Graduation Year

2025

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Andres Tejada-Martinez, Ph.D.

Co-Major Professor

Boris Galperin, Ph.D.

Committee Member

Mauricio E. Arias, Ph.D.

Committee Member

Mahmood Nachabe, Ph.D.

Committee Member

Wenbin Mao, Ph.D.

Keywords

Coastal Ocean, Computational Fluid Dynamics, Reynolds Averaged Navier Stokes, Tidal Forcing, Turbulence

Abstract

Langmuir Circulation (LC) is a wind and wave-driven process and is a key driver of vertical mixing and transport in the ocean. In coastal shelf regions, LC can span the entire water column, linking surface and bottom boundary layers and amplifying turbulent mixing. However, the influence of LC on vertical mixing and cross-shelf transport in shallow, tidally forced coastal waters remains poorly understood. This dissertation addresses this knowledge gap by developing a novel modeling framework based on the Reynolds-Averaged Navier–Stokes (RANS) equations, offering a computationally efficient alternative to more resource-intensive Large Eddy Simulations (LES).

The proposed RANS-based approach is first validated and then applied to explicitly resolve the interactions among waves, currents, and tides, providing the clearest demonstration to date of their combined impact on cross-shelf dynamics. Results show that the interaction between LC and tidal currents introduces significant complexity to flow structures, affecting momentum exchange, vertical mixing, and transport. Simulations reveal that neglecting either LC or tidal forcing results in substantial inaccuracies in predicted velocity profiles and turbulent behavior, highlighting the need to resolve both processes in inner-shelf models.

In the second phase of this work, the RANS framework is extended to better represent the temporal variability of tidal flows. Existing LES-based models typically simulate tides by imposing a constant pressure gradient; however, this approach can result in discrepancies between simulated and field-measured currents, particularly in capturing the correct vertical structure of tidal velocity. Moreover, while LES could theoretically accommodate time-dependent pressure gradients to better resolve tidal variability, such implementations are not seen in practice due to the significant computational cost involved. To address these limitations, a more efficient novel approach is implemented by prescribing time-dependent oscillating velocity profiles that capture tidal variability more realistically. Using this method, the model investigates the strength and evolution of LC—quantified via vertical velocity variance—under varying wind, wave, and tidal conditions. Unlike conventional approaches that rely on steady forcing, this technique enables direct simulation of LC–tide interactions over tidal time scales, offering insights into dynamic turbulence modulation across tidal phases.

Ultimately, the methodology presented lays the foundation for developing scaling laws and parameterizations of LC-driven mixing suitable for integration into larger-scale ocean models. These models, essential for simulating ocean–climate interactions, are typically constrained by coarse spatial and temporal resolutions. By providing physically based and computationally efficient parameterizations of LC–tide interactions, the proposed framework enhances the predictive skill of regional and global circulation models. This contributes to a broader effort to balance model accuracy with computational feasibility in oceanographic simulations, improving our capacity to simulate and manage coastal and climate systems.

Download

Share

COinS