Investigating machine learning and ensemble learning models in groundwater potential mapping in arid region: case study from Tan-Tan water-scarce region, Morocco DOI Creative Commons
Abdessamad Jari, El Mostafa Bachaoui, Soufiane Hajaj

и другие.

Frontiers in Water, Год журнала: 2023, Номер 5

Опубликована: Дек. 13, 2023

Groundwater resource management in arid regions has a critical importance for sustaining human activities and ecological systems. Accurate mapping of groundwater potential plays vital role effective water planning. This study investigates the effectiveness machine learning models, including Random Forest (RF), Adaboost, K-Nearest Neighbors (KNN), Gaussian Process (GWPM) Tan-Tan region, Morocco. Fourteen conditional factors were considered following multicollinearity test, topographical, hydrological, climatic, geological factors. Additionally, point data with 174 sites indicative occurrences incorporated. The inventory underwent random partitioning into training testing datasets at three different ratios: 55/45%, 65/35%, 75/25%. Ultimately, comprehensive ranking 13 encompassing both individual ensemble was determined using prioritization rank technique. results revealed that (EL) particularly RF Adaboost (RF-Adaboost), outperformed models mapping. Based on accuracy assessment validation dataset, RF-Adaboost EL yielded an Area Under Receiver Operating characteristic Curve (AUROC) Overall Accuracy (OA) 94.02 94%, respectively. Ensemble have been effectively applied to integrate 14 factors, capturing their intricate interrelationships, thereby enhancing robustness prediction water-scarce region. Among natural current identified lithology, structural elements (such as faults tectonic lineaments), land use significant contributors potential. However, characteristics area showing coastal position well low background prospectivity (low borehole points) are challenging GWPM. findings highlight assessing managing resources regions. Moreover, this makes contribution by demonstrating algorithms

Язык: Английский

Evaluation of groundwater potential using ANN-based mountain gazelle optimization: A framework to achieve SDGs in East El Oweinat, Egypt DOI Creative Commons
Mahmoud E. Abd-Elmaboud, Ahmed M. Saqr, Mustafa El-Rawy

и другие.

Journal of Hydrology Regional Studies, Год журнала: 2024, Номер 52, С. 101703 - 101703

Опубликована: Фев. 12, 2024

A pilot case study in East El Oweinat (PCSEO), Egypt. An artificial neural network (ANN)-based mountain gazelle optimization (MGO) model was applied to map groundwater potential zones (GWPZs). For this purpose, ten layers affecting occurrence were prepared and normalized against the drawdown (DD) map. All data divided into 70:30 for training testing. After that, sensitivity analysis adopted verify relative importance (RI) of layers. The accuracy GWPZs checked using receiver operating characteristic (ROC) curve other statistical indicators. finally propose a sustainable strategy exploration by implementing integrated MODFLOW-USG MGO framework. Over 40% PCSEO revealed high very degrees situated mostly on southwestern side. Sensitivity that significantly affected table (GWT), well density (WD), land use (LU). results also indicated ANN-based performed with an area under (AUC) ∼ 90% compared conventional models. Additionally, MODFLOW-USG-based gave spatial distribution optimal discharge well-depth zones. This finding could match SDGs relevant ending poverty, affordable groundwater, life land.

Язык: Английский

Процитировано

27
Novel Ensemble Machine Learning Modeling Approach for Groundwater Potential Mapping in Parbhani District of Maharashtra, India DOI Open Access
Md Masroor, Haroon Sajjad, Pankaj Kumar

и другие.

Water, Год журнала: 2023, Номер 15(3), С. 419 - 419

Опубликована: Янв. 19, 2023

Groundwater is an essential source of water especially in arid and semi-arid regions the world. The demand for due to exponential increase population has created stresses on available groundwater resources. Further, climate change affected quantity globally. Many parts Indian cities are experiencing scarcity. Thus, assessment potential necessary sustainable utilization management We utilized a novel ensemble approach using artificial neural network multi-layer perceptron (ANN-MLP), random forest (RF), M5 prime (M5P) support vector machine regression (SMOReg) models assessing Parbhani district Maharashtra India. Ten site-specific influencing factors, elevation, slope, aspect, drainage density, rainfall, table depth, lineament land use cover, geomorphology, soil types, were integrated preparation zones. results revealed that largest area was found under moderate category GWP zone followed by poor, good, very good poor. Spatial distribution zones showed Poor GWPZs spread over north, central southern district. Very poor mostly north-western study calls policy implications conserve manage these parts. ensembled model proved be effective outcome may help stakeholders efficiently utilize devise suitable strategies its management. Other geographical find methodology adopted this assessment.

Язык: Английский

Процитировано

30
Spatial Prediction of Wildfire Susceptibility Using Hybrid Machine Learning Models Based on Support Vector Regression in Sydney, Australia DOI Creative Commons
Arip Syaripudin Nur, Yong Je Kim, Joon Lee

и другие.

Remote Sensing, Год журнала: 2023, Номер 15(3), С. 760 - 760

Опубликована: Янв. 28, 2023

Australia has suffered devastating wildfires recently, and is predisposed to them due several factors, including topography, meteorology, vegetation, ignition sources. This study utilized a geographic information system (GIS) technique analyze understand the factors that regulate spatial distribution of wildfire incidents machine learning predict susceptibility in Sydney. Wildfire inventory data were constructed by combining fire perimeter through field surveys occurrence gathered from visible infrared imaging radiometer suite (VIIRS)-Suomi thermal anomalies product between 2011 2020 for Sydney area. Sixteen wildfire-related acquired assess potential based on support vector regression (SVR) various metaheuristic approaches (GWO PSO) mapping In addition, 2019–2020 “Black Summer” acted as validation dataset predictive capability developed model. Furthermore, gain ratio (IGR) method showed driving such land use, forest type, slope degree have large impact area, frequency (FR) represented how influence occurrence. Model evaluation area under curve (AUC) root average square error (RMSE) used, outputs hybrid-based SVR-PSO (AUC = 0.882, RMSE 0.006) model performed better than standalone SVR 0.837, 0.097) SVR-GWO 0.873, 0.080) models. Thus, optimizing with metaheuristics improved accuracy modeling The proposed framework can be an alternative approach adapted any research related different disturbances.

Язык: Английский

Процитировано

29
Leveraging machine learning in porous media DOI Creative Commons
Mostafa Delpisheh, Benyamin Ebrahimpour,

Abolfazl Fattahi

и другие.

Journal of Materials Chemistry A, Год журнала: 2024, Номер 12(32), С. 20717 - 20782

Опубликована: Янв. 1, 2024

Evaluating the advantages and limitations of applying machine learning for prediction optimization in porous media, with applications energy, environment, subsurface studies.

Язык: Английский

Процитировано

13
Performance evaluation of convolutional neural network and vision transformer models for groundwater potential mapping DOI
Behnam Sadeghi, Ali Asghar Alesheikh,

Ali Jafari

и другие.

Journal of Hydrology, Год журнала: 2025, Номер unknown, С. 132840 - 132840

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

1
Spatial variability of soil water erosion: Comparing empirical and intelligent techniques DOI Creative Commons
Ali Golkarian, Khabat Khosravi, Mahdi Panahi

и другие.

Geoscience Frontiers, Год журнала: 2022, Номер 14(1), С. 101456 - 101456

Опубликована: Авг. 22, 2022

Soil water erosion (SWE) is an important global hazard that affects food availability through soil degradation, a reduction in crop yield, and agricultural land abandonment. A map of susceptibility first vital step management conservation. Several machine learning (ML) algorithms optimized using the Grey Wolf Optimizer (GWO) metaheuristic algorithm can be used to accurately SWE susceptibility. These include Convolutional Neural Networks (CNN CNN-GWO), Support Vector Machine (SVM SVM-GWO), Group Method Data Handling (GMDH GMDH-GWO). Results obtained these compared with well-known Revised Universal Loss Equation (RUSLE) empirical model Extreme Gradient Boosting (XGBoost) ML tree-based models. We apply methods together frequency ratio (FR) Information Gain Ratio (IGR) determine relationship between historical data controlling geo-environmental factors at 116 sites Noor-Rood watershed northern Iran. Fourteen are classified topographical, hydro-climatic, cover, geological groups. next divided into two datasets, one for training (70% samples = 81 locations) other validation (30% 35 locations). Finally model-generated maps were evaluated Area under Receiver Operating Characteristic (AU-ROC) curve. Our results show elevation rainfall erosivity have greatest influence on SWE, while texture hydrology less important. The CNN-GWO (AU-ROC 0.85) outperformed models, specifically, order, SVR-GWO GMDH-GWO (AUC 0.82), CNN GMDH 0.81), SVR XGBoost 0.80), RULSE. Based RUSLE model, loss ranges from 0 2644 t ha–1yr−1.

Язык: Английский

Процитировано

31
Remote sensing and GIS-based machine learning models for spatial gully erosion prediction: A case study of Rdat watershed in Sebou basin, Morocco DOI
My Hachem Aouragh, Safae Ijlil,

Narjisse Essahlaoui

и другие.

Remote Sensing Applications Society and Environment, Год журнала: 2023, Номер 30, С. 100939 - 100939

Опубликована: Фев. 20, 2023

Язык: Английский

Процитировано

22
Deep dive into predictive excellence: Transformer's impact on groundwater level prediction DOI
Wei Sun, Li‐Chiu Chang, Fi‐John Chang

и другие.

Journal of Hydrology, Год журнала: 2024, Номер 636, С. 131250 - 131250

Опубликована: Апрель 28, 2024

Язык: Английский

Процитировано

9
Application of bagging and boosting ensemble machine learning techniques for groundwater potential mapping in a drought-prone agriculture region of eastern India DOI Creative Commons

Krishnagopal Halder,

Amit Kumar Srivastava,

Anitabha Ghosh

и другие.

Environmental Sciences Europe, Год журнала: 2024, Номер 36(1)

Опубликована: Сен. 2, 2024

Groundwater is a primary source of drinking water for billions worldwide. It plays crucial role in irrigation, domestic, and industrial uses, significantly contributes to drought resilience various regions. However, excessive groundwater discharge has left many areas vulnerable potable shortages. Therefore, assessing potential zones (GWPZ) essential implementing sustainable management practices ensure the availability present future generations. This study aims delineate with high Bankura district West Bengal using four machine learning methods: Random Forest (RF), Adaptive Boosting (AdaBoost), Extreme Gradient (XGBoost), Voting Ensemble (VE). The models used 161 data points, comprising 70% training dataset, identify significant correlations between presence absence region. Among methods, (RF) (XGBoost) proved be most effective mapping potential, suggesting their applicability other regions similar hydrogeological conditions. performance metrics RF are very good precision 0.919, recall 0.971, F1-score 0.944, accuracy 0.943. indicates strong capability accurately predict minimal false positives negatives. (AdaBoost) demonstrated comparable across all (precision: recall: F1-score: accuracy: 0.943), highlighting its effectiveness predicting accurately; whereas, outperformed slightly, higher values metrics: (0.944), (0.971), (0.958), (0.957), more refined model performance. (VE) approach also showed enhanced performance, mirroring XGBoost's 0.958, 0.957). that combining strengths individual leads better predictions. potentiality zoning varied significantly, low accounting 41.81% at 24.35%. uncertainty predictions ranged from 0.0 0.75 area, reflecting variability need targeted strategies. In summary, this highlights critical managing resources effectively advanced techniques. findings provide foundation practices, ensuring use conservation beyond.

Язык: Английский

Процитировано

7
Creation of Wildfire Susceptibility Maps in Plumas National Forest Using InSAR Coherence, Deep Learning, and Metaheuristic Optimization Approaches DOI Creative Commons
Arip Syaripudin Nur, Yong Je Kim, Chang-Wook Lee

и другие.

Remote Sensing, Год журнала: 2022, Номер 14(17), С. 4416 - 4416

Опубликована: Сен. 5, 2022

Plumas National Forest, located in the Butte and counties, has experienced devastating wildfires recent years, resulting substantial economic losses threatening safety of people. Mapping damaged areas assessing wildfire susceptibility are necessary to prevent, mitigate, manage wildfires. In this study, a map was generated using CNN metaheuristic optimization algorithms (GWO ICA) based on images by The locations were identified damage proxy (DPM) technique from Sentinel-1 synthetic aperture radar (SAR) data collected 2016 2020. DPMs’ depicting similar fire perimeters obtained California Department Forestry Fire Protection (CAL FIRE). Data regarding divided into training set (50%) for modeling testing accuracy models. Sixteen conditioning factors, categorized as topographical, meteorological, environmental, anthropological selected construct models evaluated area under receiver operating characteristic (ROC) curve (AUC) root mean square error (RMSE) analysis. evaluation results revealed that hybrid-based CNN-GWO model (AUC = 0.974, RMSE 0.334) exhibited better performance than 0.934, 0.780) CNN-ICA 0.950, 0.350) Therefore, we conclude optimizing with metaheuristics considerably increased reliability mapping study area.

Язык: Английский

Процитировано

23