DATE:

Friday, June 23, 2023

TIME:

3:30 PM

LOCATION:

Zoom

SPEAKER:

Jared Brzenski, PhD Candidate, Computational Science, San Diego State University/UC, Irvine

ABSTRACT:

Nearshore bathymetric features like steep canyons have been shown to drive nutrient mixing and create diverse hotbeds of bio-activity. However, these features are regularly left off large models because of difficulties resolving their sharp domain features or are modeled with such low fidelity to make the canyon appear nonexistent.

This project aims to employ a non-hydrostatic Boussinesq 3D curvilinear large-eddy simulator to generate accurate predictions of momentum and transport in coastal ocean areas where internal waves are caused due to sharp bathymetric features. The simulator code is crafted so that all processes, from internal dynamics to the transport of constituents, are accurately simulated. The goal is to have the fastest, most accurate calculations capable of resolving features on a curvilinear terrain-following grid and can do so in the same computational time frame as larger regional ocean models. The code for this project solves the non-hydrostatic incompressible Navier-Stokes solver using the Boussinesq approximations on 3D grids. The code is implemented in C++ and uses mimetic operators for the divergence, gradient, and Laplacian operators. It has been shown to accurately capture the physics of some primary test cases but is missing key features to make it an actual coastal ocean model.

HOST:

SDSU’s SIAM Student Chapter

VIDEO: