Quantitative assessment of Land use/land cover changes in a developing region using machine learning algorithms: A case study in the Kurdistan Region, Iraq

Heliyon, Год журнала: 2023, Номер 9(11), С. e21253 - e21253

Опубликована: Окт. 24, 2023

The identification of land use/land cover (LULC) changes is important for monitoring, evaluating, and preserving natural resources. In the Kurdistan region, utilization remotely sensed data to assess effectiveness machine learning algorithms (MLAs) LULC classification change detection analysis has been limited. This study monitors analyzes in area from 1991 2021 using a quantitative approach with multi-temporal Landsat imagery. Five MLAs were applied: Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), Extreme Gradient Boosting (XGBoost). results showed that RF algorithm produced most accurate maps three-decade period, accompanied by high kappa coefficient (0.93-0.97) compared SVM (0.91-0.95), ANN (0.91-0.96), KNN (0.92-0.96), XGBoost (0.92-0.95) algorithms. Consequently, classifier was implemented categorize all obtainable satellite images. Socioeconomic throughout these transition periods revealed results. Rangeland barren areas decreased 11.33 % (-402.03 km2) 6.68 (-236.8 km2), respectively. transmission increases 13.54 (480.18 3.43 (151.74 0.71 (25.22 occurred agricultural land, forest, built-up areas, outcomes this contribute significantly monitoring developing regions, guiding stakeholders identify vulnerable better use planning sustainable environmental protection.

Язык: Английский

---

Forecasting of flash flood susceptibility mapping using random forest regression model and geographic information systems

Heliyon, Год журнала: 2024, Номер 10(13), С. e33982 - e33982

Опубликована: Июль 1, 2024

Flash floods, rapid and devastating inundations of water, are increasingly linked to the intensifying effects climate change, posing significant challenges for both vulnerable communities sustainable environmental management. The primary goal this research is investigate predict a Flood Susceptibility Map (FSM) Ibaraki prefecture in Japan. This utilizes Random Forest (RF) regression model GIS, incorporating 11 variables (involving elevation, slope, aspect, distance stream, river, road, land cover, topographic wetness index, stream power plan profile curvature), alongside dataset comprising 224 instances flooded non-flooded locations. data was randomly classified into 70 % training set development, with remaining 30 used validation through Receiver Operating Characteristics (ROC) curve analysis. resulting map indicated that approximately two-thirds as exhibiting low very flood susceptibility, while one-fifth region categorized high susceptibility. Furthermore, RF achieved noteworthy an area under ROC 99.56 %. Ultimately, FSM serves crucial tool policymakers guiding appropriate spatial planning mitigation strategies.

Язык: Английский

---

Future groundwater potential mapping using machine learning algorithms and climate change scenarios in Bangladesh

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Май 6, 2024

Abstract The aim of the study was to estimate future groundwater potential zones based on machine learning algorithms and climate change scenarios. Fourteen parameters (i.e., curvature, drainage density, slope, roughness, rainfall, temperature, relative humidity, lineament land use cover, general soil types, geology, geomorphology, topographic position index (TPI), wetness (TWI)) were used in developing algorithms. Three artificial neural network (ANN), logistic model tree (LMT), regression (LR)) applied identify zones. best-fit selected ROC curve. Representative concentration pathways (RCP) 2.5, 4.5, 6.0, 8.5 scenarios precipitation for modeling change. Finally, identified 2025, 2030, 2035, 2040 best RCP models. According findings, ANN shows better accuracy than other two models (AUC: 0.875). predicted that 23.10 percent very high zones, whereas 33.50 extremely forecasts values under different (RCP2.6, RCP4.5, RCP6, RCP8.5) using an spatial distribution maps each scenario. sixteen generated Government officials may utilize study’s results inform evidence-based choices water management planning at national level.

Язык: Английский

---

Leveraging GIS-based AHP, remote sensing, and machine learning for susceptibility assessment of different flood types in peshawar, Pakistan

Journal of Environmental Management, Год журнала: 2024, Номер 371, С. 123094 - 123094

Опубликована: Ноя. 2, 2024

Язык: Английский

---

Meteorological drought assessment in northern Bangladesh: A machine learning-based approach considering remote sensing indices

Ecological Indicators, Год журнала: 2023, Номер 157, С. 111233 - 111233

Опубликована: Ноя. 14, 2023

Meteorological drought, driven by inadequate precipitation, has significant repercussions for water resources, agriculture, and human well-being. This study conducted an extensive assessment of meteorological drought in northern Bangladesh, employing remote sensing indices machine learning techniques. The main aim was to evaluate occurrences Bangladesh from 2010 2019, utilizing seven parameters a model. Utilizing Random Forest (RF) model, this employed the Standardized Precipitation Index (SPI) as dependent variable independent variables. Through methodology, assessed significance these generated model integrated them, culminating creation distribution map spanning 2019. approach offers novel insights probing interplay collective impacts indices, shedding light on previously unexplored aspects regional patterns Bangladesh. major findings showed that precipitation strongly influenced both short-term long-term episodes. Moreover, land surface-related such Evapotranspiration (ET) Normalized Difference Water (NDWI), exhibited more pronounced impact occurrences, while vegetation-related like Multi-band Drought (NMDI) Vegetation (NDVI) demonstrated greater influence over events. During timeframe, Rajshahi division experienced frequent extreme severe Moderate droughts abnormally dry conditions were widespread. Barind tract area consistently faced moderate droughts, with exceptions 2011, 2014, On average, 5% region had than 12% during decade. Long-term indicators (SPI 6 SPI 9) higher frequencies compared 1 3), emphasizing prolonged rainfall deficits relevance longer time frames dynamics. RF strong performance accuracy ranging 81% 95%. Low prediction errors (RMSE 6% 31%) high out-of-bag (OOB) 76% 98% highlighted its accuracy. F1 score exceeded 76%, indicating precision recall. Cross-validation values ranged 78% 94%, affirming reliable generalization new data. Incorporating findings, contributes valuable formulation targeted mitigation strategies It is imperative note scope confined generalizing other regions should be exercised caution. Nevertheless, research methodology can serve future studies related fields, advancing knowledge how assess using methods.

Язык: Английский

---

Coupling of machine learning and remote sensing for soil salinity mapping in coastal area of Bangladesh

Scientific Reports, Год журнала: 2023, Номер 13(1)

Опубликована: Окт. 10, 2023

Soil salinity is a pressing issue for sustainable food security in coastal regions. However, the coupling of machine learning and remote sensing was seldom employed soil mapping areas Bangladesh. The research aims to estimate level southwestern region Using Landsat OLI images, 13 indicators were calculated, 241 samples data collected from secondary source. This study applied three distinct models (namely, random forest, bagging with artificial neural network) salinity. best model subsequently used categorize zones into five groups. According findings, network has highest area under curve (0.921), indicating that it most potential predict detect zones. high zone covers an 977.94 km2 or roughly 413.51% total area. additional data, moderate (686.92 km2) 30.56% Satkhira, while low (582.73 25.93% Since increased adversely affects human health, agricultural production, etc., study's findings will be effective tool policymakers integrated management

Язык: Английский

---

Flood susceptibility modelling of the Teesta River Basin through the AHP-MCDA process using GIS and remote sensing

Natural Hazards, Год журнала: 2024, Номер unknown

Опубликована: Июнь 8, 2024

Язык: Английский

---

Farmers’ climate change perception, impacts and adaptation strategies in response to drought in the Northwest area of Bangladesh

Climate Services, Год журнала: 2025, Номер 38, С. 100540 - 100540

Опубликована: Фев. 3, 2025

Язык: Английский

---

Enhancing flood susceptibility mapping in Meghna River basin by introducing ensemble Naive Bayes with stacking algorithms

Geomatics Natural Hazards and Risk, Год журнала: 2025, Номер 16(1)

Опубликована: Фев. 13, 2025

Язык: Английский

---

Future flood susceptibility mapping under climate and land use change

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 11, 2025

Язык: Английский

---