Machine learning models for gully erosion susceptibility assessment in the Tensift catchment, Haouz Plain, Morocco for sustainable development

Journal of African Earth Sciences, Journal Year: 2024, Volume and Issue: 213, P. 105229 - 105229

Published: March 11, 2024

Gully erosion is a widespread environmental danger, threatening global socio-economic stability and sustainable development. This study comprehensively applied seven machine learning (ML) models including SVM, KNN, RF, XGBoost, ANN, DT, LR, evaluated gully susceptibility in the Tensift catchment predict it within Haouz plain, Morocco. To ensure reliability of findings, employed robust combination inventory, sentinel images, Digital Surface Model. Eighteen predictors, encompassing topographical, geomorphological, environmental, hydrological factors, were selected after multicollinearity analyses. The revealed that approximately 28.18% at very high risk erosion. Furthermore, 15.13% 31.28% are categorized as low respectively. These findings extend to where 7.84% surface area highly risking erosion, while 18.25% 55.18% characterized areas. gauge performance ML models, an array metrics specificity, precision, sensitivity, accuracy employed. highlights XGBoost KNN most promising achieving AUC ROC values 0.96 0.93 test phase. remaining namely RF (AUC = 0.89), LR 0.80), SVM 0.81), DT 0.86), ANN 0.78), also displayed commendable performance. novelty this research its innovative approach combat through cutting edge offering practical solutions for watershed conservation, management, prevention land degradation. insights invaluable addressing challenges posed by region, beyond geographical boundaries can be used defining appropriate mitigation strategies local national scale.

Language: Английский

---

Towards multi-model soil erosion modelling: An evaluation of the erosion potential method (EPM) for global soil erosion assessments

CATENA, Journal Year: 2023, Volume and Issue: 234, P. 107596 - 107596

Published: Oct. 14, 2023

Soil erosion is expected to increase in the future due climate change. models are useful tools that can be used by decision makers and other stakeholders deal with soil problems or implementation of protection measures. Most modelling applications using Universal Loss Equation (USLE)-type models. In this study, we evaluate applicability Erosion Potential Model (EPM) its modified version (mEPM) for estimation gross net rates at a global scale. The sensitivity analysis shows model results have highest variability (land cover) coefficient followed erodibility parameter. models' evaluations indicate EPM cannot applied cold regions while mEPM overcomes issue. based on were 1.5–2.5 times larger than ones obtained from mEPM. Increasing number catchment properties as inputs may help improving performance tested Moreover, comparison losses long-term suspended sediment yield data 116 catchments located around globe indicates median bias less 10%, although 1/3 was above 100%. Furthermore, direct such USLE-type not possible since do take into consideration processes slumps gully just sheet rill erosion. Therefore, expected, higher compared Hence, mEPM, despite limitations, could regarded an interesting approach describing should further small- medium-sized various zones.

Language: Английский

---

Modelling Water Erosion and Mass Movements (Wet) by Using GIS-Based Multi-Hazard Susceptibility Assessment Approaches: A Case Study—Kratovska Reka Catchment (North Macedonia)

Atmosphere, Journal Year: 2023, Volume and Issue: 14(7), P. 1139 - 1139

Published: July 12, 2023

Kratovska Reka is a short (17.3 km) left tributary of Kriva Reka, whose watershed (68.5 km2) located on the northwestern slopes Osogovo Mountains (North Macedonia). Due to favorable natural conditions and anthropogenic factors, catchment under high risk hazards, especially water erosion landslide occurrences. For this reason, paper presents an approach modelling potential areas susceptible above-mentioned hydro-meteorological hazards in River catchment. Firstly, study analyzed main geographical features that contribute intensive processes area. Then, using Gavrilović EPM method, average value 0.56 was obtained for coefficient Z, indicating prone risk. Furthermore, by susceptibility analysis (LSA), terrains landslides were identified. The results shows 1/3 very mass movements wet (landslides). According combined multi-hazard model, 3.13% total area both at severe erosion. significantly endangered excessive (39.86%), steep valley sides, i.e., are completely exposed, sparse vegetation, open effects distribution/concentration rainfall amounts throughout year. Identifying locations with highest serves as initial step defining implementing appropriate mitigation measures across local regional scales, thus enhancing overall resilience environmental challenges.

Language: Английский

---

NITIAL RESULTS OF COMPARATIVE ASSESSMENT OF SOIL EROSION INTENSITY USING THE WIntErO MODEL: A CASE STUDY OF POLIMLJE AND SHIRINDAREH DRAINAGE BASINS

Carpathian Journal of Earth and Environmental Sciences, Journal Year: 2023, Volume and Issue: 18(2), P. 385 - 404

Published: Aug. 31, 2023

This work aims to determine the current state of sediment production and propose land use measures that will affect reducing intensity soil erosion for areas Polimlje drainage basin on territories Montenegro Serbia, small Shirindareh sub-basin Iran. The approach is based field laboratory methods, which are processed by Web-based Intensity Erosion Outflow (WIntErO)model used calculate intensity. By using computer-graphical method "WIntErO" software, in study intensity, surface values (watershed surface, between isohypsies, etc.) length, i.e. deviations from map (length main watercourse, length watershed line, very precisely, was not case before when mechanical instruments, planimeters curvimeters. new WIntErO model an integrated computer-graphic program package third-generation earlier generations modelling tools "River basins", IntErO calculates amount sediment, erosion, as well maximum runoff basin, according EPM Gavrilović.During procedure, accuracy assessment conducted with measurements reservoir deposition. These were performed April 2017 professional hydrographic recording equipment, following same methodology 2012. measurement point locations a GPS receiver Trimble R6 base station. reservoir's depth measured single-frequency portable echo sounder, specifically Odom Hydro Track. initial shows fairly acceptable results implemented both areas. Z coefficient ranging 0.01 1.00 observed period indicate river basins delineate varying levels susceptibility water processes—ranging low moderate high risk—within studied basins. Based analysis, we found average 331.78 m³ km⁻² year⁻¹ per square kilometer. For sub-basins, actual losses kilometer 201 km⁻². Given these findings, it evident require prompt implementation conservation measures.

Language: Английский

---

An Index for Snowmelt-Induced Landslide Prediction for Zavoj Lake, Serbia

Atmosphere, Journal Year: 2024, Volume and Issue: 15(3), P. 256 - 256

Published: Feb. 21, 2024

In February 1963, a huge landslide (ca. 1,950,000 m3) blocked the Visočica River and, thus, formed Zavoj Lake. The primary objective of this research was to investigate importance snowmelt in relation occurrence and define critical climatic conditions that may trigger massive winter landslides. We used monthly precipitation average maximum temperature data from meteorological stations basin (Dojkinci) immediate proximity Lake (Pirot, Dimitrovgrad Topli Do) as inputs Snow-Melt Landslide (SML) index. It considers summed for previous months continuously have an below 0 °C. According method, event at stands out among all other values past 72 years. After applying SML index, showed >300 mm which we consider threshold value potential landslides appearance. addition data, applied index Coordinated Regional Downscaling Experiment (CORDEX) regional climate model outputs region 2022 2100. As expected, change will influenced values, especially during winter. Conversely, study area is experiencing drastic changes land use caused by depopulation, leading reduced risk basin. suggest future make it more likely experience extreme summer events, might large method can be implemented landscapes snowy winters, providing information timely manner so local residents react properly when probability rises. grounded essential principles, provides tailor-made, data-driven methodology applicable across varied geographical settings. Its utility extends mitigating hydro-meteorological hazards on scales ranging national scales, offering diverse effective early warning solutions.

Language: Английский

---

Universal Snow Avalanche Modeling Index Based on SAFI–Flow-R Approach in Poorly-Gauged Regions

ISPRS International Journal of Geo-Information, Journal Year: 2024, Volume and Issue: 13(9), P. 315 - 315

Published: Sept. 1, 2024

Most high-mountain regions worldwide are susceptible to snow avalanches during the winter or all year round. In this study, a Universal Snow Avalanche Modeling Index is developed, suitable for determining avalanche hazard in mountain regions. The first step research collection of data field and their processing geographic information systems remote sensing. period 2023–2024, were mapped field, later, as points (GIS) overlapped with dominant natural conditions study area. second involves main criteria (snow cover, terrain slope, land use) evaluating values obtain Formation (SAFI). Thresholds obtained through formation inventory used develop SAFI index. index applied aim identifying locations (source areas). calculation include Normalized Difference (NDSI > 0.6), slope (20–60°) use (pastures, meadows). third presents analysis meteorological (winter precipitation air temperature). fourth modeling propagation (simulation) other parts Flow-R software 2.0. results show that 282.9 km2 area (Šar Mountains, Serbia) avalanches, thickness potentially triggered layer being 50 cm. With 5 m thick snowpack, 299.9 would be susceptible. validation using ROC-AUC method confirms very high predictive power (0.94). SAFI–Flow-R approach offers which no available, representing an advance areas where historical do not exist. can planning, zoning vulnerable areas, adopting adequate environmental protection measures.

Language: Английский

---

Spatial Distribution, Temporal Behaviour, and Trends of Rainfall Erosivity in Central Italy Using Coarse Data

Water, Journal Year: 2025, Volume and Issue: 17(6), P. 801 - 801

Published: March 11, 2025

This study examines the spatio-temporal dynamics of rainfall erosivity, R, in Umbria region (central Italy), based on a 20-year dataset 30 min precipitation records from 54 stations. Using RUSLE2 framework, models varying complexity were evaluated to estimate R-factor: original model (Model A), and solely event depth he or daily hd. All show consistency spatial temporal patterns higher erosivity is observed southern northwestern areas, while summer contributes most annual due high average intensity events. Trend analyses indicate stationarity across Compared Model A (mean 1840 MJ mm ha−1 h−1 y−1), underestimate R-factor by about 15%, whereas derived hd-dependent almost equivalent. The also approximately 20% than that previous conducted more limited dataset. likely reason for this difference appears be formula used estimating R-factor. highlights practicality simplified models, which offer viable alternative contexts where high-resolution data are unavailable. It demonstrates benefits denser station networks longer observation periods, particularly regions characterised complex terrains.

Language: Английский

---

Geospatial Modeling of Suitable Sites for Solar Power Plants Based on GIS and BWM: A Case Study of the City of Kraljevo, Serbia

Lecture notes in operations research, Journal Year: 2025, Volume and Issue: unknown, P. 129 - 142

Published: Jan. 1, 2025

Language: Английский

---

Soil Erosion Estimation Using the E.P.M Model in the Ouaouizeght River Basin—Morocco

Environmental science and engineering, Journal Year: 2025, Volume and Issue: unknown, P. 285 - 305

Published: Jan. 1, 2025

Language: Английский

---

Erosion susceptibility mapping of a loess-covered region using Analytic Hierarchy Process – A case study: Kalat-e-Naderi, northeast Iran

Hungarian Geographical Bulletin, Journal Year: 2024, Volume and Issue: 72(4), P. 339 - 364

Published: Jan. 12, 2024

In this study, the Analytic Hierarchy Process (AHP) is applied to generate erosion susceptibility maps in four basins of Kalat-e-Naderi county, namely Archangan, Kalat, Qaratigan, and Chahchaheh basins, situated northeast Iran. The region characterized by a partial coverage loess. Given agricultural significance loess its erosion, research focuses specifically on regions covered Geographic Information System (GIS) tools, including ArcMap Quantum (QGIS), were utilized facilitate creation maps. Seven factors, slope, aspect, elevation, drainage density, lithology, Normalized Difference Vegetation Index (NDVI), precipitation selected for consideration. Recognizing variability vegetation cover across different seasons, seasonal data specified factors employed. Consequently, generated basis. Pairwise comparison tables revealed that precipitation, slope emerged as dominant contributing region. resultant distinctly delineate with higher values, unresistant lithology (such loess, high porosity permeability), steeper slopes, exhibiting heightened (Archangan Kalat basins). credibility findings was examined through on-site observations. outcomes study may provide pertinent insights decision-makers planners. This information can be effectively employed formulating strategies aimed at conserving soil quality areas vulnerable hazards.

Language: Английский

---