Amanuel Kumsa Bojer
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Amanuel Kumsa Bojer has a total of 76 co-authors. Here are the top co-authors by number of shared publications:
- Ayad M. Fadhil Al-Quraishi
Petroleum and Mining Engineering Department, Tishk International University, Erbil, Iraq
8 shared publications Saralees Nadarajah
Department of Mathematics, University of Manchester, Manchester, UK
5 shared publicationsTaye Girma Debelee
4 shared publicationsWorku Gachena Negera
4 shared publicationsDesta Jula Bekalo
3 shared publicationsKena Wendimu Gebissa
3 shared publications
Latest Publications
- Open Access
- 2 April 2026
- Published by IWA Publishing in Water Science & Technology
- Vol. 93 (8) , 1144-1168
Development and deployment of a groundwater quality prediction tool for Béchar, Southwest Algeria: benchmarking seven machine learning algorithms
Four-panel flow diagram showing the study workflow: Béchar aquifer dataset input with 621 groundwater samples and 11 physicochemical parameters; seven supervised machine learning models benchmarked using Bayesian hyperparameter optimisation and five-fold cross-validation; ANN achieving 99.47% accuracy and SVM 97.88%; deployment as a MATLAB-based groundwater quality prediction GUI tool. Accurate prediction of groundwater quality is essential for environmental monitoring and public health protection, particularly in arid regions such as Béchar in southwest Algeria. This study applied a root mean square–based water quality index (RMS–WQI) to evaluate groundwater quality using 621 samples characterized by physicochemical parameters, including pH, electrical conductivity, total dissolved solids, major cations, major anions, and nitrate. Seven supervised machine learning algorithms K-nearest neighbors, artificial neural network (ANN), support vector machine (SVM), ensemble trees (EN), discriminant analysis, Naïve Bayes, and decision trees were trained and optimized in MATLAB using the Classification Learner Toolbox with Bayesian optimization and fivefold cross-validation. Among the tested models, ANN and SVM achieved the highest predictive performance, with accuracies of 99.47 and 97.88%, respectively, along with superior precision, recall, F1-score, and Cohen's Kappa values, indicating strong agreement with observed RMS–WQI classes. Compared to conventional RMS–WQI assessment and previously reported nonoptimized models, the proposed framework demonstrates improved classification accuracy and robustness. Additionally, an operational graphical user interface was developed to facilitate rapid groundwater quality estimation using routine measurements. The findings highlight the effectiveness of optimized ANN and SVM models as reliable decision support tools for groundwater quality management in data-scarce arid environments.
- 2 April 2026
- Published by IWA Publishing in Water Science & Technology
- Vol. 93 (8) , 1144-1168
Open AccessDevelopment and deployment of a groundwater quality prediction tool for Béchar, Southwest Algeria: benchmarking seven machine learning algorithms
Four-panel flow diagram showing the study workflow: Béchar aquifer dataset input with 621 groundwater samples and 11 physicochemical parameters; seven supervised machine learning models benchmarked using Bayesian hyperparameter optimisation and five-fold cross-validation; ANN achieving 99.47% accuracy and SVM 97.88%; deployment as a MATLAB-based groundwater quality prediction GUI tool. Accurate prediction of groundwater quality is essential for environmental monitoring and public health protection, particularly in arid regions such as Béchar in southwest Algeria. This study applied a root mean square–based water quality index (RMS–WQI) to evaluate groundwater quality using 621 samples characterized by physicochemical parameters, including pH, electrical conductivity, total dissolved solids, major cations, major anions, and nitrate. Seven supervised machine learning algorithms K-nearest neighbors, artificial neural network (ANN), support vector machine (SVM), ensemble trees (EN), discriminant analysis, Naïve Bayes, and decision trees were trained and optimized in MATLAB using the Classification Learner Toolbox with Bayesian optimization and fivefold cross-validation. Among the tested models, ANN and SVM achieved the highest predictive performance, with accuracies of 99.47 and 97.88%, respectively, along with superior precision, recall, F1-score, and Cohen's Kappa values, indicating strong agreement with observed RMS–WQI classes. Compared to conventional RMS–WQI assessment and previously reported nonoptimized models, the proposed framework demonstrates improved classification accuracy and robustness. Additionally, an operational graphical user interface was developed to facilitate rapid groundwater quality estimation using routine measurements. The findings highlight the effectiveness of optimized ANN and SVM models as reliable decision support tools for groundwater quality management in data-scarce arid environments.
- 22 February 2026
- Published by Springer Nature in Natural Resources Research
- Vol. 35 (3) , 2091-2124
Assessing and Predicting Land-Use/Land-Cover Changes, and Land Surface Temperature Impacts on Gilgel Gibe Water Resources, Ethiopia
- 22 February 2026
- Published by Springer Nature in Natural Resources Research
- Vol. 35 (3) , 2091-2124
Assessing and Predicting Land-Use/Land-Cover Changes, and Land Surface Temperature Impacts on Gilgel Gibe Water Resources, Ethiopia
- Top Cited PapersOpen Access
- 28 January 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Remote sensing assessment of vegetation and moisture dynamics in semi-arid regions
This study investigates changes in land use and vegetation cover in the Oued Louza watershed, Sidi Bel Abbès province, Algeria, from 1987 to 2020, using remote sensing and Geographic Information Systems (GIS) to assess spatio-temporal dynamics. The analysis employed Landsat-derived vegetation and moisture indices, including the Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), and Normalized Difference Water Index (NDWI), along with the Topographic Wetness Index (TWI) derived from the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM). Results show a dramatic decline in vegetation cover, from 42% in 1987 to 10% in 2020, a 32% decrease, while urban areas expanded by 27%. The reduction in vegetation was linked to a 22% decrease in rainfall and a 6.5% reduction in relative humidity, both of which exacerbated vegetation loss and soil moisture decline. The study also revealed a strong relationship between areas with higher moisture retention and denser vegetation, as indicated by TWI values. Land use and land cover classification was validated with a kappa coefficient of 0.84 in 1987 and 0.91 in 2020, confirming the accuracy of the analysis. A majority-voting technique was used to combine multiple spectral indices to improve classification reliability. Despite the methodology’s effectiveness, limitations exist, particularly the reliance on satellite-derived climatic data from the NASA POWER database, given the limited availability of ground-based meteorological stations in the region. Additionally, the spatial resolution of Landsat images may not capture small-scale land use changes, although it is suitable for large-scale assessments. The findings underscore the impact of both climatic and anthropogenic factors on vegetation dynamics and highlight the potential of remote sensing and GIS for land use and environmental monitoring in semi-arid regions. This study provides essential insights for sustainable land and water resource management, and future research could build on these findings by incorporating higher-resolution imagery, local meteorological data, and advanced machine learning techniques to enable more detailed land-use change predictions.
- 28 January 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Top Cited PapersOpen AccessRemote sensing assessment of vegetation and moisture dynamics in semi-arid regions
This study investigates changes in land use and vegetation cover in the Oued Louza watershed, Sidi Bel Abbès province, Algeria, from 1987 to 2020, using remote sensing and Geographic Information Systems (GIS) to assess spatio-temporal dynamics. The analysis employed Landsat-derived vegetation and moisture indices, including the Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), and Normalized Difference Water Index (NDWI), along with the Topographic Wetness Index (TWI) derived from the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM). Results show a dramatic decline in vegetation cover, from 42% in 1987 to 10% in 2020, a 32% decrease, while urban areas expanded by 27%. The reduction in vegetation was linked to a 22% decrease in rainfall and a 6.5% reduction in relative humidity, both of which exacerbated vegetation loss and soil moisture decline. The study also revealed a strong relationship between areas with higher moisture retention and denser vegetation, as indicated by TWI values. Land use and land cover classification was validated with a kappa coefficient of 0.84 in 1987 and 0.91 in 2020, confirming the accuracy of the analysis. A majority-voting technique was used to combine multiple spectral indices to improve classification reliability. Despite the methodology’s effectiveness, limitations exist, particularly the reliance on satellite-derived climatic data from the NASA POWER database, given the limited availability of ground-based meteorological stations in the region. Additionally, the spatial resolution of Landsat images may not capture small-scale land use changes, although it is suitable for large-scale assessments. The findings underscore the impact of both climatic and anthropogenic factors on vegetation dynamics and highlight the potential of remote sensing and GIS for land use and environmental monitoring in semi-arid regions. This study provides essential insights for sustainable land and water resource management, and future research could build on these findings by incorporating higher-resolution imagery, local meteorological data, and advanced machine learning techniques to enable more detailed land-use change predictions.
- Open Access
- 15 January 2026
- Published by Elsevier in Trees, Forests and People
- Vol. 23
Integrating geo-spatial analysis and gray relational approach for wildfire susceptibility mapping in Bale Mountain National Park, Ethiopia
- 15 January 2026
- Published by Elsevier in Trees, Forests and People
- Vol. 23
Open AccessIntegrating geo-spatial analysis and gray relational approach for wildfire susceptibility mapping in Bale Mountain National Park, Ethiopia
- Open Access
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Science, Measurement & Technology
- Vol. 20 (1)
Influence of Applied Voltage Waveforms on Ozone Generation and Operating Condition Optimisation in Surface DBD Reactors
Surface dielectric barrier discharge (SDBD) ozone generators are a promising technology for industrial water treatment as well as air purification; however, they are under‐studied. In this experimental study, three types of AC voltage waveforms (triangular, rectangular and sinusoidal) are compared to ozone generation and energy performance in the SDBD reactor. The optimal waveform was triangular, with an ozone concentration of 12 mg/L and an efficiency of 120 g/kWh. The second was the sinusoidal waveform (120 mg/L and 95 g/kWh, respectively). The most typical was the square waveform, with the rectangular one less effective. The optimum operating conditions are as follows (by response surface methodology (RSM)): V = 7.84 kV, f = 202.86 Hz (η = 114.76 g/kWh) and V = 8.99 kV, f = 849.7 Hz (OC = 132.163 mg/L). This study highlights the importance of optimal waveform selection and systematic parameter tuning to improve ozone generation performance while minimising energy consumption for practical industrial SDBD applications.
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Science, Measurement & Technology
- Vol. 20 (1)
Open AccessInfluence of Applied Voltage Waveforms on Ozone Generation and Operating Condition Optimisation in Surface DBD Reactors
Surface dielectric barrier discharge (SDBD) ozone generators are a promising technology for industrial water treatment as well as air purification; however, they are under‐studied. In this experimental study, three types of AC voltage waveforms (triangular, rectangular and sinusoidal) are compared to ozone generation and energy performance in the SDBD reactor. The optimal waveform was triangular, with an ozone concentration of 12 mg/L and an efficiency of 120 g/kWh. The second was the sinusoidal waveform (120 mg/L and 95 g/kWh, respectively). The most typical was the square waveform, with the rectangular one less effective. The optimum operating conditions are as follows (by response surface methodology (RSM)): V = 7.84 kV, f = 202.86 Hz (η = 114.76 g/kWh) and V = 8.99 kV, f = 849.7 Hz (OC = 132.163 mg/L). This study highlights the importance of optimal waveform selection and systematic parameter tuning to improve ozone generation performance while minimising energy consumption for practical industrial SDBD applications.
- Open Access
- 25 September 2025
- Published by Wiley in Journal of Flood Risk Management
- Vol. 18 (4)
Spatial Modeling of Flood Hazard in Addis Ababa Using Geographic Information System ( GIS ) and Information Gain Ratio ( IGR ) Method
Floods are one of nature's most disturbing catastrophes, resulting in infrastructure damage, property devastation, and mortality. In Addis Ababa, flooding has significantly impacted residents and caused millions' worth of property damage in the last decade alone. It is continuously threatening and affecting city residents. This study focused on the spatial modeling of floods and the identification of areas susceptible to flood hazards in the city. Geographic information system (GIS) techniques combined with the information gain ratio (IGR) method were employed in this study. Five major flood hazard factors were identified: elevation, slope, rainfall, drainage density, and distance from drainage channels. The results show that 1.3% (7.1 km 2 ) of the area is highly susceptible to floods, 29.4% (159 km 2 ) is highly susceptible to heavy rains, 56% (302 km 2 ) of the area is moderately susceptible, 12.5% (67.3 km 2 ) of the area has low susceptibility, and less than 1% (4.2 km 2 ) has very low susceptibility. Slope is the most influential factor (42.74%), followed by drainage density (28.21%), distance from drainage channels (18.8%), rainfall (7.69%), and elevation (2.56%). The sub‐cities of Nifas Silk Lafto and Akaki Kality are the most susceptible to flood hazards; areas with steep slopes trigger high runoff during heavy rainy periods and cause flood hazards on gentle slope surfaces. It is recommended that to improve the accuracy of identifying susceptible flood‐hazard locations, flooding simulation should be performed in conjunction with other variables and rainfall data (such as rainfall duration and intensity). Nevertheless, this research provides recommendations to municipal administration decision‐makers regarding strategic management in the prioritization of flood‐hazard zones.
- 25 September 2025
- Published by Wiley in Journal of Flood Risk Management
- Vol. 18 (4)
Open AccessSpatial Modeling of Flood Hazard in Addis Ababa Using Geographic Information System ( GIS ) and Information Gain Ratio ( IGR ) Method
Floods are one of nature's most disturbing catastrophes, resulting in infrastructure damage, property devastation, and mortality. In Addis Ababa, flooding has significantly impacted residents and caused millions' worth of property damage in the last decade alone. It is continuously threatening and affecting city residents. This study focused on the spatial modeling of floods and the identification of areas susceptible to flood hazards in the city. Geographic information system (GIS) techniques combined with the information gain ratio (IGR) method were employed in this study. Five major flood hazard factors were identified: elevation, slope, rainfall, drainage density, and distance from drainage channels. The results show that 1.3% (7.1 km 2 ) of the area is highly susceptible to floods, 29.4% (159 km 2 ) is highly susceptible to heavy rains, 56% (302 km 2 ) of the area is moderately susceptible, 12.5% (67.3 km 2 ) of the area has low susceptibility, and less than 1% (4.2 km 2 ) has very low susceptibility. Slope is the most influential factor (42.74%), followed by drainage density (28.21%), distance from drainage channels (18.8%), rainfall (7.69%), and elevation (2.56%). The sub‐cities of Nifas Silk Lafto and Akaki Kality are the most susceptible to flood hazards; areas with steep slopes trigger high runoff during heavy rainy periods and cause flood hazards on gentle slope surfaces. It is recommended that to improve the accuracy of identifying susceptible flood‐hazard locations, flooding simulation should be performed in conjunction with other variables and rainfall data (such as rainfall duration and intensity). Nevertheless, this research provides recommendations to municipal administration decision‐makers regarding strategic management in the prioritization of flood‐hazard zones.
- Open Access
- 23 September 2025
- Published by Elsevier in International Journal of Applied Earth Observation and Geoinformation
- Vol. 144
Enhancing land use and land cover classification with deep learning-based satellite imagery segmentation
- 23 September 2025
- Published by Elsevier in International Journal of Applied Earth Observation and Geoinformation
- Vol. 144
Open AccessEnhancing land use and land cover classification with deep learning-based satellite imagery segmentation
- Open Access
- 29 August 2025
- Published by Elsevier in Agricultural Water Management
- Vol. 319
Evaluating irrigation suitability for wheat in East Shewa, Ethiopia using remote sensing and AHP
- 29 August 2025
- Published by Elsevier in Agricultural Water Management
- Vol. 319
Open AccessEvaluating irrigation suitability for wheat in East Shewa, Ethiopia using remote sensing and AHP
- 6 July 2025
- Published by Elsevier in Advances In Space Research
- Vol. 76 (8) , 4001-4015
Geospatial analysis and Revised Universal Soil Loss Equation (RUSLE) for predicting soil erosion: a case study in Deder District, Oromia, Ethiopia
- 6 July 2025
- Published by Elsevier in Advances In Space Research
- Vol. 76 (8) , 4001-4015
Geospatial analysis and Revised Universal Soil Loss Equation (RUSLE) for predicting soil erosion: a case study in Deder District, Oromia, Ethiopia
- Open Access
- 2 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-25
Seasonal groundwater quality analysis in a drought prone agricultural region using GIS and IWQI for nitrate contamination insights
The Hennaya Plain’s groundwater resources, which are crucial for agricultural irrigation, face significant quality challenges driven by human-induced environmental transformations. This study comprehensively evaluated groundwater quality during both dry and wet seasons. The assessment used an integrated approach that synergistically combined the Irrigation Water Quality Index (IWQI) with high-resolution nitrate content analysis as complementary diagnostic tools. This combined methodology enabled a detailed evaluation of seasonal groundwater dynamics. Thirty-nine groundwater samples were collected from wells distributed across the plain. The research systematically evaluated physicochemical parameters and their seasonal variations. The investigation revealed contrasting seasonal patterns in groundwater quality parameters. NO3−–N contamination increased markedly from the dry to wet season, with the percentage of samples exceeding permissible levels rising from 12.82 to 69.23%. Conversely, chloride contamination decreased during the wet season, with exceedances dropping from 23.07 to 10.26% of samples. IWQI analysis indicated a marginal improvement in overall water quality during the wet season, which contrasted with a simultaneous escalation in nitrate pollution from moderate to severe levels. The integrated assessment categorized irrigation water quality as moderate to poor, highlighting the intricate interactions between agricultural practices, seasonal precipitation, and groundwater chemistry. These findings underscore the urgent need for comprehensive groundwater management strategies that address seasonal variability and anthropogenic influences on water resources.
- 2 July 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-25
Open AccessSeasonal groundwater quality analysis in a drought prone agricultural region using GIS and IWQI for nitrate contamination insights
The Hennaya Plain’s groundwater resources, which are crucial for agricultural irrigation, face significant quality challenges driven by human-induced environmental transformations. This study comprehensively evaluated groundwater quality during both dry and wet seasons. The assessment used an integrated approach that synergistically combined the Irrigation Water Quality Index (IWQI) with high-resolution nitrate content analysis as complementary diagnostic tools. This combined methodology enabled a detailed evaluation of seasonal groundwater dynamics. Thirty-nine groundwater samples were collected from wells distributed across the plain. The research systematically evaluated physicochemical parameters and their seasonal variations. The investigation revealed contrasting seasonal patterns in groundwater quality parameters. NO3−–N contamination increased markedly from the dry to wet season, with the percentage of samples exceeding permissible levels rising from 12.82 to 69.23%. Conversely, chloride contamination decreased during the wet season, with exceedances dropping from 23.07 to 10.26% of samples. IWQI analysis indicated a marginal improvement in overall water quality during the wet season, which contrasted with a simultaneous escalation in nitrate pollution from moderate to severe levels. The integrated assessment categorized irrigation water quality as moderate to poor, highlighting the intricate interactions between agricultural practices, seasonal precipitation, and groundwater chemistry. These findings underscore the urgent need for comprehensive groundwater management strategies that address seasonal variability and anthropogenic influences on water resources.