Exploring the Performance and Interpretability of an Enhanced Data-Driven Model to Assess Surface Flooding Susceptibility DOI Open Access

Chenlei Ye,

Zongxue Xu, Weihong Liao

и другие.

Sustainability, Год журнала: 2025, Номер 17(7), С. 3065 - 3065

Опубликована: Март 30, 2025

The effects of climate change and increasing urbanization mean that urban areas are facing a greater risk serious flooding. paper aimed to adopt data-driven approach capture surface flood-prone features, providing basis for flood susceptibility. This research developed an enhanced framework En-XGBoost, which consists three modules: the core module, preprocessing postprocessing module. Data augmentation, random extraction strategies, local enhancement were introduced improve model’s performance. En-XGBoost was tested in Fuzhou, China. main findings as follows: (1) Neighborhood information strategy outperformed extracting detailed producing clearer boundaries between different susceptibility levels, refining areas. (2) Crucial explanatory variables identified major drivers risk, with location-specific factors influencing causes, necessitating localized analysis specific sites. (3) enhancement, data strategies improved model performance, augmentation proving more effective stronger models having limited impact on weaker ones. Model performance requires appropriate alignment complexity complexity. provided support capturing features.

Язык: Английский

Integration of hydrogeological data, GIS and AHP techniques applied to delineate groundwater potential zones in sandstone, limestone and shales rocks of the Damoh district, (MP) central India DOI
Kanak N. Moharir,

Chaitanya B. Pande,

Vinay Kumar Gautam

и другие.

Environmental Research, Год журнала: 2023, Номер 228, С. 115832 - 115832

Опубликована: Апрель 11, 2023

Язык: Английский

Процитировано

122
Groundwater potential zone mapping using GIS and Remote Sensing based models for sustainable groundwater management DOI Creative Commons
Abdur Rehman, Fakhrul Islam, Aqil Tariq

и другие.

Geocarto International, Год журнала: 2024, Номер 39(1)

Опубликована: Янв. 1, 2024

The present research is conducted in the southern region of Khyber Pakhtunkhwa, Pakistan, to identify groundwater potential zones (GWPZ). We used three models including Weight Evidence (WOE), Frequency Ratio (FR), and Information Value (IV) with twelve parameters (elevation, slope, aspect, curvature, drainage network, LULC, precipitation, geology, Lineament, NDVI, road, soil texture, that have been prepared integrated into ArcGIS 10.8. reliability applied models' results was validated using Area Under Receiver Operating Characteristics (AUROC). GWPZ were reclassified five classes, i.e. very low, medium, high, high zone. area occupied by mentioned classes WOE are low (10.14%), (19.58%), medium (26.75%), (27.10%), (16.40%), while FR (20.93%), (32.38%), (18.92%), (13.13%), (14.61%) IV (14.41%), (17.17%), (29.01%), (25.85%), High (13.53%). Success Rate Curve WOE, FR, 0.86, 0.91, 0.87, Predicted values 0.89, 0.93, 0.90, respectively. revealed all statistical performed well delineate GWPZ. However, use technique strongly encouraged evaluate GWPZ, its findings especially useful for managing resources urban planning. Our approaches assessing mapping can be any similar scenarios recommended as a helpful tool policymakers manage groundwater.

Язык: Английский

Процитировано

18
Mapping groundwater potential zone by robust machine learning algorithm & remote sing techniques in agriculture dominated area, Bangladesh. DOI Creative Commons
M. M. Shah Porun Rana, Muhammad Tauhidur Rahman, Md. Fuad Hassan

и другие.

Cleaner Water, Год журнала: 2025, Номер unknown, С. 100064 - 100064

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

2
Groundwater potential mapping using multi-criteria decision, bivariate statistic and machine learning algorithms: evidence from Chota Nagpur Plateau, India DOI Creative Commons
Md Hasanuzzaman, Mehedi Hasan Mandal, Md Hasnine

и другие.

Applied Water Science, Год журнала: 2022, Номер 12(4)

Опубликована: Март 9, 2022

Abstract Increased consumption of water resource due to rapid growth population has certainly reduced the groundwater storage beneath earth which leads certain challenges human being in recent time. For optimal management this vital resource, exploration potential zone (GWPZ) become essential. We have applied Analytical Hierarchy Process (AHP), Frequency Ratio (FR) and two machine learning techniques specifically Random Forest (RF) Naïve Bayes (NB) here delineate GWPZ Gandheswari River Basin Chota Nagpur Plateau, India. To achieve goal study, twelve factors that determine occurrence been selected for inter-thematic correlations overlaid with location wells. These include elevation, drainage density, slope, lithology, geomorphology, topographical wetness index (TWI), distance from river, rainfall, lineament Normalized Difference Vegetation Index (NDVI), soil, Land use cover (LULC). A total 170 points including 85 well site non-well randomly allocated into parts: training testing at share 70:30. The implemented methods significantly provided five GWPZs Very Good (VG), (G), Moderate (M), Poor (P) (VP) high acceptable accuracy. study also finds rainfall elevation greater importance shaping than LULC, NDVI, etc. Model performance tested receiver operator characteristics (ROC), Accuracy (ACC), Kappa Coefficient, MAE, RMSE, etc., methods. Area under curve (AUC) ROC revealed accuracy level AHP, FR, RF NB is 78.8%, 81%, 85.3% 85.5, respectively. coupled AHP FR unveil effective delineation area said river basin by genetically offers low primary porosity lithological constrains. Therefore, can be helpful watershed identifying appropriate wells future.

Язык: Английский

Процитировано

59
GIS-Based Frequency Ratio and Analytic Hierarchy Process for Forest Fire Susceptibility Mapping in the Western Region of Syria DOI Open Access
Hazem Ghassan Abdo, Hussein Almohamad, Ahmed Abdullah Al Dughairi

и другие.

Sustainability, Год журнала: 2022, Номер 14(8), С. 4668 - 4668

Опубликована: Апрель 13, 2022

Forest fires are among the most major causes of global ecosystem degradation. The integration spatial information from various sources using statistical analyses in GIS environment is an original tool managing spread forest fires, which one significant natural hazards western region Syria. Moreover, Syria characterized by a lack data to assess fire susceptibility as consequences current war. This study aimed conduct performance comparison frequency ratio (FR) and analytic hierarchy process (AHP) techniques delineating distribution Al-Draikich region, located An inventory map historical events was produced spatially digitizing 32 incidents during summers 2019, 2020, 2021. were divided into training dataset with 70% (22 events) test 30% (10 events). Subsequently, FR AHP used associate set 13 driving factors: slope, aspect, curvature, elevation, Normalized Difference Vegetation Index (NDVI), Moisture (NDMI), Topographic Wetness (TWI), rainfall, temperature, wind speed, TWI, distance settlements, rivers roads. accuracy maps resulting modeling checked validation receiver operating characteristics (ROC) curves area under curve (AUC). method AUC = 0.864 achieved highest value compared 0.838. outcomes this assessment provide constructive insights for adopting management strategies area, especially light

Язык: Английский

Процитировано

50
Machine learning and GIS-RS-based algorithms for mapping the groundwater potentiality in the Bundelkhand region, India DOI
Mukesh Kumar, Pitam Singh,

Priyamvada Singh

и другие.

Ecological Informatics, Год журнала: 2023, Номер 74, С. 101980 - 101980

Опубликована: Янв. 5, 2023

Язык: Английский

Процитировано

40
Mapping of groundwater potential zones in a drought prone Marathwada Region using frequency ratio and statistical index methods, India DOI Creative Commons
Uttam Pawar, Worawit Suppawimut, Upaka Rathnayake

и другие.

Results in Engineering, Год журнала: 2024, Номер 22, С. 101994 - 101994

Опубликована: Март 21, 2024

Groundwater potential investigation is one of the most significant and crucial aspects in arid semi-arid regions like Marathwada. Marathwada often susceptible to water scarcity drought due monsoon variability. Therefore, it indispensable identify groundwater zones (GWPZs) meet human needs. The present research used frequency ratio (FR) statistical index (SI) models recognize GWPZs region. Accordingly, eight factors such as geomorphology, elevation, slope, geology, drainage density, rainfall, soil types, land use/land cover applied delineate GWPZ About 309 wells study area were randomly subset into training (70%) testing (30%) datasets. maps classified five (very low, moderate, high, very high). FR showed about 53% areas having high potentiality groundwater, whereas 11% comes under a low groundwater. However, SI model 65% while 8% fall category medium substantial observed central eastern part region with elevation pediment pediplain complex rainfall covered Chromic Vertisols (Vc) soil. results receiver operating characteristic (ROC) curve (AUC) revealed that performance (AUC = 75.14%) better than (61.20%). suitable identity map outcomes derived from this are appropriate for resources management planning

Язык: Английский

Процитировано

8
Novel hybrid models to enhance the efficiency of groundwater potentiality model DOI Creative Commons
Swapan Talukdar, Javed Mallick, Showmitra Kumar Sarkar

и другие.

Applied Water Science, Год журнала: 2022, Номер 12(4)

Опубликована: Март 9, 2022

Abstract The present study aimed to create novel hybrid models produce groundwater potentiality (GWP) in the Teesta River basin of Bangladesh. Six ensemble machine learning (EML) algorithms, such as random forest (RF), subspace, dagging, bagging, naïve Bayes tree (NBT), and stacking, coupled with fuzzy logic (FL) a ROC-based weighting approach have been used for creating integrated GWP. GWP was then verified using both parametric nonparametric receiver operating characteristic curves (ROC), empirical ROC (eROC) binormal curve (bROC). We conducted an RF-based sensitivity analysis compute relevancy conditioning variables modeling. very high potential regions were predicted 831–1200 km 2 521–680 areas based on six EML models. Based area under ROC, NBT (eROC: 0.892; bROC: 0.928) model outperforms rest GPMs considered next step turned into crisp layers membership function, approach. Subsequently four operators assimilate layers, including AND, OR, GAMMA0.8, GAMMA 0.9, well GAMMA0.9. Thus, we created FL model. results eROC bROC showed that 0.9 operator outperformed other operators-based terms accuracy. According validation outcomes, performance. will aid enhancing efficiency preparing viable planning management.

Язык: Английский

Процитировано

34
Mapping of groundwater productivity potential with machine learning algorithms: A case study in the provincial capital of Baluchistan, Pakistan DOI
Umair Rasool, Xinan Yin, Zongxue Xu

и другие.

Chemosphere, Год журнала: 2022, Номер 303, С. 135265 - 135265

Опубликована: Июнь 9, 2022

Язык: Английский

Процитировано

30
Groundwater potential mapping in the Central Highlands of Vietnam using spatially explicit machine learning DOI

Tran Xuan Bien,

Abolfazl Jaafari, Tran Van Phong

и другие.

Earth Science Informatics, Год журнала: 2023, Номер 16(1), С. 131 - 146

Опубликована: Янв. 3, 2023

Язык: Английский

Процитировано

18