Tritium plume observations in Lake Huron: measurements of dispersion and comparison of lateral plume structure with the normal distribution of the Gaussian Plume Model
{"title":"Tritium plume observations in Lake Huron: measurements of dispersion and comparison of lateral plume structure with the normal distribution of the Gaussian Plume Model","authors":"Volodymyr Korolevych, Dan Festarini","doi":"10.1016/j.envc.2025.101271","DOIUrl":null,"url":null,"abstract":"<div><div>This study describes observations of the near-shore transport and dispersion of routine but intermittent tritium emissions from the Bruce Nuclear Generating Station into Lake Huron, which are used as a tracer for inferring dispersion parameters. Study region covers 45 km along the shore northward, 38 km southward, and 18 km off the shore. Currents were recorded by GPS drifters deployed in the study region, and tritium concentrations in the effluent plume were collected both along drifter trajectories and across them. Two episodes were analysed: one with the plume going into the headwind and one with a regular wind-driven plume. <em>In situ</em> dispersion was calculated from tritium plume observations. The highest concentrations encountered at each distance from the source were combined to locate the plume centreline and deduce scaling (dependence on the distance to the source) according to a power-law fit to these concentrations. Lateral distribution of concentration in the plume was Gaussian Plume Model parameterised with this <em>in situ</em> dispersion and then applied to remaining sampling locations, that is, to all off-centreline observations to model predictions.</div><div>The headwind episode yielded a scaling exponent of dispersion <em>α</em> = 0.711 and magnitude S<sub>y</sub> = 11.3, and the episode of a regular wind-driven plume yielded <em>α</em> = 0.411 and S<sub>y</sub> = 25.9. Gaussian Plume Model from SRS-19 was deployed with these parameters. Across the analyzed region model predictions were characterized by R<sup>2</sup> = 0.34 and Pearson <strong><em>r</em></strong> = 0.36 during the headwind episode and R<sup>2</sup> = 0.72 with <strong><em>r</em></strong> = 0.71 during the regular wind-driven episode.</div></div>","PeriodicalId":34794,"journal":{"name":"Environmental Challenges","volume":"20 ","pages":"Article 101271"},"PeriodicalIF":0.0000,"publicationDate":"2025-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Environmental Challenges","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2667010025001908","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Environmental Science","Score":null,"Total":0}
引用次数: 0
Abstract
This study describes observations of the near-shore transport and dispersion of routine but intermittent tritium emissions from the Bruce Nuclear Generating Station into Lake Huron, which are used as a tracer for inferring dispersion parameters. Study region covers 45 km along the shore northward, 38 km southward, and 18 km off the shore. Currents were recorded by GPS drifters deployed in the study region, and tritium concentrations in the effluent plume were collected both along drifter trajectories and across them. Two episodes were analysed: one with the plume going into the headwind and one with a regular wind-driven plume. In situ dispersion was calculated from tritium plume observations. The highest concentrations encountered at each distance from the source were combined to locate the plume centreline and deduce scaling (dependence on the distance to the source) according to a power-law fit to these concentrations. Lateral distribution of concentration in the plume was Gaussian Plume Model parameterised with this in situ dispersion and then applied to remaining sampling locations, that is, to all off-centreline observations to model predictions.
The headwind episode yielded a scaling exponent of dispersion α = 0.711 and magnitude Sy = 11.3, and the episode of a regular wind-driven plume yielded α = 0.411 and Sy = 25.9. Gaussian Plume Model from SRS-19 was deployed with these parameters. Across the analyzed region model predictions were characterized by R2 = 0.34 and Pearson r = 0.36 during the headwind episode and R2 = 0.72 with r = 0.71 during the regular wind-driven episode.