Unit:
Κατεύθυνση Φυσική ΩκεανογραφίαLibrary of the School of Science
Author:
Bolioudakis Tilemachos
Supervisors info:
Σαράντης Σοφιανός, Αναπληρωτής Καθηγητής, Τμήμα Φυσικής, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών.
Original Title:
Implicit large eddy simulation of a model ocean surface stratified turbulent Ekman layer
Translated title:
Implicit large eddy simulation of a model ocean surface stratified turbulent Ekman layer
Summary:
In the present study, through use of parallel high-resolution/accuracy im- plicit Large-Eddy Simulation (LES), we examine the development of turbu- lence and its impact on the vertical structure of background stratification and sheared currents, which have been established through interaction with rotation in the oceanic near-surface zone. Inspiration for the problem set- up has originated from large-scale regional ocean model results and the associated turbulent mixing parameterization. The simulations rely on a high-accuracy hybrid modal-spectral-element/Fourier method incompress- ible Navier-Stokes flow solver. One of our main goals is to validate this code on a new HPC platform, and, to accurately reproduce and initialize a theoret- ical Ekman spiral projected on a 3-dimensional domain. Through a selective exploration of parameter space, simulations investigate the impact of stratifi- cation on a complex Ekman-spiral-driven turbulent velocity field, including both qualitative and quantitative metrics. The physical time of each simu- lation is 40 minutes. The computational domain consists of a horizontally- periodic cubic box characterized by 2pm length in every direction. Analysis of results focuses on comparing scale-dependent turbulent processes of one unstratified and one stratified case, based on flow structure visualization and on monitoring the associated kinetic energy evolution. Finally, a discussion on extrapolating the findings to the oceanic scales is conducted.
Main subject category:
Science
Keywords:
Large Eddy Simulation, Turbulence, Stratified flow, Oceanic surface Ekman layer,
