Physics and Chemistry of the Earth最新文献_第4页

Bridging data scarcity in groundwater quality studies: A systematic evaluation of statistical and deep learning-based generators 弥合地下水质量研究中的数据短缺：基于统计和深度学习的生成器的系统评估

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-06-01 Epub Date: 2026-02-05 DOI: 10.1016/j.pce.2026.104327

C.D. Aju , Bhupendra Bahadur Singh , A.L. Achu , Madhusudan Ingale , Mangesh M. Goswami , M.C. Raicy , L. Elango

Large-scale groundwater quality assessments are often hindered by the limited availability of hydrochemical data. Synthetic data generation provides a means to augment small datasets; however, the reliability of these methods and their implications for predictive modeling remain underexplored in environmental studies, particularly in the context of groundwater sustainability. We systematically evaluated six approaches, including bootstrap sampling, Gaussian noise perturbation, Monte Carlo sampling, SMOGN, CTGAN, and TVAE, using a groundwater quality dataset from southern India. Synthetic datasets were evaluated for their similarity to real data using the Kolmogorov–Smirnov test, the Wasserstein distance, moment differences, Pearson correlation, kernel density estimation plots, and principal component analysis. The practical utility of the synthetic data was evaluated by training a Random Forest model to predict total dissolved solids (TDS) from major ions. The model performance on the real dataset was assessed using R², RMSE, and MAE. Bootstrap delivered near-perfect agreement with the real data (R² = 0.999, NSE = 0.999, RMSE = 41.5 mg L⁻¹), with SMOGN being competitive. Gaussian perturbation was acceptable, while TVAE was moderate. Monte Carlo and CTGAN performed poorly, with negative NSE indicating performance worse than predicting the mean. SHAP-based feature importance analysis confirmed that the best-performing synthetic methods preserved the dominant hydrochemical drivers. Overall, traditional resampling approaches (Bootstrap, SMOGN) outperformed complex deep generative models on small-sample groundwater datasets. This methodology can support risk assessments by improving the accuracy of water-quality predictive models, thereby facilitating effective resource management and pollution control. This study provides practical guidance for assessing and managing groundwater quality by recommending synthetic data augmentation strategies tailored to dataset characteristics, particularly in data-limited regions.

大规模地下水质量评价常常因水化学数据有限而受到阻碍。合成数据生成提供了一种增强小数据集的方法；然而，在环境研究中，特别是在地下水可持续性的背景下，这些方法的可靠性及其对预测模型的影响仍未得到充分探讨。我们系统地评估了六种方法，包括自举采样、高斯噪声扰动、蒙特卡罗采样、SMOGN、CTGAN和TVAE，使用了印度南部的地下水质量数据集。使用Kolmogorov-Smirnov检验、Wasserstein距离、矩差、Pearson相关、核密度估计图和主成分分析来评估合成数据集与真实数据的相似性。通过训练随机森林模型来预测主要离子的总溶解固体（TDS），评估了合成数据的实际效用。使用R2、RMSE和MAE评估模型在真实数据集上的性能。Bootstrap与实际数据接近完美吻合（R2 = 0.999, NSE = 0.999, RMSE = 41.5 mg L−1），SMOGN具有竞争力。高斯扰动是可以接受的，而TVAE是中等的。蒙特卡罗和CTGAN表现不佳，负NSE表明性能比预测平均值更差。基于shap的特征重要性分析证实，性能最好的合成方法保留了主要的水化学驱动因素。总体而言，传统的重采样方法（Bootstrap、SMOGN）在小样本地下水数据集上的表现优于复杂的深层生成模型。这种方法可以通过提高水质预测模型的准确性来支持风险评估，从而促进有效的资源管理和污染控制。本研究通过推荐适合数据集特征的综合数据增强策略，特别是在数据有限的地区，为评估和管理地下水质量提供了实用指导。

{"title":"Bridging data scarcity in groundwater quality studies: A systematic evaluation of statistical and deep learning-based generators","authors":"C.D. Aju , Bhupendra Bahadur Singh , A.L. Achu , Madhusudan Ingale , Mangesh M. Goswami , M.C. Raicy , L. Elango","doi":"10.1016/j.pce.2026.104327","DOIUrl":"10.1016/j.pce.2026.104327","url":null,"abstract":"<div><div>Large-scale groundwater quality assessments are often hindered by the limited availability of hydrochemical data. Synthetic data generation provides a means to augment small datasets; however, the reliability of these methods and their implications for predictive modeling remain underexplored in environmental studies, particularly in the context of groundwater sustainability. We systematically evaluated six approaches, including bootstrap sampling, Gaussian noise perturbation, Monte Carlo sampling, SMOGN, CTGAN, and TVAE, using a groundwater quality dataset from southern India. Synthetic datasets were evaluated for their similarity to real data using the Kolmogorov–Smirnov test, the Wasserstein distance, moment differences, Pearson correlation, kernel density estimation plots, and principal component analysis. The practical utility of the synthetic data was evaluated by training a Random Forest model to predict total dissolved solids (TDS) from major ions. The model performance on the real dataset was assessed using R<sup>2</sup>, RMSE, and MAE. Bootstrap delivered near-perfect agreement with the real data (R<sup>2</sup> = 0.999, NSE = 0.999, RMSE = 41.5 mg L<sup>−1</sup>), with SMOGN being competitive. Gaussian perturbation was acceptable, while TVAE was moderate. Monte Carlo and CTGAN performed poorly, with negative NSE indicating performance worse than predicting the mean. SHAP-based feature importance analysis confirmed that the best-performing synthetic methods preserved the dominant hydrochemical drivers. Overall, traditional resampling approaches (Bootstrap, SMOGN) outperformed complex deep generative models on small-sample groundwater datasets. This methodology can support risk assessments by improving the accuracy of water-quality predictive models, thereby facilitating effective resource management and pollution control. This study provides practical guidance for assessing and managing groundwater quality by recommending synthetic data augmentation strategies tailored to dataset characteristics, particularly in data-limited regions.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"143 ","pages":"Article 104327"},"PeriodicalIF":4.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174283","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}

引用次数: 0

Generative AI-enabled forecasting and green supply chain sustainability assessment: Evidence from China's palm oil trade with ASEAN 生成式人工智能预测和绿色供应链可持续性评估：来自中国与东盟棕榈油贸易的证据

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-06-01 Epub Date: 2026-02-06 DOI: 10.1016/j.pce.2026.104341

Yuhan Liu , Xiangzheng Deng , Yunxiao Gao

Food systems and agri-commodity supply chains face increasing pressures from climate change, trade volatility, and environmental degradation, palm oil positioned at the center of sustainability debates. China, as a major importer, and Indonesia, the leading producer, play crucial roles in influencing the environmental footprint and resilience of cross-border palm-oil trade. This study integrates generative AI-based demand forecasting with spatial sustainability assessment to evaluate the alignment between China's future palm-oil demand and Indonesia's sustainability-compliant supply. Using Time-series Generative Adversarial Networks (TimeGAN), we generate scenario-rich forecasts of China's palm-oil imports through 2030. A province-level Green Supply Chain Sustainability Index (GSCI) for Indonesia, incorporating deforestation intensity, land-use efficiency, zero-deforestation commitments, and RSPO certification, supports traceability-based allocation. Results show that more than 60% of China's projected imports can be met by high-GSCI provinces, indicating strong potential for deforestation-free procurement without undermining supply security. The results demonstrate how AI-enabled forecasting combined with spatial sustainability indicators can inform environmentally responsible sourcing strategies and enhance resilience in cross-border palm-oil supply systems.

粮食系统和农产品供应链面临来自气候变化、贸易波动和环境退化的越来越大的压力，棕榈油处于可持续性辩论的中心。作为主要进口国的中国和主要生产国的印度尼西亚，在影响跨境棕榈油贸易的环境足迹和弹性方面发挥着至关重要的作用。本研究将基于生成式人工智能的需求预测与空间可持续性评估相结合，以评估中国未来棕榈油需求与印尼符合可持续性的供应之间的一致性。利用时间序列生成对抗网络（TimeGAN），我们对中国到2030年的棕榈油进口进行了场景丰富的预测。印度尼西亚的省级绿色供应链可持续发展指数（GSCI）综合了毁林强度、土地利用效率、零毁林承诺和RSPO认证，支持基于可追溯性的分配。结果显示，中国60%以上的预计进口量可以由高gsci省份满足，这表明在不破坏供应安全的情况下实现无砍伐采购的巨大潜力。研究结果表明，人工智能预测与空间可持续性指标相结合，可以为环境负责任的采购策略提供信息，并增强跨境棕榈油供应系统的复原力。

{"title":"Generative AI-enabled forecasting and green supply chain sustainability assessment: Evidence from China's palm oil trade with ASEAN","authors":"Yuhan Liu , Xiangzheng Deng , Yunxiao Gao","doi":"10.1016/j.pce.2026.104341","DOIUrl":"10.1016/j.pce.2026.104341","url":null,"abstract":"<div><div>Food systems and agri-commodity supply chains face increasing pressures from climate change, trade volatility, and environmental degradation, palm oil positioned at the center of sustainability debates. China, as a major importer, and Indonesia, the leading producer, play crucial roles in influencing the environmental footprint and resilience of cross-border palm-oil trade. This study integrates generative AI-based demand forecasting with spatial sustainability assessment to evaluate the alignment between China's future palm-oil demand and Indonesia's sustainability-compliant supply. Using Time-series Generative Adversarial Networks (TimeGAN), we generate scenario-rich forecasts of China's palm-oil imports through 2030. A province-level Green Supply Chain Sustainability Index (GSCI) for Indonesia, incorporating deforestation intensity, land-use efficiency, zero-deforestation commitments, and RSPO certification, supports traceability-based allocation. Results show that more than 60% of China's projected imports can be met by high-GSCI provinces, indicating strong potential for deforestation-free procurement without undermining supply security. The results demonstrate how AI-enabled forecasting combined with spatial sustainability indicators can inform environmentally responsible sourcing strategies and enhance resilience in cross-border palm-oil supply systems.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"143 ","pages":"Article 104341"},"PeriodicalIF":4.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174280","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}

引用次数: 0

Mapping seawater intrusion for sustainable groundwater management in arid regions: A geophysical approach from Al-Hofuf, Eastern Saudi Arabia 绘制干旱地区可持续地下水管理的海水入侵图：来自沙特阿拉伯东部Al-Hofuf的地球物理方法

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-06-01 Epub Date: 2026-01-20 DOI: 10.1016/j.pce.2026.104294

Kamal Abdelrahman , Alfahad A. Almakrami , Saad S. Alarifi

Seawater intrusion affects the sustainability of coastal aquifers, particularly in arid regions such as Eastern Saudi Arabia, where groundwater is the primary source of water. This study maps and characterizes seawater intrusion in Al-Hofuf area using Vertical Electrical Sounding (VES) with a Schlumberger array at 20 sites along a coastal–inland transect (AB/2 up to 400 m). Resistivity data were inverted through 1D modeling and statistically validated (R² > 0.97), revealing four geoelectric layers. The upper heterogeneous Quaternary sediments exhibited variable resistivity, while a low-resistivity clay-rich layer (3.4–39 Ω m) acted as an aquitard. The underlying fractured and karstified limestone (12.2–132 Ω m) forms the principal freshwater aquifer. Very low resistivity (<10 Ω m) near the coast (VES-20, VES-16, VES-10) indicated active intrusion, whereas inland points (VES-7, VES-8) showed high resistivity, denoting fresh or dry sands. Spatial analysis revealed an intrusion extending up to 35 km inland, along permeable sands and fractured limestone. Structural geology, particularly north–south trending anticlines and synclines, modulates saline migration, acting as partial flow barriers. Integrating geoelectrical results with structural mapping produced the first high-resolution seawater intrusion model for Al-Hofuf, identifying vulnerable zones and potential sites for Managed Aquifer Recharge (MAR). This cost-effective and scalable approach supports sustainable groundwater management in arid coastal environments, aligning with Sustainable Development Goal 6.

海水入侵影响沿海含水层的可持续性，特别是在干旱地区，如沙特阿拉伯东部，地下水是主要的水源。本研究利用垂直电测深（VES）和斯伦贝谢阵列在沿海-内陆样带的20个地点（AB/2至400米）绘制了Al-Hofuf地区海水入侵的地图和特征。通过一维建模反演电阻率数据并进行统计验证（R2 > 0.97），揭示了4个地电层。上部非均质第四纪沉积物表现为可变电阻率，而低电阻率富粘土层（3.4-39 Ω m）则为引水层。下伏的裂隙岩溶灰岩（12.2-132 Ω m）形成了主要的淡水含水层。靠近海岸（vs -20、vs -16、vs -10）的极低电阻率（<10 Ω m）表明侵入活动，而内陆点（vs -7、vs -8）的电阻率较高，表明为新鲜或干砂。空间分析显示，沿渗透性砂岩和裂缝性石灰岩向内陆延伸了35公里。构造地质，特别是南北走向的背斜和向斜，调节了盐运移，起到部分流动障碍的作用。将地电结果与构造测绘相结合，生成了Al-Hofuf的第一个高分辨率海水入侵模型，确定了脆弱区域和管理含水层补给（MAR）的潜在地点。这种具有成本效益和可扩展的方法支持干旱沿海环境中的可持续地下水管理，符合可持续发展目标6。

{"title":"Mapping seawater intrusion for sustainable groundwater management in arid regions: A geophysical approach from Al-Hofuf, Eastern Saudi Arabia","authors":"Kamal Abdelrahman , Alfahad A. Almakrami , Saad S. Alarifi","doi":"10.1016/j.pce.2026.104294","DOIUrl":"10.1016/j.pce.2026.104294","url":null,"abstract":"<div><div>Seawater intrusion affects the sustainability of coastal aquifers, particularly in arid regions such as Eastern Saudi Arabia, where groundwater is the primary source of water. This study maps and characterizes seawater intrusion in Al-Hofuf area using Vertical Electrical Sounding (VES) with a Schlumberger array at 20 sites along a coastal–inland transect (AB/2 up to 400 <em>m</em>). Resistivity data were inverted through 1D modeling and statistically validated (R<sup>2</sup> > 0.97), revealing four geoelectric layers. The upper heterogeneous Quaternary sediments exhibited variable resistivity, while a low-resistivity clay-rich layer (3.4–39 Ω m) acted as an aquitard. The underlying fractured and karstified limestone (12.2–132 Ω m) forms the principal freshwater aquifer. Very low resistivity (<10 Ω m) near the coast (VES-20, VES-16, VES-10) indicated active intrusion, whereas inland points (VES-7, VES-8) showed high resistivity, denoting fresh or dry sands. Spatial analysis revealed an intrusion extending up to 35 km inland, along permeable sands and fractured limestone. Structural geology, particularly north–south trending anticlines and synclines, modulates saline migration, acting as partial flow barriers. Integrating geoelectrical results with structural mapping produced the first high-resolution seawater intrusion model for Al-Hofuf, identifying vulnerable zones and potential sites for Managed Aquifer Recharge (MAR). This cost-effective and scalable approach supports sustainable groundwater management in arid coastal environments, aligning with Sustainable Development Goal 6.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"143 ","pages":"Article 104294"},"PeriodicalIF":4.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174276","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}

引用次数: 0

Remote sensing applications for assessing climate change impacts on deciduous forests—A systematic review 气候变化对阔叶林影响的遥感评估——系统综述

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-06-01 Epub Date: 2026-01-28 DOI: 10.1016/j.pce.2026.104321

Sude Gül Yel , Hasan Burak Özmen , Siğnem Öney Birol , Esra Tunç Görmüş , Gordana Kaplan

Deciduous forests play a vital role in global carbon sequestration, hydrological regulation, and biodiversity conservation, yet they are increasingly vulnerable to the adverse effects of climate change. This review synthesizes recent advances in remote sensing applications for assessing climate-induced changes in phenology, productivity, and resilience of deciduous forests. A systematic literature analysis was conducted using Web of Science, incorporating 70 peer-reviewed studies that employed satellite-based datasets such as MODIS, Landsat, and Sentinel-2. The review identifies key remote sensing metrics, including NDVI, EVI, LAI, and SIF, which are used to monitor phenological shifts, drought impacts, and changes in forest productivity. Analytical approaches such as time series analysis, regression models, and machine learning (e.g., RF, LMEM) were evaluated for their effectiveness in modeling forest-climate interactions. Results highlight distinct regional patterns in phenological responses, sensor preferences across climate zones, and a growing reliance on multi-sensor data fusion to enhance monitoring precision. Despite advancements in remote sensing technologies, critical research gaps persist, especially regarding the biological interpretation of spectral indices and the assessment of forest resilience mechanisms at finer spatial scales. This review underscores the need for interdisciplinary approaches and advanced remote sensing frameworks to support future forest conservation and climate adaptation strategies.

落叶森林在全球固碳、水文调节和生物多样性保护方面发挥着至关重要的作用，但它们越来越容易受到气候变化的不利影响。本文综述了近年来遥感应用在评估气候引起的落叶森林物候、生产力和恢复力变化方面的最新进展。利用Web of Science进行了系统的文献分析，纳入了70项同行评议的研究，这些研究采用了基于卫星的数据集，如MODIS、Landsat和Sentinel-2。该综述确定了用于监测物候变化、干旱影响和森林生产力变化的关键遥感指标，包括NDVI、EVI、LAI和SIF。对时间序列分析、回归模型和机器学习（如RF、LMEM）等分析方法在模拟森林-气候相互作用方面的有效性进行了评估。结果突出了物候响应的不同区域模式，不同气候带的传感器偏好，以及越来越多地依赖多传感器数据融合来提高监测精度。尽管遥感技术取得了进步，但关键的研究差距仍然存在，特别是在光谱指数的生物学解释和在更精细的空间尺度上评估森林恢复力机制方面。这一综述强调需要跨学科方法和先进的遥感框架来支持未来的森林保护和气候适应战略。

{"title":"Remote sensing applications for assessing climate change impacts on deciduous forests—A systematic review","authors":"Sude Gül Yel , Hasan Burak Özmen , Siğnem Öney Birol , Esra Tunç Görmüş , Gordana Kaplan","doi":"10.1016/j.pce.2026.104321","DOIUrl":"10.1016/j.pce.2026.104321","url":null,"abstract":"<div><div>Deciduous forests play a vital role in global carbon sequestration, hydrological regulation, and biodiversity conservation, yet they are increasingly vulnerable to the adverse effects of climate change. This review synthesizes recent advances in remote sensing applications for assessing climate-induced changes in phenology, productivity, and resilience of deciduous forests. A systematic literature analysis was conducted using Web of Science, incorporating 70 peer-reviewed studies that employed satellite-based datasets such as MODIS, Landsat, and Sentinel-2. The review identifies key remote sensing metrics, including NDVI, EVI, LAI, and SIF, which are used to monitor phenological shifts, drought impacts, and changes in forest productivity. Analytical approaches such as time series analysis, regression models, and machine learning (e.g., RF, LMEM) were evaluated for their effectiveness in modeling forest-climate interactions. Results highlight distinct regional patterns in phenological responses, sensor preferences across climate zones, and a growing reliance on multi-sensor data fusion to enhance monitoring precision. Despite advancements in remote sensing technologies, critical research gaps persist, especially regarding the biological interpretation of spectral indices and the assessment of forest resilience mechanisms at finer spatial scales. This review underscores the need for interdisciplinary approaches and advanced remote sensing frameworks to support future forest conservation and climate adaptation strategies.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"143 ","pages":"Article 104321"},"PeriodicalIF":4.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174649","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}

引用次数: 0

Investigating the planform dynamics and LULC changes with predictive insights in a Sub-Himalayan Transboundary River Basin, India 印度亚喜马拉雅跨界河流流域台地动力学和LULC变化的预测研究

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-02-01 Epub Date: 2025-12-18 DOI: 10.1016/j.pce.2025.104261

Jiarul Alam , Deepak Kumar Mandal , Jahurul Mandal , Harekrishna Manna , Sanjoy Barman , Anoop Kumar Shukla

Rapid population growth and human activities have significantly altered planform dynamics in river basins worldwide. Land Use Land Cover (LULC) changes pose significant environmental challenges by impacting hydrological processes and the sustainable management of land and water resources. This study examined the planform and LULC changes in the Raidak River basin using the CA-Markov model and geospatial techniques. The modified normalized difference water index was applied to delineate land-water boundaries, while riverbank migration, accretion, and erosion were assessed from 1930 to 2024. LULC classification for the period 1973–2024 was conducted using supervised classification methods. The results revealed that river erosion-accretion influenced land area and the rapid growth of built-up and croplands with declining vegetation, swamps, and water bodies. During 1930–2024, the most massive area was eroded by 114.91 sq. Km, where the river bank migrated an average of 801.01 m and 936.89 m by Raidak I and Raidak II, respectively. During 1973–2024, areas under different land uses were mostly converted to built-up and agricultural areas. As compared to 2024, the cropland, sandbar, and built-up areas were expected to increase by 0.14, 11.55, and 7.39 %, respectively, while dense vegetation, open vegetation, swamp, and water bodies were predicted to shrink by 5.80, 7.07, 23.56, and 4.91 %, respectively, by 2034. These insights allow regional authorities to develop evidence-based policies for sustainable environmental management. Additionally, the projected LULC patterns can enhance hybrid models, enabling a comprehensive assessment of their impact on natural resources.

人口的快速增长和人类活动显著改变了世界各地河流流域的台地动态。土地利用和土地覆盖变化通过影响水文过程以及土地和水资源的可持续管理，构成了重大的环境挑战。本研究利用CA-Markov模型和地理空间技术研究了来达克河流域的平台和LULC变化。采用修正的归一化水差指数来划定陆地-水边界，并对1930 - 2024年的河岸迁移、加积和侵蚀进行了评估。采用监督分类方法对1973-2024年的LULC进行分类。结果表明，河流侵蚀增积影响了土地面积，影响了植被减少、沼泽和水体减少的建成区和耕地的快速增长。1930-2024年间，最大面积的侵蚀面积为114.91平方公里。Raidak I和Raidak II的平均河岸迁移量分别为801.01 m和936.89 m。在1973-2024年期间，不同土地利用方式下的区域大多转变为建成区和农业区。与2024年相比，到2034年，耕地、沙洲和建成区面积分别增加0.14%、11.55%和7.39%，而茂密植被、开阔植被、沼泽和水体面积分别减少5.80%、7.07、23.56%和4.91%。这些见解使地区当局能够制定以证据为基础的可持续环境管理政策。此外，预估的LULC模式可以增强混合模式，从而能够全面评估其对自然资源的影响。

{"title":"Investigating the planform dynamics and LULC changes with predictive insights in a Sub-Himalayan Transboundary River Basin, India","authors":"Jiarul Alam , Deepak Kumar Mandal , Jahurul Mandal , Harekrishna Manna , Sanjoy Barman , Anoop Kumar Shukla","doi":"10.1016/j.pce.2025.104261","DOIUrl":"10.1016/j.pce.2025.104261","url":null,"abstract":"<div><div>Rapid population growth and human activities have significantly altered planform dynamics in river basins worldwide. Land Use Land Cover (LULC) changes pose significant environmental challenges by impacting hydrological processes and the sustainable management of land and water resources. This study examined the planform and LULC changes in the Raidak River basin using the CA-Markov model and geospatial techniques. The modified normalized difference water index was applied to delineate land-water boundaries, while riverbank migration, accretion, and erosion were assessed from 1930 to 2024. LULC classification for the period 1973–2024 was conducted using supervised classification methods. The results revealed that river erosion-accretion influenced land area and the rapid growth of built-up and croplands with declining vegetation, swamps, and water bodies. During 1930–2024, the most massive area was eroded by 114.91 sq. Km, where the river bank migrated an average of 801.01 m and 936.89 m by Raidak I and Raidak II, respectively. During 1973–2024, areas under different land uses were mostly converted to built-up and agricultural areas. As compared to 2024, the cropland, sandbar, and built-up areas were expected to increase by 0.14, 11.55, and 7.39 %, respectively, while dense vegetation, open vegetation, swamp, and water bodies were predicted to shrink by 5.80, 7.07, 23.56, and 4.91 %, respectively, by 2034. These insights allow regional authorities to develop evidence-based policies for sustainable environmental management. Additionally, the projected LULC patterns can enhance hybrid models, enabling a comprehensive assessment of their impact on natural resources.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104261"},"PeriodicalIF":4.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884574","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}

引用次数: 0

Robust monthly precipitation forecasts using VMD-decomposed hybrid LSSVM–BiLSTM model optimized by Harris Hawks algorithm 基于Harris Hawks算法优化的vmd分解混合LSSVM-BiLSTM模型的稳健月降水预报

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-02-01 Epub Date: 2025-12-29 DOI: 10.1016/j.pce.2025.104257

Behnoush Farokhzadeh , Sepideh Choobeh , Rasool Imani

Accurate prediction of monthly precipitation is crucial for sustainable water resources management in climate-sensitive regions such as the Urmia Lake Basin, northwestern Iran. This study proposes an optimized hybrid forecasting framework integrating Variational Mode Decomposition (VMD), Least Squares Support Vector Machine (LSSVM), Bidirectional Long Short-Term Memory (BiLSTM) networks, and the Harris Hawks Optimization (HHO) algorithm. The normalized monthly precipitation series (1980–2024) was decomposed into five Intrinsic Mode Functions (IMFs) and a residual component using VMD. LSSVM was applied to model the high- and medium-frequency IMFs, while BiLSTM captured the long-term residual trend. The HHO algorithm was employed to optimize model hyperparameters, minimizing reconstruction and prediction errors. Model evaluation revealed outstanding predictive performance, with Nash–Sutcliffe Efficiency (NSE) exceeding 0.97 and Root Mean Square Error (RMSE) below 1.6 mm across both training and testing phases, and strong correlation between observed and simulated rainfall. Bootstrap-based uncertainty analysis confirmed the model's stability and reliability. The optimized hybrid model was further applied to project monthly precipitation for 2030–2050, indicating an approximate 7 % increase in mean annual rainfall relative to the historical average (1980–2024). These results demonstrate that the VMD–HHO–LSSVM–BiLSTM hybrid framework effectively captures nonlinear, nonstationary rainfall dynamics across multiple temporal scales and provides a robust tool for hydrological forecasting and adaptive water management in the Urmia Lake Basin.

月降水量的准确预测对于气候敏感地区（如伊朗西北部的乌尔米亚湖盆地）的可持续水资源管理至关重要。本文提出了一种结合变分模态分解（VMD）、最小二乘支持向量机（LSSVM）、双向长短期记忆（BiLSTM）网络和哈里斯鹰优化（HHO）算法的优化混合预测框架。将逐月降水序列（1980—2024）归一化，利用VMD分解为5个本征模态函数（IMFs）和1个残差分量。LSSVM用于高频和中频IMFs的建模，而BiLSTM捕获了长期残差趋势。采用HHO算法优化模型超参数，使重建和预测误差最小化。模型的预测效果非常好，在训练和测试阶段，Nash-Sutcliffe效率（NSE）均超过0.97，均方根误差（RMSE）均低于1.6 mm，且观测值与模拟值之间具有很强的相关性。基于bootstrap的不确定性分析验证了模型的稳定性和可靠性。将优化后的混合模型进一步应用于2030-2050年的月降水量预测，结果表明，相对于1980-2024年的历史平均值，年平均降雨量增加了约7%。这些结果表明，VMD-HHO-LSSVM-BiLSTM混合框架可以有效地捕获跨时间尺度的非线性、非平稳降雨动态，为乌尔米亚湖流域的水文预报和适应性水管理提供了强有力的工具。

{"title":"Robust monthly precipitation forecasts using VMD-decomposed hybrid LSSVM–BiLSTM model optimized by Harris Hawks algorithm","authors":"Behnoush Farokhzadeh , Sepideh Choobeh , Rasool Imani","doi":"10.1016/j.pce.2025.104257","DOIUrl":"10.1016/j.pce.2025.104257","url":null,"abstract":"<div><div>Accurate prediction of monthly precipitation is crucial for sustainable water resources management in climate-sensitive regions such as the Urmia Lake Basin, northwestern Iran. This study proposes an optimized hybrid forecasting framework integrating Variational Mode Decomposition (VMD), Least Squares Support Vector Machine (LSSVM), Bidirectional Long Short-Term Memory (BiLSTM) networks, and the Harris Hawks Optimization (HHO) algorithm. The normalized monthly precipitation series (1980–2024) was decomposed into five Intrinsic Mode Functions (IMFs) and a residual component using VMD. LSSVM was applied to model the high- and medium-frequency IMFs, while BiLSTM captured the long-term residual trend. The HHO algorithm was employed to optimize model hyperparameters, minimizing reconstruction and prediction errors. Model evaluation revealed outstanding predictive performance, with Nash–Sutcliffe Efficiency (NSE) exceeding 0.97 and Root Mean Square Error (RMSE) below 1.6 mm across both training and testing phases, and strong correlation between observed and simulated rainfall. Bootstrap-based uncertainty analysis confirmed the model's stability and reliability. The optimized hybrid model was further applied to project monthly precipitation for 2030–2050, indicating an approximate 7 % increase in mean annual rainfall relative to the historical average (1980–2024). These results demonstrate that the VMD–HHO–LSSVM–BiLSTM hybrid framework effectively captures nonlinear, nonstationary rainfall dynamics across multiple temporal scales and provides a robust tool for hydrological forecasting and adaptive water management in the Urmia Lake Basin.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104257"},"PeriodicalIF":4.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884578","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}

引用次数: 0

Agro-climatic classification of Türkiye's Agricultural Basins using long-term ERA5-Land temperature data 基于era5 -陆地温度长期数据的<s:1>基耶农业流域农业气候分类

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-02-01 Epub Date: 2026-01-19 DOI: 10.1016/j.pce.2026.104307

Mehmet Ali Çelik

The 30 main agricultural basins in Türkiye are categorized in this study using long-term mean, maximum, and minimum temperature indicators that were obtained from ERA5-Land data. Four thermally coherent clusters representing different agro-climatic regimes were formed from the basins using Principal Component Analysis (PCA) and K-means clustering. Mild and steady temperatures in Cluster 1 (coastal and lowland basins) are ideal for intensive, high-value crop production. Wide diurnal ranges and low minimum temperatures are characteristics of Cluster 2 (high-altitude and continental basins), which represents cold and frost-prone agricultural zones. The highest mean and maximum temperatures are found in Cluster 3 (the basins of the Mediterranean, Aegean, and Southeastern Anatolia), which denotes hot, dry climates with significant risks of heat and water stress. Cluster 4, which has mixed agricultural potential and moderate thermal variability, serves as a transitional zone between continental and maritime climates. Temperature trends across all agricultural basins during the 1950–2023 period were positive, with some basins (Coastal Aegean, Coastal Mediterranean, Gediz and GAP basins) experiencing an average temperature increase rate of 2–7 °C per century. Analyses reveal that minimum temperatures are increasing faster than maximum temperatures, and that nocturnal warming is a dominant climate signal across Türkiye. Strong inter-station coherence in all temperature variables was confirmed by correlation analysis, indicating that regional-scale atmospheric dynamics shape Türkiye's temperature patterns more so than local microclimatic effects. These thermal clusters offer a spatially explicit framework for adaptive agricultural planning from an agro-climatological point of view, providing information on heat stress vulnerability, evapotranspiration demand, and frost risk. The findings demonstrate how temperature-based clustering can be combined with other policy tools like crop suitability mapping, irrigation control, and agro-climatic zoning. This study contributes to the development of climate-adapted and sustainable agricultural strategies in Türkiye by identifying agricultural regions with similar temperatures.

本文利用ERA5-Land数据获得的长期平均、最高和最低温度指标，对 rkiye地区30个主要农业流域进行了分类。利用主成分分析（PCA）和K-means聚类，从流域中形成了代表不同农业气候状况的4个热相干集群。集群1（沿海和低地盆地）温和而稳定的温度是集约化、高价值作物生产的理想条件。大的日变化幅度和低的最低温度是集群2（高海拔和大陆盆地）的特征，它代表了寒冷和易霜冻的农业区。平均温度和最高温度最高的是第三组（地中海、爱琴海和安纳托利亚东南部的盆地），这表示炎热、干燥的气候，有严重的高温和缺水风险。集群4具有混合的农业潜力和适度的热变率，是大陆性和海洋性气候之间的过渡区。在1950-2023年期间，所有农业盆地的温度趋势都是积极的，其中一些盆地（爱琴海沿岸、地中海沿岸、盖迪斯和GAP盆地）的平均温度增长率为每世纪2-7°C。分析表明，最低气温的上升速度快于最高气温，夜间变暖是整个 rkiye的主要气候信号。相关分析证实了各温度变量的站间一致性较强，表明区域尺度的大气动力对 rkiye温度模式的影响大于局地小气候效应。从农业气候学的角度来看，这些热集群为适应性农业规划提供了明确的空间框架，提供了有关热胁迫脆弱性、蒸散需求和霜冻风险的信息。研究结果表明，基于温度的聚类可以与作物适宜性绘图、灌溉控制和农业气候区划等其他政策工具相结合。该研究通过确定具有相似温度的农业区，有助于制定适应气候变化的可持续农业战略。

{"title":"Agro-climatic classification of Türkiye's Agricultural Basins using long-term ERA5-Land temperature data","authors":"Mehmet Ali Çelik","doi":"10.1016/j.pce.2026.104307","DOIUrl":"10.1016/j.pce.2026.104307","url":null,"abstract":"<div><div>The 30 main agricultural basins in Türkiye are categorized in this study using long-term mean, maximum, and minimum temperature indicators that were obtained from ERA5-Land data. Four thermally coherent clusters representing different agro-climatic regimes were formed from the basins using Principal Component Analysis (PCA) and K-means clustering. Mild and steady temperatures in Cluster 1 (coastal and lowland basins) are ideal for intensive, high-value crop production. Wide diurnal ranges and low minimum temperatures are characteristics of Cluster 2 (high-altitude and continental basins), which represents cold and frost-prone agricultural zones. The highest mean and maximum temperatures are found in Cluster 3 (the basins of the Mediterranean, Aegean, and Southeastern Anatolia), which denotes hot, dry climates with significant risks of heat and water stress. Cluster 4, which has mixed agricultural potential and moderate thermal variability, serves as a transitional zone between continental and maritime climates. Temperature trends across all agricultural basins during the 1950–2023 period were positive, with some basins (Coastal Aegean, Coastal Mediterranean, Gediz and GAP basins) experiencing an average temperature increase rate of 2–7 °C per century. Analyses reveal that minimum temperatures are increasing faster than maximum temperatures, and that nocturnal warming is a dominant climate signal across Türkiye. Strong inter-station coherence in all temperature variables was confirmed by correlation analysis, indicating that regional-scale atmospheric dynamics shape Türkiye's temperature patterns more so than local microclimatic effects. These thermal clusters offer a spatially explicit framework for adaptive agricultural planning from an agro-climatological point of view, providing information on heat stress vulnerability, evapotranspiration demand, and frost risk. The findings demonstrate how temperature-based clustering can be combined with other policy tools like crop suitability mapping, irrigation control, and agro-climatic zoning. This study contributes to the development of climate-adapted and sustainable agricultural strategies in Türkiye by identifying agricultural regions with similar temperatures.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104307"},"PeriodicalIF":4.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023083","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}

引用次数: 0

Numerical simulation of CO2 migration and mineralization reaction characteristics in goaf 采空区CO2运移与矿化反应特征的数值模拟

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-02-01 Epub Date: 2025-12-02 DOI: 10.1016/j.pce.2025.104220

Zheng Dai , Shugang Li , Biao Hu , Xiangguo Kong , Jingfei Zhang , Bing Zhu , Qian Wei

Coal mine goafs, as a special type of saline aquifer characterized by multi-fracture structures, remain insufficiently studied as potential CO₂ storage sites. Therefore, this study integrates the fracture evolution of mining overlying strata with the TOUGHREACT numerical simulation to construct a two-dimensional idealized goaf model. A 50-day isothermal CO₂ injection simulation was conducted to systematically assess pressure evolution, multiphase CO₂ migration, and the acidification–mineral dissolution–precipitation processes driven by the fracture network, and to reveal the key differences in sequestration mechanisms between the goaf and a continuous geological body (CGB), clarifying the mineral evolution pathways. The results show that the fracture network significantly enhances CO₂-brine contact efficiency, forming a dissolved CO₂ distribution that envelops the surrounding minerals, and leading a triangular pressure field inside the goaf. In strongly acidic zones, primary minerals such as calcite and k-feldspar undergo rapid dissolution, releasing Ca²⁺, Mg²⁺, and Al³⁺, which promote the formation of secondary minerals such as smectite and quartz. At the same spatial and temporal scale, the CO₂ dissolution storage capacity of the goaf exceeds that of the CGB by 3261 kg, demonstrating superior sequestration performance. These findings provide theoretical support for the engineering deployment of CO₂ sequestration in such unconventional reservoirs.

煤矿采空区作为一种以多裂隙结构为特征的特殊咸水含水层，作为潜在的CO2储集场所，目前研究还不充分。因此，本研究将采动覆岩裂隙演化与TOUGHREACT数值模拟相结合，构建二维理想化采空区模型。通过50 d等温CO2注入模拟，系统评价了裂隙网络驱动下的压力演化、多相CO2运移和酸化-矿物溶解-沉淀过程，揭示了采空区与连续地质体（CGB）封存机制的关键差异，阐明了矿物演化路径。结果表明：裂缝网络显著提高了CO2-卤水接触效率，形成了一个包裹周围矿物的溶解CO2分布，并在采空区内部形成了一个三角形压力场；在强酸性带中，方解石、钾长石等原生矿物快速溶解，释放出Ca2+、Mg2+和Al3+，促进蒙脱石、石英等次生矿物的形成。在相同时空尺度下，采空区的CO2溶蚀库容量比CGB大3261 kg，表现出较好的固存性能。这些发现为此类非常规储层二氧化碳封存技术的工程部署提供了理论支持。

{"title":"Numerical simulation of CO2 migration and mineralization reaction characteristics in goaf","authors":"Zheng Dai , Shugang Li , Biao Hu , Xiangguo Kong , Jingfei Zhang , Bing Zhu , Qian Wei","doi":"10.1016/j.pce.2025.104220","DOIUrl":"10.1016/j.pce.2025.104220","url":null,"abstract":"<div><div>Coal mine goafs, as a special type of saline aquifer characterized by multi-fracture structures, remain insufficiently studied as potential CO<sub>2</sub> storage sites. Therefore, this study integrates the fracture evolution of mining overlying strata with the TOUGHREACT numerical simulation to construct a two-dimensional idealized goaf model. A 50-day isothermal CO<sub>2</sub> injection simulation was conducted to systematically assess pressure evolution, multiphase CO<sub>2</sub> migration, and the acidification–mineral dissolution–precipitation processes driven by the fracture network, and to reveal the key differences in sequestration mechanisms between the goaf and a continuous geological body (CGB), clarifying the mineral evolution pathways. The results show that the fracture network significantly enhances CO<sub>2</sub>-brine contact efficiency, forming a dissolved CO<sub>2</sub> distribution that envelops the surrounding minerals, and leading a triangular pressure field inside the goaf. In strongly acidic zones, primary minerals such as calcite and k-feldspar undergo rapid dissolution, releasing Ca<sup>2+</sup>, Mg<sup>2+</sup>, and Al<sup>3+</sup>, which promote the formation of secondary minerals such as smectite and quartz. At the same spatial and temporal scale, the CO<sub>2</sub> dissolution storage capacity of the goaf exceeds that of the CGB by 3261 kg, demonstrating superior sequestration performance. These findings provide theoretical support for the engineering deployment of CO<sub>2</sub> sequestration in such unconventional reservoirs.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104220"},"PeriodicalIF":4.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685426","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}

引用次数: 0

Contamination risk assessment of shallow groundwater in Xi'an City's plain area based on groundwater vulnerability, pollution loading, and groundwater function value 基于地下水脆弱性、污染负荷和地下水功能值的西安市平原区浅层地下水污染风险评价

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-02-01 Epub Date: 2026-01-09 DOI: 10.1016/j.pce.2026.104281

Yan Liu , Jialu Pang , Weiwei Hu , Zhifeng Jia , Jia Chen , Hao Liu , Rui Lei

Groundwater contamination risk assessment is essential for effective groundwater management. This study proposes a framework for quantifying groundwater contamination risk in the Plain of Xi'an, China, integrating pollution loading, vulnerability, and function value. Pollution loading was assessed based on pollutant toxicity, likelihood of release, and potential release quantity. The DRASTIC model was optimized to reflect differences in net recharge between urban impermeable areas and agricultural areas for vulnerability assessment. Function value assessment included both in situ and extractive values. By integrating groundwater vulnerability and pollution loading, the groundwater basic contamination risk was generated, which showed that areas with high basic contamination risk comprise about 2.80 % of the plain, predominantly in overlapping pollution source zones on floodplains and alluvial fans. These zones are characterized by high vulnerability, and the presence of high-load pollution sources exacerbates the risk of contamination. Groundwater quality affirms the reliability of the basic contamination risk assessment. To further assess the consequences of pollution, a value-weighted contamination risk map was generated by introducing groundwater function value. High value-weighted contamination risk areas account for approximately 0.60 % of the plain, concentrated in floodplain zones with both high basic risk and function value. These areas require prioritized pollution prevention and control measures. The findings help identify priority areas for risk management and control, providing critical support for groundwater protection.

地下水污染风险评价是地下水有效管理的基础。基于污染负荷、脆弱性和功能价值，提出了西安平原地下水污染风险量化框架。根据污染物的毒性、释放的可能性和潜在的释放量来评估污染负荷。对DRASTIC模型进行了优化，以反映城市不透水区和农业区的净补给差异，用于脆弱性评价。功能价值评估包括原位价值和提取价值。综合地下水脆弱性和污染负荷，得出地下水基本污染风险值，表明高基本污染风险区约占平原面积的2.80%，主要集中在洪泛平原和冲积扇的污染源叠加区。这些地区的特点是高脆弱性，高负荷污染源的存在加剧了污染的风险。地下水水质验证了基本污染风险评价的可靠性。为了进一步评估污染后果，引入地下水功能值，生成了价值加权污染风险图。高价值加权污染风险区约占平原面积的0.60%，主要集中在基本风险和功能价值均较高的洪泛平原地区。这些地区需要优先采取污染防治措施。研究结果有助于确定风险管理和控制的重点领域，为地下水保护提供关键支持。

{"title":"Contamination risk assessment of shallow groundwater in Xi'an City's plain area based on groundwater vulnerability, pollution loading, and groundwater function value","authors":"Yan Liu , Jialu Pang , Weiwei Hu , Zhifeng Jia , Jia Chen , Hao Liu , Rui Lei","doi":"10.1016/j.pce.2026.104281","DOIUrl":"10.1016/j.pce.2026.104281","url":null,"abstract":"<div><div>Groundwater contamination risk assessment is essential for effective groundwater management. This study proposes a framework for quantifying groundwater contamination risk in the Plain of Xi'an, China, integrating pollution loading, vulnerability, and function value. Pollution loading was assessed based on pollutant toxicity, likelihood of release, and potential release quantity. The DRASTIC model was optimized to reflect differences in net recharge between urban impermeable areas and agricultural areas for vulnerability assessment. Function value assessment included both in situ and extractive values. By integrating groundwater vulnerability and pollution loading, the groundwater basic contamination risk was generated, which showed that areas with high basic contamination risk comprise about 2.80 % of the plain, predominantly in overlapping pollution source zones on floodplains and alluvial fans. These zones are characterized by high vulnerability, and the presence of high-load pollution sources exacerbates the risk of contamination. Groundwater quality affirms the reliability of the basic contamination risk assessment. To further assess the consequences of pollution, a value-weighted contamination risk map was generated by introducing groundwater function value. High value-weighted contamination risk areas account for approximately 0.60 % of the plain, concentrated in floodplain zones with both high basic risk and function value. These areas require prioritized pollution prevention and control measures. The findings help identify priority areas for risk management and control, providing critical support for groundwater protection.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104281"},"PeriodicalIF":4.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977168","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}

引用次数: 0

Understanding groundwater mineralization and controls source apportioning using multivariate statistics, modeling, and sensitivity analysis in the Eastern parts of Ghana 利用多元统计、建模和敏感性分析了解加纳东部地下水矿化和控制源分配

IF 4.1 3区地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY

Physics and Chemistry of the Earth

Pub Date : 2026-02-01 Epub Date: 2026-01-14 DOI: 10.1016/j.pce.2026.104301

Mahamuda Abu, Musah Saeed Zango, Maxwell Anim-Gyampo

Groundwater chemistry has been continuously modified by natural and human activities, justifying the need for continuous evaluation of groundwater geochemistry. The hydrochemistry of groundwater in the southeastern parts of Ghana suggests highly evolved groundwater. However, the controlling factors of the hydrochemical modifications are not well constrained. Hence, to characterize the mineralization and factor-sensitive processes controlling the hydrochemical changes in the area, the study adopted factor analysis (FA), prediction modeling, sensitivity analysis (for the first time), and spatial mapping. The following: TDS, HCO₃⁻, K⁺, Na⁺, and F⁻ are found to be higher than their respective WHO standards in 81 %, 63 %, 44 %, 46 %, and 11 % of the area, respectively. These have an implication of poor taste, odor, and palatability (TDS), corrosiveness of plumbing materials (HCO₃⁻), hyperkalemia (K⁺), kidney and heart-related problems (Na⁺), and high F⁻ health-related issues. The ratio proxies and discriminant plots reveal dissolution of silicate and carbonate minerals as the main controlling processes of the groundwater mineralization. This was corroborated by 3 principal components (PCs) with 40.8 %, 14.5 %, and 10.4 % of variance for PCs 1, 2, and 3, respectively. The prediction model and sensitivity analysis indicated that SO₄²⁻ has a positive influence on TDS and HCO₃⁻. In contrast, HCO₃⁻ and Mg²⁺ have a reverse influence on TDS and HCO₃⁻, respectively. Also, K⁺ is controlled by Na⁺ and Na⁺ by K⁺ in the groundwater. Hot spots of TDS, HCO₃⁻, K⁺, Na⁺, and F⁻ are observed at Asuotwene, Tabibian, Koforidua, Asokore, Dome, and Amasaman.

由于自然和人类活动不断改变地下水化学性质，因此需要对地下水地球化学进行持续评价。加纳东南部地下水的水化学表明，这是一种高度演化的地下水。然而，水化学改性的控制因素并没有得到很好的约束。为此，采用因子分析（FA）、预测建模、敏感性分析（首次）和空间填图等方法，对控制该区水化学变化的矿化和因子敏感过程进行了表征。TDS、HCO3−、K+、Na+和F−分别有81%、63%、44%、46%和11%的地区高于WHO标准。这些问题包括味道、气味和适口性差（TDS）、管道材料的腐蚀性（HCO3−）、高钾血症（K+）、肾脏和心脏相关问题（Na+）以及高F−与健康相关的问题。比值指标和判别图显示，硅酸盐和碳酸盐矿物的溶蚀作用是地下水成矿作用的主要控制过程。这被3个主成分（PCs）证实，PCs 1、2和3的方差分别为40.8%、14.5%和10.4%。预测模型和敏感性分析表明，SO42−对TDS和HCO3−有正向影响。相反，HCO3−和Mg2+对TDS和HCO3−的影响相反。地下水中K+受Na+控制，Na+受K+控制。在Asuotwene、Tabibian、Koforidua、Asokore、Dome和Amasaman观测到TDS、HCO3−、K+、Na+和F−的热点。

{"title":"Understanding groundwater mineralization and controls source apportioning using multivariate statistics, modeling, and sensitivity analysis in the Eastern parts of Ghana","authors":"Mahamuda Abu, Musah Saeed Zango, Maxwell Anim-Gyampo","doi":"10.1016/j.pce.2026.104301","DOIUrl":"10.1016/j.pce.2026.104301","url":null,"abstract":"<div><div>Groundwater chemistry has been continuously modified by natural and human activities, justifying the need for continuous evaluation of groundwater geochemistry. The hydrochemistry of groundwater in the southeastern parts of Ghana suggests highly evolved groundwater. However, the controlling factors of the hydrochemical modifications are not well constrained. Hence, to characterize the mineralization and factor-sensitive processes controlling the hydrochemical changes in the area, the study adopted factor analysis (FA), prediction modeling, sensitivity analysis (for the first time), and spatial mapping. The following: TDS, HCO<sub>3</sub><sup>−</sup>, K<sup>+</sup>, Na<sup>+</sup>, and F<sup>−</sup> are found to be higher than their respective WHO standards in 81 %, 63 %, 44 %, 46 %, and 11 % of the area, respectively. These have an implication of poor taste, odor, and palatability (TDS), corrosiveness of plumbing materials (HCO<sub>3</sub><sup>−</sup>), hyperkalemia (K<sup>+</sup>), kidney and heart-related problems (Na<sup>+</sup>), and high F<sup>−</sup> health-related issues. The ratio proxies and discriminant plots reveal dissolution of silicate and carbonate minerals as the main controlling processes of the groundwater mineralization. This was corroborated by 3 principal components (PCs) with 40.8 %, 14.5 %, and 10.4 % of variance for PCs 1, 2, and 3, respectively. The prediction model and sensitivity analysis indicated that SO<sub>4</sub><sup>2−</sup> has a positive influence on TDS and HCO<sub>3</sub><sup>−</sup>. In contrast, HCO<sub>3</sub><sup>−</sup> and Mg<sup>2+</sup> have a reverse influence on TDS and HCO<sub>3</sub><sup>−</sup>, respectively. Also, K<sup>+</sup> is controlled by Na<sup>+</sup> and Na<sup>+</sup> by K<sup>+</sup> in the groundwater. Hot spots of TDS, HCO<sub>3</sub><sup>−</sup>, K<sup>+</sup>, Na<sup>+</sup>, and F<sup>−</sup> are observed at Asuotwene, Tabibian, Koforidua, Asokore, Dome, and Amasaman.</div></div>","PeriodicalId":54616,"journal":{"name":"Physics and Chemistry of the Earth","volume":"142 ","pages":"Article 104301"},"PeriodicalIF":4.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977766","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}

引用次数: 0