Pub Date : 2026-01-10DOI: 10.1016/j.watres.2026.125371
Yong Wang , Liang Cui , Zhou Zhou , Renju Liu , Wanpeng Wang , Lijing Jiang , Xuesong Yi , Zongze Shao
Preserved fruit processing wastewater, characterized by high levels of salinity, acidity, and organic load, presents significant challenges to conventional biological treatment technologies, which typically necessitate pH adjustment and salinity reduction. For the first time, a fungal consortium predominantly composed of Candida, Pichia, and Saccharomyces was employed in this study to directly deal with highly saline and acidic preserved plum processing wastewater, in order to achieve COD removal and pH enhancement. COD fractionation in preserved plum processing wastewater by virtue of particle size distribution and oxygen uptake rate shows that 84% of COD was soluble, and 97% of COD were degradable for the inoculated fungal consortium which was characterized by an extremely low biomass yield coefficient of 0.28 mg COD-biomass∙(mg COD-wastewater)-1. In contrast to the conventional COD:N:P ratio of 100:5:1, fungal degradation utilized significantly less nitrogen and phosphorus, exhibiting a COD:N:P ratio of 386–770:4.4:1. Based on the results of pH and salinity effects, this fungal consortium demonstrated optimal performance within a pH range of 3–4 and a salinity range of 3%-5%, which aligns well with the typical conditions of preserved fruit processing wastewater. In the full-scale preserved plum processing wastewater treatment facility with a capacity of 50 m3∙d-1 employing a novel combined process of fungal degradation and activated sludge treatment, the COD removal efficiencies during fungal degradation and subsequent activated sludge treatment were 67% and 27%, respectively, with pH being elevated from 3.42 to 6.52 solely through fungal degradation. Microbial structure analysis indicated that the inoculated fungal consortium was firmly established in the fungal degradation tank, constituting a substantial proportion of the microbial community. Species interaction network analysis revealed that positive correlations predominated within fungal and bacterial communities, particularly among Candida, Pichia, and Saccharomyces, suggesting a collaborative relationship among these species. The findings of this study confirm that fungal degradation is a viable and promising alternative to traditional biological treatment technologies for highly saline and acidic wastewater from preserved fruit processing.
{"title":"Can fungal degradation replace conventional biological processes for treatment of highly acidic and saline preserved fruit processing wastewater by virtue of Candida, Pichia and Saccharomyces?","authors":"Yong Wang , Liang Cui , Zhou Zhou , Renju Liu , Wanpeng Wang , Lijing Jiang , Xuesong Yi , Zongze Shao","doi":"10.1016/j.watres.2026.125371","DOIUrl":"10.1016/j.watres.2026.125371","url":null,"abstract":"<div><div>Preserved fruit processing wastewater, characterized by high levels of salinity, acidity, and organic load, presents significant challenges to conventional biological treatment technologies, which typically necessitate pH adjustment and salinity reduction. For the first time, a fungal consortium predominantly composed of <em>Candida, Pichia, and Saccharomyces</em> was employed in this study to directly deal with highly saline and acidic preserved plum processing wastewater, in order to achieve COD removal and pH enhancement. COD fractionation in preserved plum processing wastewater by virtue of particle size distribution and oxygen uptake rate shows that 84% of COD was soluble, and 97% of COD were degradable for the inoculated fungal consortium which was characterized by an extremely low biomass yield coefficient of 0.28 mg COD-biomass∙(mg COD-wastewater)<sup>-1</sup>. In contrast to the conventional COD:N:P ratio of 100:5:1, fungal degradation utilized significantly less nitrogen and phosphorus, exhibiting a COD:N:P ratio of 386–770:4.4:1. Based on the results of pH and salinity effects, this fungal consortium demonstrated optimal performance within a pH range of 3–4 and a salinity range of 3%-5%, which aligns well with the typical conditions of preserved fruit processing wastewater. In the full-scale preserved plum processing wastewater treatment facility with a capacity of 50 m<sup>3</sup>∙d<sup>-1</sup> employing a novel combined process of fungal degradation and activated sludge treatment, the COD removal efficiencies during fungal degradation and subsequent activated sludge treatment were 67% and 27%, respectively, with pH being elevated from 3.42 to 6.52 solely through fungal degradation. Microbial structure analysis indicated that the inoculated fungal consortium was firmly established in the fungal degradation tank, constituting a substantial proportion of the microbial community. Species interaction network analysis revealed that positive correlations predominated within fungal and bacterial communities, particularly among <em>Candida, Pichia, and Saccharomyces</em>, suggesting a collaborative relationship among these species. The findings of this study confirm that fungal degradation is a viable and promising alternative to traditional biological treatment technologies for highly saline and acidic wastewater from preserved fruit processing.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125371"},"PeriodicalIF":12.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-10DOI: 10.1016/j.watres.2026.125372
Hanxin Zhang , Haiping Zhang , Jia Liu , Kaiming Peng , Ru Guo , Feiyun Sun , Xiangfeng Huang , Chen Cai
Accurate prediction of anomalous effluent quality fluctuations in wastewater treatment plants (WWTPs) is essential for safeguarding aquatic environments, and the main challenge lies in limited time-series data and imbalanced samples. However, existing transfer learning methods have difficulty identifying anomalous fluctuations in this heterogeneous system due to influent variability and process diversity. To address this, a causal-informed temporal domain adaptation model enhanced with cross-attention (CA-TDA) is proposed for time-series prediction. Unlike conventional methods that focus on distribution alignment, CA-TDA dynamically captures and aligns causal structures across heterogeneous domains. Temporal causal features are extracted via a variational autoencoder (VAE), upon which cross-attention identifies key driving factors for effective cross-domain adaptation. Experiments show that even when the target domain contains only 8.35 % of source samples, CA-TDA achieves superior prediction of multiple effluent quality indicators (EQIs), with an average R² of 0.9669 and a peak R² of 0.9820 for TN. Compared to baselines, F1- and F2-scores for consecutive anomalous fluctuations are improved by 29–87 % and 30–89 %, respectively. These results confirm that CA-TDA effectively captures complex, non-stationary fluctuations, substantially enhancing the accuracy and robustness of cross-domain effluent anomaly prediction, demonstrating broad application potential.
{"title":"Causal-informed domain adaptation with cross-attention for predicting anomalous effluent quality fluctuations in wastewater treatment plants","authors":"Hanxin Zhang , Haiping Zhang , Jia Liu , Kaiming Peng , Ru Guo , Feiyun Sun , Xiangfeng Huang , Chen Cai","doi":"10.1016/j.watres.2026.125372","DOIUrl":"10.1016/j.watres.2026.125372","url":null,"abstract":"<div><div>Accurate prediction of anomalous effluent quality fluctuations in wastewater treatment plants (WWTPs) is essential for safeguarding aquatic environments, and the main challenge lies in limited time-series data and imbalanced samples. However, existing transfer learning methods have difficulty identifying anomalous fluctuations in this heterogeneous system due to influent variability and process diversity. To address this, a causal-informed temporal domain adaptation model enhanced with cross-attention (CA-TDA) is proposed for time-series prediction. Unlike conventional methods that focus on distribution alignment, CA-TDA dynamically captures and aligns causal structures across heterogeneous domains. Temporal causal features are extracted via a variational autoencoder (VAE), upon which cross-attention identifies key driving factors for effective cross-domain adaptation. Experiments show that even when the target domain contains only 8.35 % of source samples, CA-TDA achieves superior prediction of multiple effluent quality indicators (EQIs), with an average R² of 0.9669 and a peak R² of 0.9820 for TN. Compared to baselines, F1- and F2-scores for consecutive anomalous fluctuations are improved by 29–87 % and 30–89 %, respectively. These results confirm that CA-TDA effectively captures complex, non-stationary fluctuations, substantially enhancing the accuracy and robustness of cross-domain effluent anomaly prediction, demonstrating broad application potential.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125372"},"PeriodicalIF":12.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-10DOI: 10.1016/j.watres.2026.125370
Yuanyuan Xu , Shiqi Tian , Ruihua Cao , Wei Lin , Jun Ma , Gang Wen
The occurrence of harmful algal blooms (HABs) poses serious ecological concerns. Photocatalytic oxidation has emerged as an ideal alternative to conventional algal removal strategies, owing to its minimal energy requirements. Here, we prepared a novel photocatalyst by fabricating molybdenum disulfide nanosheets with bismuth oxyhalide (MS/BBI). Successful construction of the heterojunction effectively facilitated the separation and migration of photogenerated carriers, thereby significantly enhancing the photocatalytic performance of the catalyst. The MS/BBI photocatalyst achieved 96.8% removal efficiency for Microcystis aeruginosa (M. aeruginosa) after 3 h light irradiation, which was higher than the performance of MoS2 and BBI alone. Comprehensive physiological analyses revealed substantial changes in cell activity, accompanied by electrolyte leakage, lipid peroxidation, disruption of photosynthetic and antioxidant systems, and depletion of intracellular organic matter. Notably, only the light-irradiated MS/BBI system resulted in irreversible damage to algal cells, completely suppressing the regrowth potential of algae. Experiments identified superoxide radicals as the predominant reactive oxygen species that mediated the defensive response in algal cells, triggering the production of extracellular polymeric substances, which in turn facilitated the self-flocculation of algal cells into flocs. The synergistic effect of photo-oxidation and flocculation enabled the efficient removal of M. aeruginosa. Furthermore, the solution after photocatalytic treatment demonstrated good biocompatibility, indicating a negligible environmental risk. This study reports a novel photocatalyst designed for the rapid removal of algae. It identifies a flocculation mechanism present during the photocatalytic process, providing fundamental understanding for future research on HABs control via photocatalysis.
{"title":"Rapid elimination of harmful cyanobacteria via synergistic effects of photo-oxidation and flocculation","authors":"Yuanyuan Xu , Shiqi Tian , Ruihua Cao , Wei Lin , Jun Ma , Gang Wen","doi":"10.1016/j.watres.2026.125370","DOIUrl":"10.1016/j.watres.2026.125370","url":null,"abstract":"<div><div>The occurrence of harmful algal blooms (HABs) poses serious ecological concerns. Photocatalytic oxidation has emerged as an ideal alternative to conventional algal removal strategies, owing to its minimal energy requirements. Here, we prepared a novel photocatalyst by fabricating molybdenum disulfide nanosheets with bismuth oxyhalide (MS/BBI). Successful construction of the heterojunction effectively facilitated the separation and migration of photogenerated carriers, thereby significantly enhancing the photocatalytic performance of the catalyst. The MS/BBI photocatalyst achieved 96.8% removal efficiency for <em>Microcystis aeruginosa</em> (<em>M. aeruginosa</em>) after 3 h light irradiation, which was higher than the performance of MoS<sub>2</sub> and BBI alone. Comprehensive physiological analyses revealed substantial changes in cell activity, accompanied by electrolyte leakage, lipid peroxidation, disruption of photosynthetic and antioxidant systems, and depletion of intracellular organic matter. Notably, only the light-irradiated MS/BBI system resulted in irreversible damage to algal cells, completely suppressing the regrowth potential of algae. Experiments identified superoxide radicals as the predominant reactive oxygen species that mediated the defensive response in algal cells, triggering the production of extracellular polymeric substances, which in turn facilitated the self-flocculation of algal cells into flocs. The synergistic effect of photo-oxidation and flocculation enabled the efficient removal of <em>M. aeruginosa</em>. Furthermore, the solution after photocatalytic treatment demonstrated good biocompatibility, indicating a negligible environmental risk. This study reports a novel photocatalyst designed for the rapid removal of algae. It identifies a flocculation mechanism present during the photocatalytic process, providing fundamental understanding for future research on HABs control via photocatalysis.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125370"},"PeriodicalIF":12.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-10DOI: 10.1016/j.watres.2026.125366
Xiao-Na Zhao, Zi-Yi Han, Lu Wang, Zhi Gao, Zhuang-Song Huang, Yu-Lei Liu, Jun Ma
Pivotal roles of high-valent intermediate iron species [Fe(V) and Fe(IV)] in ferrate system have been widely recognized, yet their individual contributions remain unclear. Herein, we conducted systematic investigations to differentiate Fe(VI), Fe(V) and Fe(IV) in ferrate systems. By using pyrophosphate to sequester Fe(V), we observed three distinct variation trends in the apparent rate constant (kapp,OCs,PP) with respect to the reactant molar ratio: positive correlation, negative correlation and no correlation (remaining constant). Kinetic model simulations revealed that the variation trends of kapp,OCs,PP depended on the relative reactivity of Fe(IV), and further established a relationship between kapp,OCs,PP and the actual rate constants for the reactions of Fe(VI) species with contaminants (kFe(VI)-OCs). Based on these findings, we proposed a generalized procedure to determine kFe(VI)-OCs. Subsequently, we developed a novel approach, termed the targeted sequestered kinetic method, to evaluate the individual contributions of Fe(VI), Fe(V), and Fe(IV). Experimental verification using methyl phenyl sulfoxide as a model compound demonstrated that the targeted sequestered kinetic method reliably quantifies the contributions of individual high-valent iron species. Overall, this study offers a robust method to identify high-valent iron species and advances the in-depth understanding of ferrate chemistry.
{"title":"Quantifying the contributions of Fe(VI), Fe(V), and Fe(IV) in contaminant oxidation by ferrate: A kinetic insight","authors":"Xiao-Na Zhao, Zi-Yi Han, Lu Wang, Zhi Gao, Zhuang-Song Huang, Yu-Lei Liu, Jun Ma","doi":"10.1016/j.watres.2026.125366","DOIUrl":"10.1016/j.watres.2026.125366","url":null,"abstract":"<div><div>Pivotal roles of high-valent intermediate iron species [Fe(V) and Fe(IV)] in ferrate system have been widely recognized, yet their individual contributions remain unclear. Herein, we conducted systematic investigations to differentiate Fe(VI), Fe(V) and Fe(IV) in ferrate systems. By using pyrophosphate to sequester Fe(V), we observed three distinct variation trends in the apparent rate constant (<em>k</em><sub>app,OCs,PP</sub>) with respect to the reactant molar ratio: positive correlation, negative correlation and no correlation (remaining constant). Kinetic model simulations revealed that the variation trends of <em>k</em><sub>app,OCs,PP</sub> depended on the relative reactivity of Fe(IV), and further established a relationship between <em>k</em><sub>app,OCs,PP</sub> and the actual rate constants for the reactions of Fe(VI) species with contaminants (<em>k</em><sub>Fe(VI)-OCs</sub>). Based on these findings, we proposed a generalized procedure to determine <em>k</em><sub>Fe(VI)-OCs</sub>. Subsequently, we developed a novel approach, termed the targeted sequestered kinetic method, to evaluate the individual contributions of Fe(VI), Fe(V), and Fe(IV). Experimental verification using methyl phenyl sulfoxide as a model compound demonstrated that the targeted sequestered kinetic method reliably quantifies the contributions of individual high-valent iron species. Overall, this study offers a robust method to identify high-valent iron species and advances the in-depth understanding of ferrate chemistry.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125366"},"PeriodicalIF":12.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956945","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-10DOI: 10.1016/j.watres.2026.125367
Mehran Mahdian, Roohollah Noori, Mohammad Javad Saravani, Ali Reza Shahvaran, Mohsen Shahmohammad, Paul P.J. Gaffney, Mohammad Milad Salamattalab, Milad Shamsi Anboohi, Majid Hosseinzadeh, Fan Xia, Yongqiang Zhou, Yunlin Zhang, Mikko Kolehmainen, Soroush Abolfathi
Understanding water quality in the deeper layers of stratified lakes is critical, as these zones govern ecosystem stability and biodiversity health. The hypolimnion, the bottommost layer of a stratified lake, plays a pivotal role by limiting vertical circulation and promoting pollutant accumulation, thus serving as an indicator of long-term lake condition. While surface water quality can now be routinely monitored using <ce:italic>in-situ</ce:italic> sensors and satellite observations, assessing hypolimnetic conditions remains costly and logistically challenging, particularly in the Arctic regions with severe environmental conditions and logistical constraints. Using a long-term dataset spanning 1979–2022 collected at a single monitoring station where both epilimnetic and hypolimnetic profiles are measured, this study develops and compares five machine and deep learning models including artificial neural networks (ANN), random forest (RF), extreme gradient boosting (XGBoost), support vector regression (SVR), and Kolmogorov–Arnold networks (KAN) to estimate hypolimnetic total nitrogen (TN), total phosphorus (TP), and dissolved oxygen (DO) in Lake Inari, Finland, based on readily available epilimnetic water quality predictors. Model performance was evaluated using five-fold cross-validation and assessed with the Nash–Sutcliffe efficiency (<ce:italic>NSE</ce:italic>), normalized mean absolute error (<ce:italic>NMAE</ce:italic>), and coefficient of determination (<ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic>). For TN, the RF model achieved the best overall performance, with mean cross-validation values of <ce:italic>NSE</ce:italic> = 0.52, <ce:italic>NMAE</ce:italic> = 0.11, and <ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic> = 0.52, outperforming the other models, which yielded <ce:italic>NSE</ce:italic> values of 0.38–0.52, <ce:italic>NMAE</ce:italic> of 0.12–0.14, and <ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic> of 0.47–0.51. For TP, ANN showed superior predictive skill (<ce:italic>NSE</ce:italic> = 0.49, <ce:italic>NMAE</ce:italic> = 0.13, <ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic> = 0.55), compared with <ce:italic>NSE</ce:italic> values of 0.17–0.46, <ce:italic>NMAE</ce:italic> of 0.13–0.18, and <ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic> of 0.32–0.51 for the remaining models. For DO, RF consistently outperformed all other approaches, achieving <ce:italic>NSE</ce:italic> = 0.76, <ce:italic>NMAE</ce:italic> = 0.09, and <ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic> = 0.77, whereas competing models produced <ce:italic>NSE</ce:italic> values of 0.62–0.72, <ce:italic>NMAE</ce:italic> of 0.10–0.11, and <ce:italic>R<ce:sup loc="post">2</ce:sup></ce:italic> of 0.68–0.73. Overall, all five models demonstrated their strongest performance for DO prediction. Permutation importance analysis revealed that surface TN, TP and water temperature were key predictors for hypolimnetic TN, TP, and DO, respectively.
{"title":"Linking Hypolimnion to Epilimnion in a Stratified Arctic Lake: Machine Learning-Based Estimation of Hypolimnetic Water Quality from Epilimnetic Measurements","authors":"Mehran Mahdian, Roohollah Noori, Mohammad Javad Saravani, Ali Reza Shahvaran, Mohsen Shahmohammad, Paul P.J. Gaffney, Mohammad Milad Salamattalab, Milad Shamsi Anboohi, Majid Hosseinzadeh, Fan Xia, Yongqiang Zhou, Yunlin Zhang, Mikko Kolehmainen, Soroush Abolfathi","doi":"10.1016/j.watres.2026.125367","DOIUrl":"https://doi.org/10.1016/j.watres.2026.125367","url":null,"abstract":"Understanding water quality in the deeper layers of stratified lakes is critical, as these zones govern ecosystem stability and biodiversity health. The hypolimnion, the bottommost layer of a stratified lake, plays a pivotal role by limiting vertical circulation and promoting pollutant accumulation, thus serving as an indicator of long-term lake condition. While surface water quality can now be routinely monitored using <ce:italic>in-situ</ce:italic> sensors and satellite observations, assessing hypolimnetic conditions remains costly and logistically challenging, particularly in the Arctic regions with severe environmental conditions and logistical constraints. Using a long-term dataset spanning 1979–2022 collected at a single monitoring station where both epilimnetic and hypolimnetic profiles are measured, this study develops and compares five machine and deep learning models including artificial neural networks (ANN), random forest (RF), extreme gradient boosting (XGBoost), support vector regression (SVR), and Kolmogorov–Arnold networks (KAN) to estimate hypolimnetic total nitrogen (TN), total phosphorus (TP), and dissolved oxygen (DO) in Lake Inari, Finland, based on readily available epilimnetic water quality predictors. Model performance was evaluated using five-fold cross-validation and assessed with the Nash–Sutcliffe efficiency (<ce:italic>NSE</ce:italic>), normalized mean absolute error (<ce:italic>NMAE</ce:italic>), and coefficient of determination (<ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic>). For TN, the RF model achieved the best overall performance, with mean cross-validation values of <ce:italic>NSE</ce:italic> = 0.52, <ce:italic>NMAE</ce:italic> = 0.11, and <ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic> = 0.52, outperforming the other models, which yielded <ce:italic>NSE</ce:italic> values of 0.38–0.52, <ce:italic>NMAE</ce:italic> of 0.12–0.14, and <ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic> of 0.47–0.51. For TP, ANN showed superior predictive skill (<ce:italic>NSE</ce:italic> = 0.49, <ce:italic>NMAE</ce:italic> = 0.13, <ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic> = 0.55), compared with <ce:italic>NSE</ce:italic> values of 0.17–0.46, <ce:italic>NMAE</ce:italic> of 0.13–0.18, and <ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic> of 0.32–0.51 for the remaining models. For DO, RF consistently outperformed all other approaches, achieving <ce:italic>NSE</ce:italic> = 0.76, <ce:italic>NMAE</ce:italic> = 0.09, and <ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic> = 0.77, whereas competing models produced <ce:italic>NSE</ce:italic> values of 0.62–0.72, <ce:italic>NMAE</ce:italic> of 0.10–0.11, and <ce:italic>R<ce:sup loc=\"post\">2</ce:sup></ce:italic> of 0.68–0.73. Overall, all five models demonstrated their strongest performance for DO prediction. Permutation importance analysis revealed that surface TN, TP and water temperature were key predictors for hypolimnetic TN, TP, and DO, respectively. ","PeriodicalId":443,"journal":{"name":"Water Research","volume":"56 1","pages":""},"PeriodicalIF":12.8,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-10DOI: 10.1016/j.watres.2026.125368
Chuanshun Zhi , Dahai Wang , Baonan He , Guohua Hou , Maosheng Gao , Hui Mu , Ruchun Wei , Xiancang Wu , Jing Bai , Yufei Jiao , Xiaonong Hu
Arsenic (As) mobilization in deltaic aquifers is regulated by tightly linked C–N–S–Fe–As biogeochemical processes, yet the influence of salinity on these interactions remains poorly resolved. Here, we investigated high-salinity groundwater from the Yellow River Delta, where total dissolved solids range from 1 to 35 g/L and As concentrations reach 303 μg/L. By integrating metagenomic sequencing, metagenome-assembled genomes (MAGs), and nitrogen and sulfur isotopic measurements, we characterized how salinity and redox gradients restructure microbial functional potential and regulate As cycling. Functional-gene profiles show a transition from nitrate- and Fe(III)-coupled metabolisms in low-salinity groundwater to sulfate- and sulfite-driven anaerobic pathways under high-salinity conditions, consistent with δ15N–NH4+, δ15N–NO3–, and δ34S–SO42– signatures. Genome-resolved analyses further reveal that Pseudomonadota and Desulfobacterota dominate carbon oxidation, nitrogen and iron reduction, and sulfur-intermediate reduction, while Muiribacteriota and Planctomycetota specialize in sulfite reduction and anammox/Feammox, respectively. Together, these results show that microbial communities reorganize along the salinity gradient, with arsenic mobilization associated with nitrogen- and iron-coupled reductive processes in low-salinity groundwater and with sulfur-driven reduction under high-salinity conditions. By linking microbial metabolic interactions to salinity-related redox environments, this study provides a process-based basis for anticipating how arsenic mobility may change as coastal aquifers undergo salinization.
{"title":"Metabolic coupling of arsenic, carbon, nitrogen, sulfur and iron in high-salinity groundwater in the Yellow River Delta: Insights from metagenomic analyses","authors":"Chuanshun Zhi , Dahai Wang , Baonan He , Guohua Hou , Maosheng Gao , Hui Mu , Ruchun Wei , Xiancang Wu , Jing Bai , Yufei Jiao , Xiaonong Hu","doi":"10.1016/j.watres.2026.125368","DOIUrl":"10.1016/j.watres.2026.125368","url":null,"abstract":"<div><div>Arsenic (As) mobilization in deltaic aquifers is regulated by tightly linked C–N–S–Fe–As biogeochemical processes, yet the influence of salinity on these interactions remains poorly resolved. Here, we investigated high-salinity groundwater from the Yellow River Delta, where total dissolved solids range from 1 to 35 g/L and As concentrations reach 303 μg/L. By integrating metagenomic sequencing, metagenome-assembled genomes (MAGs), and nitrogen and sulfur isotopic measurements, we characterized how salinity and redox gradients restructure microbial functional potential and regulate As cycling. Functional-gene profiles show a transition from nitrate- and Fe(III)-coupled metabolisms in low-salinity groundwater to sulfate- and sulfite-driven anaerobic pathways under high-salinity conditions, consistent with δ<sup>15</sup>N–NH<sub>4</sub><sup>+</sup>, δ<sup>15</sup>N–NO<sub>3</sub><sup>–</sup>, and δ<sup>34</sup>S–SO<sub>4</sub><sup>2–</sup> signatures. Genome-resolved analyses further reveal that Pseudomonadota and Desulfobacterota dominate carbon oxidation, nitrogen and iron reduction, and sulfur-intermediate reduction, while Muiribacteriota and Planctomycetota specialize in sulfite reduction and anammox/Feammox, respectively. Together, these results show that microbial communities reorganize along the salinity gradient, with arsenic mobilization associated with nitrogen- and iron-coupled reductive processes in low-salinity groundwater and with sulfur-driven reduction under high-salinity conditions. By linking microbial metabolic interactions to salinity-related redox environments, this study provides a process-based basis for anticipating how arsenic mobility may change as coastal aquifers undergo salinization.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125368"},"PeriodicalIF":12.4,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145949960","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-09DOI: 10.1016/j.watres.2025.125313
Le Min Chen, Sunanda Keisham, Hiroaki Tateno, Gijs Y. Kleine, Martin Pabst, Mario Pronk, Mark C.M. van Loosdrecht, Yuemei Lin
The long-term effects of environmental conditions, such as seawater salinity, on the extracellular investigated EPS changes during a stepwise increase in salinity (0–4%), renewing over 90% of biomass at each condition. Stable granulation, complete anaerobic acetate uptake, and phosphate removal were maintained throughout. FT-IR of granules showed significant changes in glycans (1025 cm⁻¹) and sialic acid (1730 cm⁻¹), which were reflected in the EPS. Lectin microarray revealed that increasing salinity reduced glycan diversity in EPS glycoproteins, while increasing negatively charged groups, including sialic acids and sulfated groups. At 4% salinity, EPS negative charge increased by 19.8% compared to 0%. Microbial community composition shifted from a diverse mix (Dechloromonas; 23%, “Candidatus Competibacter”; 13%, “Candidatus Accumulibacter”; 28%) at 0% to a dominant (69% – 75%) unclassified Accumulibacter clade I species at 1 - 4% salinity. Metaproteomic analysis showed strong upregulation of genes of “Ca. Accumulibacter” involved in monosaccharide, lipopolysaccharide, and peptidoglycan biosynthesis from 3% - 4% salinity, indicating its adaptation to salinity stress. Dechloromonas and “Ca. Competibacter” represented a minor or a non-significant fraction of those proteins related to glycan synthesis across the salinities. Despite that no glycoprotein biosynthesis pathways were identified in the metaproteomic data, three putative glycoproteins produced by “Ca. Accumulibacter” were detected across all conditions. They were downregulated as the salinity increased. These findings highlight how “Ca. Accumulibacter” dynamically adapts its EPS, particularly glycoprotein glycans, in response to increasing salinity, offering new insights into EPS adaptation under environmental stress.
{"title":"Extracellular Polymeric Substances in Aerobic Granular Sludge Under Increasing Salinity Conditions","authors":"Le Min Chen, Sunanda Keisham, Hiroaki Tateno, Gijs Y. Kleine, Martin Pabst, Mario Pronk, Mark C.M. van Loosdrecht, Yuemei Lin","doi":"10.1016/j.watres.2025.125313","DOIUrl":"https://doi.org/10.1016/j.watres.2025.125313","url":null,"abstract":"The long-term effects of environmental conditions, such as seawater salinity, on the extracellular investigated EPS changes during a stepwise increase in salinity (0–4%), renewing over 90% of biomass at each condition. Stable granulation, complete anaerobic acetate uptake, and phosphate removal were maintained throughout. FT-IR of granules showed significant changes in glycans (1025 cm⁻¹) and sialic acid (1730 cm⁻¹), which were reflected in the EPS. Lectin microarray revealed that increasing salinity reduced glycan diversity in EPS glycoproteins, while increasing negatively charged groups, including sialic acids and sulfated groups. At 4% salinity, EPS negative charge increased by 19.8% compared to 0%. Microbial community composition shifted from a diverse mix (Dechloromonas; 23%, “<ce:italic>Candidatus</ce:italic> Competibacter”; 13%, “<ce:italic>Candidatus</ce:italic> Accumulibacter”; 28%) at 0% to a dominant (69% – 75%) unclassified Accumulibacter clade I species at 1 - 4% salinity. Metaproteomic analysis showed strong upregulation of genes of “<ce:italic>Ca.</ce:italic> Accumulibacter” involved in monosaccharide, lipopolysaccharide, and peptidoglycan biosynthesis from 3% - 4% salinity, indicating its adaptation to salinity stress. Dechloromonas and <ce:italic>“Ca.</ce:italic> Competibacter” represented a minor or a non-significant fraction of those proteins related to glycan synthesis across the salinities. Despite that no glycoprotein biosynthesis pathways were identified in the metaproteomic data, three putative glycoproteins produced by <ce:italic>“Ca.</ce:italic> Accumulibacter” were detected across all conditions. They were downregulated as the salinity increased. These findings highlight how “Ca. Accumulibacter” dynamically adapts its EPS, particularly glycoprotein glycans, in response to increasing salinity, offering new insights into EPS adaptation under environmental stress.","PeriodicalId":443,"journal":{"name":"Water Research","volume":"26 1","pages":""},"PeriodicalIF":12.8,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-09DOI: 10.1016/j.watres.2026.125360
Bin Ma , Xiaoning Zhang , Zhang Wen , Lili Liang , Menggui Jin
Ammonium (NH₄⁺) and nitrous oxide (N₂O) are key reactive nitrogen species in aquatic systems, with implications for water quality and greenhouse gas emissions. However, their sources and transformation pathways in shallow groundwater remain poorly constrained, especially for dynamic redox groundwater in hill-plain transition zones. To address this gap, we investigated nitrogen sources and cycling in the piedmont hilly–plain zone of the Jianghan Plain (central China) by integrating hydrochemistry, nitrogen isotopes (δ¹⁵NNH₄⁺, δ¹⁵NNO₃⁻, δ¹⁸ONO₃⁻), and N₂O isotopomers (δ¹⁵N, δ¹⁸O, and site preference). A total of 56 water samples were collected during pre- and post-monsoon periods from rivers, lakes, and shallow aquifers. Previously unrecognized spatial patterns and underlying mechanisms controlling nitrate, ammonium, and N₂O along the groundwater flow path from hilly recharge zones to plain discharge areas are identified. δ²H–δ¹⁸O signatures showed that groundwater primarily derives from precipitation, with stronger evaporation effects in surface water. Dissolved N₂O concentrations were significantly higher in groundwater than in surface water, suggesting active subsurface nitrogen cycling. N₂O site preference (SP) and δ¹⁸O–N₂O values revealed coexistence of nitrification and denitrification, with enhanced N₂O reduction in plain aquifers exhibiting more reducing conditions. We quantitatively characterized the distinct N₂O production pathways across the hilly–plain transition zone, revealing the spatial variability in the contributions of nitrification and denitrification to N₂O generation. These results demonstrate that coupling hydro-geochemistry with N₂O isotopomer analysis provides a powerful approach for resolving nitrogen sources and transformation pathways in hydrogeologically complex zones. The findings offer critical insights for managing nitrogen pollution in rapidly developing agricultural plains.
{"title":"Ammonium and N₂O production pathways in quaternary hill–plain groundwater: Evidence from multi-isotope and isotopomer analysis","authors":"Bin Ma , Xiaoning Zhang , Zhang Wen , Lili Liang , Menggui Jin","doi":"10.1016/j.watres.2026.125360","DOIUrl":"10.1016/j.watres.2026.125360","url":null,"abstract":"<div><div>Ammonium (NH₄⁺) and nitrous oxide (N₂O) are key reactive nitrogen species in aquatic systems, with implications for water quality and greenhouse gas emissions. However, their sources and transformation pathways in shallow groundwater remain poorly constrained, especially for dynamic redox groundwater in hill-plain transition zones. To address this gap, we investigated nitrogen sources and cycling in the piedmont hilly–plain zone of the Jianghan Plain (central China) by integrating hydrochemistry, nitrogen isotopes (δ¹⁵N<img>NH₄⁺, δ¹⁵N<img>NO₃⁻, δ¹⁸O<img>NO₃⁻), and N₂O isotopomers (δ¹⁵N, δ¹⁸O, and site preference). A total of 56 water samples were collected during pre- and post-monsoon periods from rivers, lakes, and shallow aquifers. Previously unrecognized spatial patterns and underlying mechanisms controlling nitrate, ammonium, and N₂O along the groundwater flow path from hilly recharge zones to plain discharge areas are identified. δ²H–δ¹⁸O signatures showed that groundwater primarily derives from precipitation, with stronger evaporation effects in surface water. Dissolved N₂O concentrations were significantly higher in groundwater than in surface water, suggesting active subsurface nitrogen cycling. N₂O site preference (SP) and δ¹⁸O–N₂O values revealed coexistence of nitrification and denitrification, with enhanced N₂O reduction in plain aquifers exhibiting more reducing conditions. We quantitatively characterized the distinct N₂O production pathways across the hilly–plain transition zone, revealing the spatial variability in the contributions of nitrification and denitrification to N₂O generation. These results demonstrate that coupling hydro-geochemistry with N₂O isotopomer analysis provides a powerful approach for resolving nitrogen sources and transformation pathways in hydrogeologically complex zones. The findings offer critical insights for managing nitrogen pollution in rapidly developing agricultural plains.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125360"},"PeriodicalIF":12.4,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-09DOI: 10.1016/j.watres.2026.125361
Zhendan Wang , Dunxian She , Shaoda Liu , Xiaoyu Zhang , Zhihong Song , Jun Xia
River networks release significant amounts of carbon dioxide (CO2) into the atmosphere via gas exchanges at water-air interface, profoundly affecting the global carbon cycle. Hydrological drought, characterized by negative anomalies in river discharge, can affect the transport and decomposition of organic matter, thereby significantly impacting riverine CO2 emissions. However, the extent to which hydrological droughts affect CO2 emission fluxes remains unanswered. In this study, we developed a framework to quantify the impact of hydrological droughts on CO2 emissions from the Yangtze River networks. We found that hydrological droughts led to approximately 33 % reduction in CO2 evasions in Yangtze River networks compared with non-drought periods. Moreover, the reduction in CO2 emissions across all stream orders of rivers showed significantly positive correlations to drought severity (p < 0.001). Notably, the emission reduction primarily resulted from river width contraction, which diminished the water-air interface area and consequently limited CO2 evasion. These findings highlight the importance of deepening our understanding of the impact of hydrological droughts on riverine CO2 emissions.
{"title":"Hydrological droughts reduce carbon dioxide emission from the Yangtze River networks","authors":"Zhendan Wang , Dunxian She , Shaoda Liu , Xiaoyu Zhang , Zhihong Song , Jun Xia","doi":"10.1016/j.watres.2026.125361","DOIUrl":"10.1016/j.watres.2026.125361","url":null,"abstract":"<div><div>River networks release significant amounts of carbon dioxide (CO<sub>2</sub>) into the atmosphere via gas exchanges at water-air interface, profoundly affecting the global carbon cycle. Hydrological drought, characterized by negative anomalies in river discharge, can affect the transport and decomposition of organic matter, thereby significantly impacting riverine CO<sub>2</sub> emissions. However, the extent to which hydrological droughts affect CO<sub>2</sub> emission fluxes remains unanswered. In this study, we developed a framework to quantify the impact of hydrological droughts on CO<sub>2</sub> emissions from the Yangtze River networks. We found that hydrological droughts led to approximately 33 % reduction in CO<sub>2</sub> evasions in Yangtze River networks compared with non-drought periods. Moreover, the reduction in CO<sub>2</sub> emissions across all stream orders of rivers showed significantly positive correlations to drought severity (<em>p</em> < 0.001). Notably, the emission reduction primarily resulted from river width contraction, which diminished the water-air interface area and consequently limited CO<sub>2</sub> evasion. These findings highlight the importance of deepening our understanding of the impact of hydrological droughts on riverine CO<sub>2</sub> emissions.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125361"},"PeriodicalIF":12.4,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-09DOI: 10.1016/j.watres.2026.125358
Kexin Li , Zhendong Lei , Yujia Zhang , Jinling He , Ming Zheng , Ai Zhang , Pin Gao , Xiang Li , Gang Xue
Anaerobic fermentation of waste activated sludge (WAS) into short-chain fatty acids (SCFAs) is a promising valorization strategy, yet it is often constrained by inefficient hydrolysis and the presence of contaminants. This study reports a synergistic electrochemical activation coupling a boron-doped diamond anode (BDD) with peroxymonosulfate (PMS), i.e., B/P, which enhances SCFA production and pollutant degradation. The integrated B/P system achieved a maximal SCFA yield, which was 2.0-fold higher than the control and significantly surpassed mono-pretreatments (BDD, PMS) and their sequential combination (B + P). Furthermore, it optimized product composition, increasing the proportion of high-value acetic acid by 2.7-fold. Mechanistic investigations revealed that the electrochemical synergy generated reactive species (e.g., •OH, SO4•⁻, 1O2), which profoundly accelerated sludge solubilization (supported by 10.7 times of SCOD produced by B/P relative to the sum of SCOD by mono-BDD and mono-PMS). Microbial analysis demonstrated that B/P pretreatment restructured the community to enrich key hydrolytic and acidogenic phyla. This was corroborated by the upregulation of genes associated with carbohydrate hydrolysis and acidogenesis, alongside the suppression of methanogenesis. Beyond SCFA enhancement, the B/P system also improved sludge dewaterability (24 % reduction in capillary suction time), efficient removal of sulfamethoxazole (62.7 %), and substantially transferred heavy metals from the sludge solids into the liquid phase (e.g., 71.8 % of total Cu was transferred from the solid phase to the supernatant). These findings highlight the B/P process as a multifunctional pretreatment for efficient sludge bioconversion, pollutant removal, and environmental risk mitigation.
{"title":"Electrochemical activation of peroxymonosulfate with a boron-doped diamond anode boosts short-chain fatty acid production from waste activated sludge","authors":"Kexin Li , Zhendong Lei , Yujia Zhang , Jinling He , Ming Zheng , Ai Zhang , Pin Gao , Xiang Li , Gang Xue","doi":"10.1016/j.watres.2026.125358","DOIUrl":"10.1016/j.watres.2026.125358","url":null,"abstract":"<div><div>Anaerobic fermentation of waste activated sludge (WAS) into short-chain fatty acids (SCFAs) is a promising valorization strategy, yet it is often constrained by inefficient hydrolysis and the presence of contaminants. This study reports a synergistic electrochemical activation coupling a boron-doped diamond anode (BDD) with peroxymonosulfate (PMS), i.e., B/P, which enhances SCFA production and pollutant degradation. The integrated B/P system achieved a maximal SCFA yield, which was 2.0-fold higher than the control and significantly surpassed mono-pretreatments (BDD, PMS) and their sequential combination (B + P). Furthermore, it optimized product composition, increasing the proportion of high-value acetic acid by 2.7-fold. Mechanistic investigations revealed that the electrochemical synergy generated reactive species (e.g., •OH, SO<sub>4</sub>•⁻, <sup>1</sup>O<sub>2</sub>), which profoundly accelerated sludge solubilization (supported by 10.7 times of SCOD produced by B/P relative to the sum of SCOD by mono-BDD and mono-PMS). Microbial analysis demonstrated that B/P pretreatment restructured the community to enrich key hydrolytic and acidogenic phyla. This was corroborated by the upregulation of genes associated with carbohydrate hydrolysis and acidogenesis, alongside the suppression of methanogenesis. Beyond SCFA enhancement, the B/P system also improved sludge dewaterability (24 % reduction in capillary suction time), efficient removal of sulfamethoxazole (62.7 %), and substantially transferred heavy metals from the sludge solids into the liquid phase (e.g., 71.8 % of total Cu was transferred from the solid phase to the supernatant). These findings highlight the B/P process as a multifunctional pretreatment for efficient sludge bioconversion, pollutant removal, and environmental risk mitigation.</div></div>","PeriodicalId":443,"journal":{"name":"Water Research","volume":"292 ","pages":"Article 125358"},"PeriodicalIF":12.4,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"环境科学与生态学","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