified and divided into two distinct groups: the training dataset consisting of 1092 (70%) landslides used to develop the model and the validation dataset consisting of 490 (30%) landslides used to validate the utilized model. In this research, we have used sixteen spatial datasets, viz, topographic datasets are elevation, slope, profile curvature, topographic ruggedness index (TRI), topographic position index (TPI), topographic wetness index (TWI), stream power index (SPI), sediment transport index (STI); hydrological datasets are rainfall and distance to drainage; geological datasets are lithology, geomorphology, distance to fault; landcover datasets are NDVI, distance to road, LULC. This study aims to perform the landslide susceptibility zonation using the performance of evidence belief function (EBF) and frequency ratio (FR) model in the Darjeeling Himalayan region. The susceptibility models have been categorized into five classes, i.e., very low, low, moderate, high, and very high, where EBF model accounting 4.51, 1.22, 25.60, 38.27, and 40.41 percent, respectively, and FR model accounting 5.02, 22.14, 31.61, 33.51, and 7.71 percent, respectively. The receiver operating characteristic and area under the curve (ROC-AUC) was used to measure the success and prediction rates for validation. The result revealed that the success rate of the EBF and FR models is 0.937 and 0.936, and the prediction rate is 0.949 and 0.953, respectively.

GIS-based data-driven bivariate statistical models for landslide susceptibility prediction in Upper Tista Basin, India

Predicting landslides is becoming a crucial global challenge for sustainable development in mountainous areas. This research compares the landslide susceptibility maps (LSMs) prepared from five GIS-based data-driven bivariate statistical models, namely, (a) Frequency Ratio (FR), (b) Index of Entropy (IOE), (c) Statistical Index (SI), (d) Modified Information Value Model (MIV) and (e) Evidential Belief Function (EBF). These five models were tested in the high landslides-prone humid sub-tropical type Upper Tista basin of the Darjeeling-Sikkim Himalaya by integrating the GIS and remote sensing. The landslide inventory map consisting of 477 landslide locations was prepared, and about 70% of all landslide data was utilized for training the model, and 30% was used to validate it after training. A total of fourteen landslide triggering parameters (elevation, slope, aspect, curvature, roughness, stream power index, TWI, distance to stream, distance to road, NDVI, LULC, rainfall, modified fournier index, and lithology) were taken into consideration for preparing the LSMs. The multicollinearity statistics revealed no collinearity problem among the fourteen causative factors used in this study. Based on the FR, MIV, IOE, SI, and EBF approaches, 12.00%, 21.46%, 28.53%, 31.42%, and 14.17% areas, respectively, identified in the high and very high landslide-prone zones. The research also revealed that the IOE model has the highest training accuracy of 95.80%, followed by SI (92.60%), MIV (92.20%), FR (91.50%), and EBF (89.90%) models. Consistent with the actual distribution of landslides, the very high, high, and medium hazardous zones stretch along the Tista River and major roads. The suggested landslide susceptibility models have enough accuracy for usage in landslide mitigation and long-term land use planning in the study area. Decision-makers and local planners may utilise the study's findings. The techniques for determining landslide susceptibility can also be employed in other Himalayan regions to manage and evaluate landslide hazards.

Mapping Potential Water Resource Areas Using GIS-Based Frequency Ratio and Evidential Belief Function

Groundwater is a critical freshwater resource that is necessary for sustaining life. Thus, targeting prospective groundwater zones is crucial for the extraction, use, and management of water resources. In this study, we combined the remote sensing, GIS-based frequency ratio (FR), and evidential belief function (EBF) techniques into a model to delineate and quantify prospective groundwater zones. To accomplish this, we processed Shuttle Radar Topography Mission (SRTM), Landsat-8 Operational Land Imager (OLI), Sentinel-2, and rainfall data to reveal the geomorphic, hydrologic, and structural elements and climatic conditions of the study area, which is downstream of the Yellow River basin, China. We processed, quantified, and combined twelve factors (the elevation, slope, aspect, drainage density, lineament density, distance to rivers, NDVI, TWI, SPI, TRI, land use/cover, and rainfall intensity) that control the groundwater infiltration and occurrence using the GIS-based FR and EBF models to produce groundwater potential zones (GWPZs). We used the natural breaks classifier to categorize the groundwater likelihood at each location as very low, low, moderate, high, or very high. The FR model exhibited a better performance than the EBF model, as evidenced by the area under the curve (AUC) assessment of the groundwater potential predictions (FR AUCs of 0.707 and 0.734, and EBF AUCs of 0.665 and 0.690). Combining the FR and EBF models into the FR–EBF model increased the accuracy (AUC = 0.716 and 0.747), and it increased the areas of very high and moderate potentiality to 1.97% of the entire area, instead of the 0.39 and 0.78% of the FR and EBF models, respectively. The integration of remote sensing and GIS-data-driven techniques is crucial for the mapping of groundwater prospective zones.

A Review on Landslide Susceptibility Mapping in Malaysia: Recent Trend and Approaches

The accelerating economic growth has assisted rapid urban development and expansion of construction sites into the landslide-vulnerable zones in Malaysia. Thus landslide susceptibility mapping has now become an important part of project designing work for landslide zone areas. There are several models that are used for susceptibility mapping, especially in the peninsular region. Every model has its own set of selected computing variables and characteristics to generate a map. To date, there is no single method applicable to assess and predict all landslides, as there are variations of geomorphological conditions set by the nature. This paper has reviewed recent research publications on landslide susceptibility mapping in Malaysia. Results show that there are 16 models that are being used to describe landslide risk mapping and among them, the Fuzzy model, Neural Network combined with Fuzzy logic, evidential belief function model, probability analysis (e.g. Weights-of-Evidence, and regression), and Support Vector Machine models are proved to be effective even in the areas with limited information. It is observed that most of the susceptible models use curvature, slope angles, distance from drainage, altitude, slope gradient, road distance, aspects as variable factors, and prolonged rainfall as the prime triggering factors. Furthermore, it is observed that the maximum number of research has been conducted in Cameron Highlands (28%) and Penang (20%), because of their high frequencies of landslide occurring and vulnerabilities. Sabah and Sarawak are covered by a negligible number of susceptibility research. Further, a comparison study between the selected models presents the limitations of each model and their benefits and some suggestions are also made based on the author's recommendations works.

A Comparative Study of Forest Fire Mapping Using GIS-Based Data Mining Approaches in Western Iran

Mapping fire risk accurately is essential for the planning and protection of forests. This study aims to map fire risk (probability of ignition) in Marivan County of Kurdistan province, Iran, using the data mining approaches of the evidential belief function (EBF) and weight of evidence (WOE) models, with an emphasis placed on climatic variables. Firstly, 284 fire incidents in the region were randomly divided into two groups, including the training group (70%, 199 points) and the validation group (30%, 85 points). Given the previous studies and conditions of the region, the variables of slope percentage, slope direction, altitude, distance from rivers, distance from roads, distance from settlements, land use, slope curvature, rainfall, and maximum annual temperature were considered for zoning fire risk. Then, forest fire risk maps were prepared using the EBF and WOE models. The performance of each model was examined using the Relative Operating Characteristic (ROC) curve. The results showed that WOE and EBF are effective tools for mapping forest fire risks in the study area. However, the WOE model shows a slightly higher Area Under the Curve value (0.896) compared to that of the EBF model (0.886), indicating a slightly better performance. The results of this study can provide valuable information for preventing forest fires in the study area.

Comparison of statistical and MCDM approaches for flood susceptibility mapping in northern Iran

• This study is a comparison assessment between statistical and MCDM approach in FHSM in Mohammad Abad catchment, Iran. • The results show that MCDM model have better prediction accuracy than statistical models. • The results show that slope, distance of rivers and geology are the most influential factors in flood formation, respectively. • Evaluations show that increasing Ground Flood Points will have impact on the results of statistical models. Accurate mapping of flood risk areas is the basis for providing basic information on flood hazard reduction strategies and facilitates the relocation process. This study compared statistical approaches and multi-criteria-decision-making (MCDM) in flood hazard susceptibility mapping (FHSM). The performance of two statistical methods, the Evidential Belief Function (EBF) and Weight of Evidence (WOE), was compared with the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) as an MCDM technique. Mohammad-Abad catchment, known as one of the flood susceptible areas in northern Iran, was selected as a case study. A 100-year flood event with a peak flow of 85 m 3 /s, known as the most severe flood in the study catchment during the last 20 years (2000–2020), is considered a basis for selected modelling and evaluation. The accuracy and efficiency of the adopted methods were evaluated using the area under the receiver operating characteristic (AUROC) curve, seed cell surface index, and frequency ratio. Flood inventory maps (including 56 flooded points) and flood-related conditioning factors in the study area were prepared to establish FHSM. Elevation, slope, aspect, plan curvature, topographic wetness index (TWI), stream power index (SPI), distance from the river, drainage density, NDVI, geology, soil type, and land use/land cover were used as flood-related conditional factors. The results of the MCDM method showed that the slope of the catchment is the most important factor in flood formation (with a relative weight of 0.25). By examining the validity of the methods, TOPSIS showed the highest efficiency (AUC = 0.8423), followed by WofE (AUC = 0.7686) and EBF methods (AUC = 0.6251). Based on the frequency ratio and values ​​of seed cell surface index, the MCDM approach shows better performance than statistical methods.

Landslide susceptibility mapping using GIS-based bivariate models in the Rif chain (northernmost Morocco)

The coastline between Tetouan and Bou Ahmed in the northernmost Rif of Morocco and its hinterland has become immensely hazardous due to frequent triggering of diversified landslides from last two decades. This paper describes the potential application of a set of multisource data and the GIS platform for zoning and identifying anomalous areas prone to landsliding and its associated landslide hazards. For this purpose, Information value (IV), Statistical index SI (Wi), Weighting factors (WF) and Evidential belief function (EBF) models have been used in this study. Eleven conditioning factors such as elevation, slope, aspect, curvature, shaded/relief, proximity to streams, proximity to faults, proximity to roads, land use, lithology, annual rainfall and an inventory of 905 unstable spots were used to develop the spatial database for landslides susceptibility mapping (LSM). The factors have been used after a test of multi-collinearity. The Receiver Operating Characteristic (ROC) curve and the Area Under the Curve (AUC) methods were used for validation of the LSM. The AUC results showed good prediction accuracy for all models with a prediction rate of 78% (IV), 77% (SI), 73% (WF) and 70% (EBF) respectively. However, the results indicated that comparatively, the IV model followed by WI model is more precise and accurate for landslides susceptibility mapping than other models. According to the presented models, about 64% of the study area is located in high to very high landslide susceptible zone. The findings presented in this study are imperatively valuable especially wherein large development projects and land use planning activities are going on.

Spatial prediction of groundwater potentiality using machine learning methods with Grey Wolf and Sparrow Search Algorithms

• Four hybrid models were constructed for groundwater potential mapping and different sample datasets (D1–D3) were used to increase the confidence of the result. • The Sparrow Search Algorithm was first used in groundwater modeling. • The correlation between conditioning factors and groundwater occurrence were analyze. • The hybrid models were validated using ROC curve and related statistical indexes. • All hybrid models obtained high accuracy on the training and validation datasets, and the Sparrow Search Algorithm performs better. Groundwater is considered to be one of the most valuable natural resources in the world. However, the availability of groundwater is of concern. Therefore, understanding the potential of groundwater is very important for the utilization of water resources. The main goal of the study was to predict and assess the groundwater using hybrid machine learning and metaheuristic algorithms to automatically tune the parameters, namely the Random Forest (RF), Support Vector Machines (SVM), the Grey Wolf Algorithm (GWO), and the Sparrow Search Algorithm (SSA). A total of 608 groundwater locations were identified by field surveys. Three different sample datasets (D1-D3) were created to increase the confidence of the result, and each dataset was divided randomly into a training set (70%) and a validation set (30%). Fifteen conditioning factors involving geology, human activity, and hydrology were extracted from the available materials. The Evidential Belief Function (EBF) was employed to determine the correlation between groundwater and factors. Then fourteen relevant factors were selected by feature selection. After that, the hybrid models of RF-GWO, RF-SSA, SVM-GWO, and SVM-SSA built using the datasets (D1-D3) were applied to generate groundwater potential maps (GPMs). Results showed that the performances of the hybrid models can be considered to be stable. The global performance of these hybrid models was assessed using the area under the receiver operating characteristic curve (AUC-ROC) and related statistical indexes. According to the D1 dataset validation results, the AUC values for the RF-GWO, RF-SSA, SVM-GWO, and SVM-SSA were 0.832, 0.840, 0.790, and 0.809, respectively. The RF-SSA had the highest accuracy (0.764), with an AUC of 0.840, and the SVM-GWO showed the least accuracy (0.723), with an AUC of 0.790. The outcomes revealed that all hybrid models had a good predictive performance. However, the RF outperformed the SVM model, and the SSA algorithm is superior to the GWO algorithm. Overall, the results of this study show that the construction of the groundwater potential model using a metaheuristic optimization algorithm is a feasible approach.

Evidential belief functions modeling of geophysical and multi-element geochemical data for Pb-Zn mineral potential targeting

The Evidential Belief Functions model was used for investigating the geochemical and geophysical anomalies and delineating the Pb–Zn mineralization zones in the Chichakloo area, east of Takab, Iran. In addition, we evaluated the uncertainty of geochemical and geophysical results in the detection of mineralization zones. For this purpose, the individual belief, disbelief, plausibility, and uncertainty functions of geophysical and geochemical predictor maps were assigned and then integrated using belief function equations. The obtained belief map reveals a remarkable potential for Pb–Zn mineralization in the east part of the study area with a conservative estimation of more than 50% and uncertainty of less than 10%. An increase in conservative estimate and a decrease in the uncertainty rate in this zone, compared to any of the other individual exploration predictor maps, indicate remarkable overlapping anomalies. The results of drilled boreholes proved the existence of the Pb–Zn mineralization and confirmed the obtained mineral potential map.

Assessment of groundwater vulnerability in an urban area: a comparative study based on DRASTIC, EBF, and LR models.

The groundwater vulnerability assessment is known as a useful tool for predicting and prevention of groundwater pollution. This study targets the DRASTIC, evidential belief function (EBF), and logistic regression (LR) models to assess vulnerability in Kabul aquifers, Afghanistan Country. The growth of urban sprawl, groundwater overexploitation, and lack of suitable municipal sewage systems as anthropogenic sources have been the main potential to increase groundwater contaminants such as nitrate in the study area. The vulnerability map has been developed based on various effective factors including altitude, slope (percentage rise), aspect, curvature, land-use type, drainage density, distance from river, annual mean precipitation, net recharge, geology/lithology units, the impact of the vadose zone, aquifer media, depth to water (unsaturated zone), saturated zone, drawdown, and hydraulic conductivity. To identify groundwater pollution, the spatial variation of nitrate concentration data in 2018 was considered indication of groundwater pollution. Based on descriptive statistics, the value of 2.65mg/l (the median of the pixel values of nitrate map) was selected as a threshold to differentiate the occurrence and non-occurrence of pollution. The groundwater quality data were selected and randomly divided into two datasets for training and validation, including 70% and 30%, respectively. The success-rate and prediction-rate curves were computed based on the receiver operating characteristic (ROC) curve and the area under the curve (AUC) to estimate the efficiency of models. The ROC-AUC of success rates for EBF, LR, and DRASTIC models were estimated to be 67%, 66%, and 52%, respectively. Moreover, the ROC-AUC of the prediction rates of the EBF, LR, and DRASTIC models were obtained 61%, 63%, and 55%, respectively. Based on correlation between mean nitrate concentration and the mean vulnerability indexes in each model, the EBF model is the most compatible with the current developed vulnerability zones as the role of mankind in changing the environment in real conditions in comparison to LR and DRASTIC models.