Pub Date : 2025-01-31DOI: 10.1016/j.pce.2025.103880
Omid Hazbeh , Hamzeh Ghorbani , Somayeh Tabasi , Meysam Rajabi , Pezhman Soltani Tehrani , Sahar Lajmorak , Mehdi Ahmadi Alvar , Ahmed E. Radwan
The production of gas condensate from condensate gas reservoirs (GCR) presents significant challenges in reservoir engineering management, production, and operations. A crucial factor affecting the production and transport of condensate gas is the condensate liquid viscosity (μlc), which is vital for equations of state and for establishing relationships between PVT properties. This study aims to predict viscosity using five input variables: condensate gravity (API), initial gas-to-condensate ratio (RS), pressure (P), gas specific gravity (γg), and temperature (T). To achieve this, four robust models, previously unused in this domain, were used. Data from 2160 records were gathered from Iranian reservoirs, with 2114 data sets retained after outlier elimination using k-means clustering. The study combines multilayer perceptron (MLP) and distance-weighted k-nearest neighbor (DWKNN) networks with two optimizers, independent component analysis (ICA) and the gravitational search algorithm (GSA), to predict μlc. The results indicate that the AI-based hybrid model achieves significantly greater accuracy than the four empirical equations evaluated, with the DWKNN-GSA model outperforming the others in terms of accuracy (R2 = 0.9998, RMSE = 0.0037 cP). Correlation analysis reveals that P, API, and RS highly influence μlc, whereas T and γg have a low impact. A heat map diagram further highlights that γg exerts the highest effect, while API has the lowest impact on μlc. The approach used in this study demonstrates significantly higher accuracy in predicting μlc compared to other published methods and could be applied to the prediction of similar parameters.
{"title":"Robust computational approach to determine condensate liquid viscosity","authors":"Omid Hazbeh , Hamzeh Ghorbani , Somayeh Tabasi , Meysam Rajabi , Pezhman Soltani Tehrani , Sahar Lajmorak , Mehdi Ahmadi Alvar , Ahmed E. Radwan","doi":"10.1016/j.pce.2025.103880","DOIUrl":"10.1016/j.pce.2025.103880","url":null,"abstract":"<div><div>The production of gas condensate from condensate gas reservoirs (GCR) presents significant challenges in reservoir engineering management, production, and operations. A crucial factor affecting the production and transport of condensate gas is the condensate liquid viscosity (μlc), which is vital for equations of state and for establishing relationships between PVT properties. This study aims to predict viscosity using five input variables: condensate gravity (API), initial gas-to-condensate ratio (RS), pressure (P), gas specific gravity (γg), and temperature (T). To achieve this, four robust models, previously unused in this domain, were used. Data from 2160 records were gathered from Iranian reservoirs, with 2114 data sets retained after outlier elimination using k-means clustering. The study combines multilayer perceptron (MLP) and distance-weighted k-nearest neighbor (DWKNN) networks with two optimizers, independent component analysis (ICA) and the gravitational search algorithm (GSA), to predict μlc. The results indicate that the AI-based hybrid model achieves significantly greater accuracy than the four empirical equations evaluated, with the DWKNN-GSA model outperforming the others in terms of accuracy (R<sup>2</sup> = 0.9998, RMSE = 0.0037 cP). Correlation analysis reveals that P, API, and RS highly influence μlc, whereas T and γg have a low impact. A heat map diagram further highlights that γg exerts the highest effect, while API has the lowest impact on μlc. The approach used in this study demonstrates significantly higher accuracy in predicting μlc compared to other published methods and could be applied to the prediction of similar parameters.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103880"},"PeriodicalIF":3.0,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-30DOI: 10.1016/j.pce.2025.103881
Samuel Chukwujindu Nwokolo , Anthony Umunnakwe Obiwulu , Paul C. Okonkwo , Rita Orji , Theyab R. Alsenani , Ibrahim B. Mansir , Chukwuka Orji
<div><div>This study explores the hybridization of machine learning models with analytical techniques to evaluate the impact of climate change on active power predictions for single-axis (PAS) and dual-axis (PAD) tracking solar-concentrated photovoltaic (CPV) systems in Alice Springs, Australia. It integrates meteorological factors, including global solar irradiation, extraterrestrial solar radiation, clearness index, ambient temperature, and relative humidity, to assess their influence on system performance. Descriptive statistics for PAS and PAD reveal slight variations in these factors, with PAD experiencing higher values for solar irradiation and ambient temperature. To enhance forecasting accuracy, machine learning models, including SARIMA, CARIMA, MLP, RBF, boosting (BOT), and bagging (BAG), are utilized to predict future performance of CPV systems in Alice Springs. These models consider historical weather data and system performance metrics to make accurate predictions for optimal system operation and maintenance. A novel hybrid model, CARIMA-SARIMA-LG, integrates SARIMA, CARIMA, and the logistic distribution (LG) model, showing exceptional performance. The CARIMA-SARIMA-LG model achieves R<sup>2</sup> values of 0.9833% for PAS and 0.9777% for PAD during training, surpassing other models in error metrics such as MAPE, RMSE, and nRMSE. In contrast, traditional machine learning models like MLP and RBF exhibit diminished predictive capabilities. Ensemble methods, while useful, do not achieve the accuracy of statistical models. Furthermore, the study evaluates the efficacy of the CARIMA-SARIMA-LG hybrid model across five African locations (Kano, Accra, Johannesburg, Nairobi, and Dar es Salaam), demonstrating PAD's adaptability to diverse climates. The findings illustrate PAD's outstanding accuracy and adaptability across a variety of climatic situations, providing breakthrough insights for renewable energy generation and grid integration. The influence of changing climate conditions on solar photovoltaic systems in Alice Springs, Australia, across SSP126, SSP245, and SSP585 scenarios demonstrates notable differences in active power generation for both single-axis tracking (PAS) and dual-axis tracking (PAD) systems. Under SSP126, the projections indicate slight variations in power, peaking in DJF with an increase of +3.145% from 2015 to 2030, followed by a decrease of −3.775% in the late century from 2071 to 2099. PAD systems show comparable seasonal patterns, experiencing increases in the initial phases (+5.463%, DJF, 2015–2030) followed by declines as the century progresses (−0.578%, ANN, 2071–2099). Under the SSP245 scenario, moderate losses are projected, with PAS facing a decrease of −2.138% (annual average, 2031–2050) and PAD encountering significant declines by the period of 2071–2099, amounting to −5.001% (annual average). SSP585 indicates the most drastic declines, with PAS facing a reduction of −3.294% (ANN, 2015–2030) and PAD seeing
{"title":"Machine learning and physics-based model hybridization to assess the impact of climate change on single- and dual-axis tracking solar-concentrated photovoltaic systems","authors":"Samuel Chukwujindu Nwokolo , Anthony Umunnakwe Obiwulu , Paul C. Okonkwo , Rita Orji , Theyab R. Alsenani , Ibrahim B. Mansir , Chukwuka Orji","doi":"10.1016/j.pce.2025.103881","DOIUrl":"10.1016/j.pce.2025.103881","url":null,"abstract":"<div><div>This study explores the hybridization of machine learning models with analytical techniques to evaluate the impact of climate change on active power predictions for single-axis (PAS) and dual-axis (PAD) tracking solar-concentrated photovoltaic (CPV) systems in Alice Springs, Australia. It integrates meteorological factors, including global solar irradiation, extraterrestrial solar radiation, clearness index, ambient temperature, and relative humidity, to assess their influence on system performance. Descriptive statistics for PAS and PAD reveal slight variations in these factors, with PAD experiencing higher values for solar irradiation and ambient temperature. To enhance forecasting accuracy, machine learning models, including SARIMA, CARIMA, MLP, RBF, boosting (BOT), and bagging (BAG), are utilized to predict future performance of CPV systems in Alice Springs. These models consider historical weather data and system performance metrics to make accurate predictions for optimal system operation and maintenance. A novel hybrid model, CARIMA-SARIMA-LG, integrates SARIMA, CARIMA, and the logistic distribution (LG) model, showing exceptional performance. The CARIMA-SARIMA-LG model achieves R<sup>2</sup> values of 0.9833% for PAS and 0.9777% for PAD during training, surpassing other models in error metrics such as MAPE, RMSE, and nRMSE. In contrast, traditional machine learning models like MLP and RBF exhibit diminished predictive capabilities. Ensemble methods, while useful, do not achieve the accuracy of statistical models. Furthermore, the study evaluates the efficacy of the CARIMA-SARIMA-LG hybrid model across five African locations (Kano, Accra, Johannesburg, Nairobi, and Dar es Salaam), demonstrating PAD's adaptability to diverse climates. The findings illustrate PAD's outstanding accuracy and adaptability across a variety of climatic situations, providing breakthrough insights for renewable energy generation and grid integration. The influence of changing climate conditions on solar photovoltaic systems in Alice Springs, Australia, across SSP126, SSP245, and SSP585 scenarios demonstrates notable differences in active power generation for both single-axis tracking (PAS) and dual-axis tracking (PAD) systems. Under SSP126, the projections indicate slight variations in power, peaking in DJF with an increase of +3.145% from 2015 to 2030, followed by a decrease of −3.775% in the late century from 2071 to 2099. PAD systems show comparable seasonal patterns, experiencing increases in the initial phases (+5.463%, DJF, 2015–2030) followed by declines as the century progresses (−0.578%, ANN, 2071–2099). Under the SSP245 scenario, moderate losses are projected, with PAS facing a decrease of −2.138% (annual average, 2031–2050) and PAD encountering significant declines by the period of 2071–2099, amounting to −5.001% (annual average). SSP585 indicates the most drastic declines, with PAS facing a reduction of −3.294% (ANN, 2015–2030) and PAD seeing ","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103881"},"PeriodicalIF":3.0,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143168059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-30DOI: 10.1016/j.pce.2025.103879
Ali Danandeh Mehr , Masood Jabarnejad , Mir Jafar Sadegh Safari
Drought is an environmental challenge, with devastating impacts across a wide range of sectors, including agriculture, economy, and ecosystems. Accurate drought forecasting models are necessary for sustainable water resources planning. Therefore, exploring the efficacy and parsimony of emerging machine learning (ML) techniques to enhance predictive drought forecasting models’ accuracy while reducing their complexity is essential. This article introduces a novel hybrid evolutionary ML model, called MOGGP, and compares its efficiency with two evolutionary models, namely gene expression programming and multigene genetic programming as well as conventional Multilayer Perceptron. The new model integrates multi-objective geometric mean optimizer with a traditional symbolic genetic programming that allows parsimonious model selection through developing Pareto optimal solutions. Grided Standardized Precipitation Evapotranspiration Index (SPEI) datasets were employed for demonstrating MOGGP and verifying its efficiency. The results showed that annual cycle is not an effective input for the evolved evolutionary SPEI model. In addition, performance appraisal analysis revealed that the MOGGP consistently exhibits parsimonious models, superior to its counterparts, and excels in addressing multi-objective hydrological modeling problems.
{"title":"MOGGP: A novel multi objective geometric genetic programming model for drought forecasting","authors":"Ali Danandeh Mehr , Masood Jabarnejad , Mir Jafar Sadegh Safari","doi":"10.1016/j.pce.2025.103879","DOIUrl":"10.1016/j.pce.2025.103879","url":null,"abstract":"<div><div>Drought is an environmental challenge, with devastating impacts across a wide range of sectors, including agriculture, economy, and ecosystems. Accurate drought forecasting models are necessary for sustainable water resources planning. Therefore, exploring the efficacy and parsimony of emerging machine learning (ML) techniques to enhance predictive drought forecasting models’ accuracy while reducing their complexity is essential. This article introduces a novel hybrid evolutionary ML model, called MOGGP, and compares its efficiency with two evolutionary models, namely gene expression programming and multigene genetic programming as well as conventional Multilayer Perceptron. The new model integrates multi-objective geometric mean optimizer with a traditional symbolic genetic programming that allows parsimonious model selection through developing Pareto optimal solutions. Grided Standardized Precipitation Evapotranspiration Index (SPEI) datasets were employed for demonstrating MOGGP and verifying its efficiency. The results showed that annual cycle is not an effective input for the evolved evolutionary SPEI model. In addition, performance appraisal analysis revealed that the MOGGP consistently exhibits parsimonious models, superior to its counterparts, and excels in addressing multi-objective hydrological modeling problems.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103879"},"PeriodicalIF":3.0,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143168060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, similar materials are used to prepare the multi-layered composited coal mass. Through drop hammer impact multi-layered combined coal mass test, it's found that each layer appears multiple radial cracks, local coal layer appears circumferential cracks to generate petal-shaped fragments. The number of radial and circumferential cracks and petal-like fragments first increases and then decreases with layers increase. Combining the dynamic strain response characteristics, the first layer radial plane produces compression wave in radial direction and tension wave in circumferential direction due to the loading wave. And then, the unloading wave leads to the tension wave in the radial direction and compression wave in the circumferential direction. The impact load propagates as spherical wave within the coal mass, resulting in compressive strains along the axial direction and tensile strains of the coal mass lateral face. Combined with the failure characteristics and dynamic strain response characteristics, it's clear that the radial cracks first appear in each layer, the process of unloading wave meets with plastic loading wave results in tensile stress wave in some layers to form circumferential cracks. The energy attenuation of stress waves affects the number of circumferential cracks and petal-like fragments. In the axial direction, the coal mass forms dense nuclei and undergoes splitting. The radial and circumferential cracks in the radial plane intersect with the boundary to form axial cracks, finally forming the lateral face axial failure mode with less axial cracks in the top and bottom layers, more axial cracks in the middle layers.
{"title":"Research on dynamic damage path of multi-layered combined coal mass under impact load","authors":"Feng Li, Chenchen Wang, Bo Xu, Dongdong Liang, Zeyu Li, Tianyi Zhang, Bincan Tian","doi":"10.1016/j.pce.2025.103872","DOIUrl":"10.1016/j.pce.2025.103872","url":null,"abstract":"<div><div>In this paper, similar materials are used to prepare the multi-layered composited coal mass. Through drop hammer impact multi-layered combined coal mass test, it's found that each layer appears multiple radial cracks, local coal layer appears circumferential cracks to generate petal-shaped fragments. The number of radial and circumferential cracks and petal-like fragments first increases and then decreases with layers increase. Combining the dynamic strain response characteristics, the first layer radial plane produces compression wave in radial direction and tension wave in circumferential direction due to the loading wave. And then, the unloading wave leads to the tension wave in the radial direction and compression wave in the circumferential direction. The impact load propagates as spherical wave within the coal mass, resulting in compressive strains along the axial direction and tensile strains of the coal mass lateral face. Combined with the failure characteristics and dynamic strain response characteristics, it's clear that the radial cracks first appear in each layer, the process of unloading wave meets with plastic loading wave results in tensile stress wave in some layers to form circumferential cracks. The energy attenuation of stress waves affects the number of circumferential cracks and petal-like fragments. In the axial direction, the coal mass forms dense nuclei and undergoes splitting. The radial and circumferential cracks in the radial plane intersect with the boundary to form axial cracks, finally forming the lateral face axial failure mode with less axial cracks in the top and bottom layers, more axial cracks in the middle layers.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103872"},"PeriodicalIF":3.0,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Carbonate rocks hold significant potential for economic activities, including mineral and hydrocarbon exploration, highlighting the importance of detailed characterization. This study focuses on delineating and characterizing deep-seated Upper Jurassic carbonate deposits (Abenaki Formation) in the Penobscot Field using model-based inversion (MBI) and machine learning (ML) algorithms. The MBI model achieved 94.3% correlation with a relatively lower error of 605.8 m/s∗g/cm3 in impedance estimation. The ML models were evaluated using metrics such as precision, recall, F1 score, accuracy, and misclassification rates. The XGB model consistently outperformed the RF, ANN, and SVM models, achieving the highest precision, recall, F1 score, and accuracy across shale, sand, and carbonate facies classifications. It recorded an overall accuracy of 0.927 and a misclass rate of 0.073, surpassing SVM (accuracy: 0.901, misclass: 0.099), RF (accuracy: 0.866 & misclass: 0.134), and ANN (accuracy: 0.838 & misclass: 0.162). Furthermore, the effective porosity volume was predicted with a correlation of 85.75% and a mean absolute error of 0.02. It was found that the Artimon Member (∼85 m) includes an upper porous carbonate reservoir unit (∼35 m) with impedance 11,500–15,000 m/s∗g/cm3,carbonate probability 70–80% and porosity 12–15%,and a deeper siliciclastic unit (∼45m) with impedance 9500–13,000 m/s∗g/cm3, carbonate probability 20–30% and porosity nearly 3–4% possibly during a significant transgressive phase of sea-level rise. The underlying Baccaro Member (∼260m) predominantly comprises thick carbonate facies with impedance 12,500–16,000 m/s∗g/cm3, carbonate probability 80–90% and porosity nearly 6–9%. This quantitative study examines how depositional environments, mineralization, and diagenesis shape the distribution of carbonate facies in the Penobscot Field. By integrating advanced seismic inversion with machine learning, it refines the characterization of deep-seated carbonate facies, offering insights for identifying potential carbonate hydrocarbon bearing zones worldwide.
{"title":"Machine learning assisted geophysical characterization of deep-seated upper jurassic carbonate deposits in Penobscot Field, Nova Scotia","authors":"Vijay Kumar , Satya Narayan , S.D. Sahoo , Brijesh Kumar , S.K. Pal","doi":"10.1016/j.pce.2025.103876","DOIUrl":"10.1016/j.pce.2025.103876","url":null,"abstract":"<div><div>Carbonate rocks hold significant potential for economic activities, including mineral and hydrocarbon exploration, highlighting the importance of detailed characterization. This study focuses on delineating and characterizing deep-seated Upper Jurassic carbonate deposits (Abenaki Formation) in the Penobscot Field using model-based inversion (MBI) and machine learning (ML) algorithms. The MBI model achieved 94.3% correlation with a relatively lower error of 605.8 m/s∗g/cm<sup>3</sup> in impedance estimation. The ML models were evaluated using metrics such as precision, recall, F1 score, accuracy, and misclassification rates. The XGB model consistently outperformed the RF, ANN, and SVM models, achieving the highest precision, recall, F1 score, and accuracy across shale, sand, and carbonate facies classifications. It recorded an overall accuracy of 0.927 and a misclass rate of 0.073, surpassing SVM (accuracy: 0.901, misclass: 0.099), RF (accuracy: 0.866 & misclass: 0.134), and ANN (accuracy: 0.838 & misclass: 0.162). Furthermore, the effective porosity volume was predicted with a correlation of 85.75% and a mean absolute error of 0.02. It was found that the Artimon Member (∼85 m) includes an upper porous carbonate reservoir unit (∼35 m) with impedance 11,500–15,000 <em>m/s∗</em><em>g/cm</em><sup><em>3</em></sup><em>,</em> <em>carbonate probability 70–80% and porosity 12–15%</em><em>,</em> <em>and a deeper siliciclastic unit (∼45</em> <em>m) with</em> impedance <em>9500–13</em>,000 m/s∗g/cm<sup>3</sup>, carbon<em>ate probability</em> 20–30% <em>and porosity</em> nearly 3–4% possibly during a significant transgressive phase of sea-level rise. <em>The underlying Baccaro Member (∼260</em> <em>m) predominantly comprises thick carbonate facies with impedance 12,500–16,</em>000 m/s∗g/cm<sup>3</sup>, carbonate probability 80–90% and porosity nearly 6–9%. This quantitative study examines how depositional environments, mineralization, and diagenesis shape the distribution of carbonate facies in the Penobscot Field. By integrating advanced seismic inversion with machine learning, it refines the characterization of deep-seated carbonate facies, offering insights for identifying potential carbonate hydrocarbon bearing zones worldwide.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103876"},"PeriodicalIF":3.0,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-25DOI: 10.1016/j.pce.2025.103877
Musah Saeed Zango , Sidique Gawusu , Mahamuda Abu
This study applied machine learning techniques to monitor and predict fluoride concentrations in the Vea Catchment area, a region affected by endemic fluoride contamination. The aim was to identify high-risk areas with unsafe fluoride levels, supporting public health surveillance and water management efforts. Water quality data, including fluoride concentrations and various physico-chemical parameters, were collected from multiple sampling sites. Machine learning models, including Logistic Regression, Random Forest, and Gradient Boosting, were used for regression, classification, and spatial analysis to classify regions as safe or unsafe based on World Health Organization (WHO) fluoride guidelines. The results demonstrated the effectiveness of the models, with Random Forest and Gradient Boosting achieving high accuracy in predicting unsafe fluoride levels with precision and F1-score values of 0.88 for both models, successfully identifying high-risk areas with concentrations exceeding the WHO limit of 1.5 mg/L. Anomaly detection techniques revealed localized areas of concern. This study highlights the value of machine learning in water quality management, providing a data-driven approach to predicting fluoride contamination and informing public health interventions and water management strategies. The integration of predictive modeling with spatial analysis represents a significant advancement, offering the potential for real-time water quality monitoring in fluoride-endemic regions.
{"title":"Fluoride surveillance in a fluoride endemic region using machine learning techniques: A case study of Vea Catchment, Upper East Region, Ghana","authors":"Musah Saeed Zango , Sidique Gawusu , Mahamuda Abu","doi":"10.1016/j.pce.2025.103877","DOIUrl":"10.1016/j.pce.2025.103877","url":null,"abstract":"<div><div>This study applied machine learning techniques to monitor and predict fluoride concentrations in the Vea Catchment area, a region affected by endemic fluoride contamination. The aim was to identify high-risk areas with unsafe fluoride levels, supporting public health surveillance and water management efforts. Water quality data, including fluoride concentrations and various physico-chemical parameters, were collected from multiple sampling sites. Machine learning models, including Logistic Regression, Random Forest, and Gradient Boosting, were used for regression, classification, and spatial analysis to classify regions as safe or unsafe based on World Health Organization (WHO) fluoride guidelines. The results demonstrated the effectiveness of the models, with Random Forest and Gradient Boosting achieving high accuracy in predicting unsafe fluoride levels with precision and F1-score values of 0.88 for both models, successfully identifying high-risk areas with concentrations exceeding the WHO limit of 1.5 mg/L. Anomaly detection techniques revealed localized areas of concern. This study highlights the value of machine learning in water quality management, providing a data-driven approach to predicting fluoride contamination and informing public health interventions and water management strategies. The integration of predictive modeling with spatial analysis represents a significant advancement, offering the potential for real-time water quality monitoring in fluoride-endemic regions.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103877"},"PeriodicalIF":3.0,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-24DOI: 10.1016/j.pce.2025.103875
S. Farzin, M. Valikhan Anaraki, M. Kadkhodazadeh, A. Morshed-Bozorgdel
In the present study, for the first time, a novel methodology has been introduced for constructing missed runoff data in Iran country. To this end, the convolution neural network (CNN) is developed based on various types of data, including basin characteristics, time data, the geography of each station, the statistical characteristics of river flow that do not have missing data, and the statistical characteristics of river flow. Furthermore, quantile mapping is used to correct bias in CNN results. Seven CNN structures were defined, and the results were compared with deep neural networks and machine learning algorithms. The obtained results of runoff modeling in the 1666 hydrometric station indicated the superiority of the best CNN structure (CNN4) with mean absolute error = 5.95m3/s, root mean square error = 25.61m3/s, relative root mean square error = 0.44, and Nash Sutcliffe efficiency coefficient = 0.81. In addition, the distribution of runoff modeled with CNN4 was more similar to observed runoff than those for other algorithms. Finally, the runoff time series for all stations was constructed, even for stations with 100% missing data. This study's methodology can potentially estimate missing data in runoff river data from other countries.
{"title":"Novel methodology for prediction of missing values in river flow based on convolution neural networks: Principles and application in Iran country","authors":"S. Farzin, M. Valikhan Anaraki, M. Kadkhodazadeh, A. Morshed-Bozorgdel","doi":"10.1016/j.pce.2025.103875","DOIUrl":"10.1016/j.pce.2025.103875","url":null,"abstract":"<div><div>In the present study, for the first time, a novel methodology has been introduced for constructing missed runoff data in Iran country. To this end, the convolution neural network (CNN) is developed based on various types of data, including basin characteristics, time data, the geography of each station, the statistical characteristics of river flow that do not have missing data, and the statistical characteristics of river flow. Furthermore, quantile mapping is used to correct bias in CNN results. Seven CNN structures were defined, and the results were compared with deep neural networks and machine learning algorithms. The obtained results of runoff modeling in the 1666 hydrometric station indicated the superiority of the best CNN structure (CNN4) with mean absolute error = 5.95m3/s, root mean square error = 25.61m3/s, relative root mean square error = 0.44, and Nash Sutcliffe efficiency coefficient = 0.81. In addition, the distribution of runoff modeled with CNN4 was more similar to observed runoff than those for other algorithms. Finally, the runoff time series for all stations was constructed, even for stations with 100% missing data. This study's methodology can potentially estimate missing data in runoff river data from other countries.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103875"},"PeriodicalIF":3.0,"publicationDate":"2025-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-16DOI: 10.1016/j.pce.2025.103873
Yanqing Ding , Chengliang Wang , Puyu Qi , Mengyang Sun , Zhiru Hu , Zhimin Xu
Acid mine drainage (AMD) releases heavy metals, deteriorating the regional environment. Copper (Cu) is often found in AMD, and the migration and transformation are controlled by colloidal organic matter (COM). In this study, COM in the AMD-affected river was extracted by ultrafiltration, compositions of COM were detected by ultraviolet–visible and fluorescence spectra, and the binding behavior of COM with Cu was revealed by two-dimensional correlation spectroscopy analysis of synchronous fluorescence. The results found that COM in the AMD-affected river mainly originated from autochthonous sources, including microbial-derived humic-like, fulvic-like, and tryptophan-like fluorescence components. The fluorescence proportion of the first two was relatively higher, accounting for 78.71–83.90% in river and 78.65–91.03% in sediment. AMD decreased riverine COC by 0.24 mg·L−1 and a(350) by 13.83%, but increased the aromaticity and molecular weight of riverine COM downstream, while AMD increased COM contents in river sediment. The redundancy analysis found the explanatory power of Cu was 44.4%, and thus Cu was the primary environmental driving factor in the AMD-affected river. Fluorescence quenching titration found that microbial-derived humic-like substances were the most sensitive to Cu(II) titration. Cu(II) was preferentially bound to fulvic-like substances of riverine COM and with tryptophan-like substances of sediment COM. AMD delayed the binding sequences of tryptophan-like substances in riverine COM at the inlet and fulvic-like substances in downstream sediment COM with Cu(II), and weakened the binding affinity of COM with Cu(Ⅱ) at 2.03%–6.61% in downstream sediment to increase the risks of heavy metal pollution.
{"title":"Effect of acid mine drainage on colloidal organic matter in the river: Changes in composition and copper binding behavior","authors":"Yanqing Ding , Chengliang Wang , Puyu Qi , Mengyang Sun , Zhiru Hu , Zhimin Xu","doi":"10.1016/j.pce.2025.103873","DOIUrl":"10.1016/j.pce.2025.103873","url":null,"abstract":"<div><div>Acid mine drainage (AMD) releases heavy metals, deteriorating the regional environment. Copper (Cu) is often found in AMD, and the migration and transformation are controlled by colloidal organic matter (COM). In this study, COM in the AMD-affected river was extracted by ultrafiltration, compositions of COM were detected by ultraviolet–visible and fluorescence spectra, and the binding behavior of COM with Cu was revealed by two-dimensional correlation spectroscopy analysis of synchronous fluorescence. The results found that COM in the AMD-affected river mainly originated from autochthonous sources, including microbial-derived humic-like, fulvic-like, and tryptophan-like fluorescence components. The fluorescence proportion of the first two was relatively higher, accounting for 78.71–83.90% in river and 78.65–91.03% in sediment. AMD decreased riverine COC by 0.24 mg·L<sup>−1</sup> and <em>a</em>(350) by 13.83%, but increased the aromaticity and molecular weight of riverine COM downstream, while AMD increased COM contents in river sediment. The redundancy analysis found the explanatory power of Cu was 44.4%, and thus Cu was the primary environmental driving factor in the AMD-affected river. Fluorescence quenching titration found that microbial-derived humic-like substances were the most sensitive to Cu(II) titration. Cu(II) was preferentially bound to fulvic-like substances of riverine COM and with tryptophan-like substances of sediment COM. AMD delayed the binding sequences of tryptophan-like substances in riverine COM at the inlet and fulvic-like substances in downstream sediment COM with Cu(II), and weakened the binding affinity of COM with Cu(Ⅱ) at 2.03%–6.61% in downstream sediment to increase the risks of heavy metal pollution.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103873"},"PeriodicalIF":3.0,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-16DOI: 10.1016/j.pce.2025.103874
Donald T.A. Tapfuma , Desmond Mwembe , Yogeshkumar Naik
The need for refined non-lethal techniques for the monitoring of the bioavailability and accumulation of toxic metals in aquatic ecosystems in Artisanal small-scale gold mining sites has motivated the current study. The respective fin clips and white muscle of 13 fish species from selected ASGM hotspots in the Upper uMzingwane catchment area were collected and analysed using an Inductively Coupled Plasma Mass Spectrophotometer for xenobiotics (As, Cd, Hg, Pb) and trace metals (Cr, Cu, Li, Ni) often associated with this highly unregulated, illicit trade. The oxidation of sulphide ore bodies ingrained in greenstone-hosted quartz-carbonate veins endemic in gold panning hotspots in the study area has been known to result in the leaching of xenobiotics into water bodies with their subsequent accumulation in aquatic biota. A comparison of the levels of the xenobiotics in the respective fin clips and white muscle of the fish species under study, together with an assessment of the levels in their respective aquatic environments may give a clue as to the bioavailability and mobility of the metals. Notably, As, Cd and Cr concentration in sediment samples ranged from 533.88 to 16185.03 ppb; 35.5–36.4 ppb; and 37173.36–273651.69 ppb respectively. Rock cutlets; Chiloglanis paratus and Chiloglanis pretoriae were the greatest metal accumulators. Of note was the mercury concentrations in Chiloglanis pretoriae and Chiloglanis paratus fin clips ranged from, 0.17–0.35 mg/kg and 0.23–13.35 mg/kg respectively. Chromium concentration in Chiloglanis pretoriae and Chiloglanis paratus in fin clips ranged from 43.15 to 57.87 mg/kg and 44.87–244.86 mg/kg respectively. The study concludes that fin clips can, indeed, be used as a non-lethal means to assess the bioavailability of metal species in an aquatic ecosystem.
{"title":"Non-lethal method for the assessment of bioavailable metals in aquatic ecosystems surrounding ASGM activity","authors":"Donald T.A. Tapfuma , Desmond Mwembe , Yogeshkumar Naik","doi":"10.1016/j.pce.2025.103874","DOIUrl":"10.1016/j.pce.2025.103874","url":null,"abstract":"<div><div>The need for refined non-lethal techniques for the monitoring of the bioavailability and accumulation of toxic metals in aquatic ecosystems in Artisanal small-scale gold mining sites has motivated the current study. The respective fin clips and white muscle of 13 fish species from selected ASGM hotspots in the Upper uMzingwane catchment area were collected and analysed using an Inductively Coupled Plasma Mass Spectrophotometer for xenobiotics (As, Cd, Hg, Pb) and trace metals (Cr, Cu, Li, Ni) often associated with this highly unregulated, illicit trade. The oxidation of sulphide ore bodies ingrained in greenstone-hosted quartz-carbonate veins endemic in gold panning hotspots in the study area has been known to result in the leaching of xenobiotics into water bodies with their subsequent accumulation in aquatic biota. A comparison of the levels of the xenobiotics in the respective fin clips and white muscle of the fish species under study, together with an assessment of the levels in their respective aquatic environments may give a clue as to the bioavailability and mobility of the metals. Notably, As, Cd and Cr concentration in sediment samples ranged from 533.88 to 16185.03 ppb; 35.5–36.4 ppb; and 37173.36–273651.69 ppb respectively. Rock cutlets; <em>Chiloglanis paratus</em> and <em>Chiloglanis pretoriae</em> were the greatest metal accumulators. Of note was the mercury concentrations in <em>Chiloglanis pretoriae</em> and <em>Chiloglanis paratus</em> fin clips ranged from, 0.17–0.35 mg/kg and 0.23–13.35 mg/kg respectively. Chromium concentration in <em>Chiloglanis pretoriae</em> and <em>Chiloglanis paratus</em> in fin clips ranged from 43.15 to 57.87 mg/kg and 44.87–244.86 mg/kg respectively. The study concludes that fin clips can, indeed, be used as a non-lethal means to assess the bioavailability of metal species in an aquatic ecosystem.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103874"},"PeriodicalIF":3.0,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-11DOI: 10.1016/j.pce.2025.103871
Isaac Nyambiya , Lazarus Chapungu , Lawrence Sawunyama , Eustina V. Musvoto , Luxon Nhamo , John N. Zvimba
The Circular Economy (CE) emerges as an alternative to the long-standing linear economy. While beneficial, the linear model has fallen short in addressing current cross-sectoral challenges. This study identifies the drivers, opportunities, and barriers for transitioning to a circular economy within the wastewater services sector in low-to medium-income countries (LMICs). Utilizing case studies from LMICs, the study provides pathways to transition from the current linear model to the circular economy within the wastewater services sector in LMICs. Key barriers identified include financial constraints, uncertainty among stakeholders, weak governance structures, and limited public awareness leading to health concerns and marketability issues. Drivers for circularity include recognizing wastewater as a valuable resource for revenue generation, leveraging smart technologies, and ensuring sustainability compliance. Opportunities include improving sector governance, enhancing water security, and promoting human and environmental health through demand management, resource diversification, and nutrient recovery. The study advocates for proactive management, effective partnerships, and sustainable business practices to accelerate the transition to a circular economy, emphasizing the need for prioritized resources in awareness, training, and innovative infrastructure development.
{"title":"Circular economy drivers, opportunities, and barriers, for wastewater services within low- and medium-income countries","authors":"Isaac Nyambiya , Lazarus Chapungu , Lawrence Sawunyama , Eustina V. Musvoto , Luxon Nhamo , John N. Zvimba","doi":"10.1016/j.pce.2025.103871","DOIUrl":"10.1016/j.pce.2025.103871","url":null,"abstract":"<div><div>The Circular Economy (CE) emerges as an alternative to the long-standing linear economy. While beneficial, the linear model has fallen short in addressing current cross-sectoral challenges. This study identifies the drivers, opportunities, and barriers for transitioning to a circular economy within the wastewater services sector in low-to medium-income countries (LMICs). Utilizing case studies from LMICs, the study provides pathways to transition from the current linear model to the circular economy within the wastewater services sector in LMICs. Key barriers identified include financial constraints, uncertainty among stakeholders, weak governance structures, and limited public awareness leading to health concerns and marketability issues. Drivers for circularity include recognizing wastewater as a valuable resource for revenue generation, leveraging smart technologies, and ensuring sustainability compliance. Opportunities include improving sector governance, enhancing water security, and promoting human and environmental health through demand management, resource diversification, and nutrient recovery. The study advocates for proactive management, effective partnerships, and sustainable business practices to accelerate the transition to a circular economy, emphasizing the need for prioritized resources in awareness, training, and innovative infrastructure development.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"138 ","pages":"Article 103871"},"PeriodicalIF":3.0,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143279798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号