Modelling of Meddy-topography interactions
In the
description of the research project
the following two main components of the research work are mentioned:
(i) application of a numerical code, and
(ii) a series of laboratory experiments.
Here follows the text on these components as written in the
project proposal.
(i) Numerical modelling
The numerical modelling of the influence of topography on Meddies
will be done with a code that solves the two-dimensional (2D) viscous
Navier-Stokes equation in the vorticity-streamfunction formulation
using a finite difference method,
i.e. using a grid in a certain domain with appropriate boundary
conditions.
The code, originating from a code developed by Orlandi and Verzicco
(Rome), has been used successfully in simulating the evolution
of several kinds of 2D vortices (monopoles, dipoles, ...)
and the results compare well with those of laboratory experiments
and those of other numerical methods.
The code has proven to be very stable, even for inviscid
flows, depending only on the size of the time step with respect to
the size of the grid.
Extending the existing code by implementing a topographic function
can be done quite easily.
Due to the use of a finite difference method only more or less
smooth topographies will be possible, such as Gaussian-shaped seamounts,
smooth ridge-like features or sloping bottoms.
This will nevertheless lead to a better understanding of the effect
of topography on the motion and evolution of Meddy-like monopolar
vortices.
The required smoothness of the topography that can be used is mainly
determined by the resolution, i.e. the grid size, needed to show
the effects being studied.
There are a few additional features in the code and separate
programs available to follow specific flow characteristic
(such as the energy, enstrophy and circulation of the vortex)
in the course of time, as well as the motion of passive tracer
particles advected by the flow field.
(ii) Laboratory experiments
In the laboratory experiments, a rotating tow-tank channel facility
filled with a stratified fluid will be used to study the interactions.
Such a facility is fitted with a belt arrangement that permits topographic
features to be translated uniformly and horizontally at prescribed
velocities along the base of the tank.
In the present project, a model Meddy will be generated in the tank,
using the technique developed by Carton (1994), and a topographic feature
mounted on the movable belt will be translated towards and past the Meddy.
The interaction between the Meddy and the topography will be monitored
by the flow visualisation and direct density field measurement systems
available in the host laboratory.
Topographic elements matching the numerical simulations will be used,
in order to make direct comparisons between the two parts of the study
possible.
Quantitative measurements of the velocity and density fields within the
Meddy throughout its interaction with the topography will be made using
the data acquisition system in Davies' laboratory.
With such measurements, the nature of the topographic distortion of the
density, velocity and vorticity fields within the Meddy by the
topography can be determined.
In addition, the small-scale mixing caused to the Meddy can
be quantified and parameterised.
The experiments will be designed to cover cases of direct relevance to
oceanic counterparts and will investigate the topographic
interactions for a range of external parameters (e.g. undisturbed
Meddy vorticity, advection velocity of the Meddy, topographic height,
background stratification, Coriolis parameter).
Systematic parametric experiments will be carried out to study strong
and weak interactions and attempts will be made to classify the different
interactions that occur.
The main outputs of the work will therefore be a series of flow
classifications, sets of three dimensional and time-dependent
velocity, vorticity and density fields for Meddy-topography
interactions of varying strength and intercomparisons between laboratory
model data, numerical model results and oceanic field data where available.
Back to the Post-doc. research in Dundee page.
Jos van Geffen --
Home |
Site Map |
Contact Me
last modified: 27 May 2001