Lagrangian Observations and Modeling of Turbulence Along a Tidally Influenced River. Issue 1 (10th January 2022)

• Record Type: Journal Article
• Authors: Déjeans, Bérengère S.; Mullarney, Julia C.; MacDonald, Iain T.
• Is Part Of: Water resources research. Volume 58:Issue 1(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-01-10
• Subjects: turbulence modeling -- Lagrangian -- river; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2020WR027894
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: We examine and model the patterns of velocities and turbulence from flow‐following measurements along a tidally influenced river, collected from the heavily sediment‐laden Kaipara River, New Zealand, using Lagrangian 'Floc Drifter' platforms released at different locations and times. Numerical modeling was undertaken in Delft3D with the k − ϵ turbulence closure scheme. While model calibration using Eulerian measurements was classified as excellent, results explored in a Lagrangian framework revealed deficiencies in model performance. Generally, the model reproduced flow speeds and patterns of dissipation rates of turbulent kinetic energy ϵ of the right order of magnitude, but did not always correctly reproduce the observed ϵ, particularly around abrupt meander bends. These discrepancies were attributed partly to errors in velocity predictions, but also indicated that other processes (omitted or not accurately represented in our model) such as wind‐driven mixing and secondary flow strongly influenced turbulence dynamics. Predicted ϵ were relatively consistent across different tidal phases and days, suggesting the bathymetry and geometry of the river were the greatest control on the along‐river structure of turbulence. The vertical and cross‐sectional distributions of turbulent quantities ( ϵ, k and ν ) displayed larger variations vertically than longitudinally and agree with previous studies. Results show that flow‐following measurements offer a stringent validation for numerical modeling of hydrodynamics over large (riverine) spatial scales at a temporal and spatial resolution not possible with traditional Eulerian measurements. Ultimately, robust numerical predictions of velocities and turbulence are critical to accurately model and predict the dispersal and fate of fine‐sediments in aquatic environments. Key Points: Lagrangian measurements offer stringent validation for numerical modeling of riverine hydrodynamics not possible with Eulerian observations Flow‐following model predictions of turbulence are reasonable along the river; however, comparisons with observations are worse at bends Bathymetry and geometry of the river are suggested as the main control on the dissipation rate of turbulent kinetic energy