{"title":"Understanding Flow Transport in Dual-Water-Source Rivers: Enhancing Storage Selection Functions With Two-Gamma Distributions","authors":"Anran Liao, Tiantian Zhou, Xianfang Song, Lihu Yang, Dongmei Han, Shengtian Yang","doi":"10.1002/hyp.70121","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>Modelling solute transport in river systems using the Storage Selection (SAS) function presents challenges when solute recharge originates from two distinct water sources: a point reclaimed-water source (perpendicular to gravity) and a surface rainfall source (parallel to gravity). This study introduces three modifications to the original SAS function to address dual-source transport dynamics in a reclaimed-water-source river reach: (i) incorporating rainwater into the reclaimed water flux (<i>R</i>), (ii) introducing a mixing process to represent interactions between rainwater and reclaimed water (Mix), and (iii) modifying the original SAS function into a two-gamma-distribution (2GM) model to better capture dual-source transport behaviour. The effects of these modifications were evaluated individually and in combination. Among the modifications, the 2GM model achieved the highest increase in the Nash-Sutcliffe efficiency coefficient (NSE), followed by R and then Mix. The 2GM model regulated storage volume, thereby influencing the outflow δ<sup>2</sup><i>H</i> value (i.e., the objective function), making it the most effective modification. Combining all modifications produced the best model performance, improving NSE from 0.61 to 0.86. However, short-term δ<sup>2</sup><i>H</i> fluctuations due to river morphology-induced stochastic processes were not captured by this lumped transport model. The optimal model indicated that storm events significantly accelerated river water (mixed reclaimed water and rainwater) mobility. One day after rainfall, 53% of the water at the outlet originated from event water, either from adjacent sites or nearby subsurface flow. Subsequent rainwater and reclaimed water flowed out gradually at a very slow rate. River storage received sustained event water throughout the wet period, with most event water stagnating in storage, except for outflow occurring 1 day post-rainfall. Special attention should be given to pollutants carried by event water and the associated aquatic risks during wet periods in megacities. This study provides a reference for flood management and pollution control.</p>\n </div>","PeriodicalId":13189,"journal":{"name":"Hydrological Processes","volume":"39 4","pages":""},"PeriodicalIF":3.2000,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Hydrological Processes","FirstCategoryId":"89","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/hyp.70121","RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Environmental Science","Score":null,"Total":0}
引用次数: 0
Abstract
Modelling solute transport in river systems using the Storage Selection (SAS) function presents challenges when solute recharge originates from two distinct water sources: a point reclaimed-water source (perpendicular to gravity) and a surface rainfall source (parallel to gravity). This study introduces three modifications to the original SAS function to address dual-source transport dynamics in a reclaimed-water-source river reach: (i) incorporating rainwater into the reclaimed water flux (R), (ii) introducing a mixing process to represent interactions between rainwater and reclaimed water (Mix), and (iii) modifying the original SAS function into a two-gamma-distribution (2GM) model to better capture dual-source transport behaviour. The effects of these modifications were evaluated individually and in combination. Among the modifications, the 2GM model achieved the highest increase in the Nash-Sutcliffe efficiency coefficient (NSE), followed by R and then Mix. The 2GM model regulated storage volume, thereby influencing the outflow δ2H value (i.e., the objective function), making it the most effective modification. Combining all modifications produced the best model performance, improving NSE from 0.61 to 0.86. However, short-term δ2H fluctuations due to river morphology-induced stochastic processes were not captured by this lumped transport model. The optimal model indicated that storm events significantly accelerated river water (mixed reclaimed water and rainwater) mobility. One day after rainfall, 53% of the water at the outlet originated from event water, either from adjacent sites or nearby subsurface flow. Subsequent rainwater and reclaimed water flowed out gradually at a very slow rate. River storage received sustained event water throughout the wet period, with most event water stagnating in storage, except for outflow occurring 1 day post-rainfall. Special attention should be given to pollutants carried by event water and the associated aquatic risks during wet periods in megacities. This study provides a reference for flood management and pollution control.
期刊介绍:
Hydrological Processes is an international journal that publishes original scientific papers advancing understanding of the mechanisms underlying the movement and storage of water in the environment, and the interaction of water with geological, biogeochemical, atmospheric and ecological systems. Not all papers related to water resources are appropriate for submission to this journal; rather we seek papers that clearly articulate the role(s) of hydrological processes.