 |
| Titel |
Modelling salt finger formation using the Imperial College Ocean Model |
| VerfasserIn |
F. P. MacTavish, C. J. Cotter, M. D. Piggott |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250027445
|
|
|
|
|
|
| Zusammenfassung |
| We present numerical simulations of salt finger formation produced
using the Imperial College Ocean Model (ICOM) which is a finite
element model using adaptive meshing. Our aim is to validate the model
against published data and to develop the capability to simulate salt
finger formation using adaptive meshes.
Salt fingering is a form of double-diffusion which occurs because heat
diffuses more quickly than salt. When an area of warm, salty
water overlies an area of colder, fresher water, an initial
perturbation can lead to some of the water from the lower layer moving
into the top layer. Its temperature then increases more quickly than
its salinity, so that the water is less dense than its surroundings
and it will rise up more. This process repeats to form salt fingers,
with salt fingers also forming in the downward direction.
Salt fingers play a role in oceanic mixing, in particular they are
responsible for maintaining thermohaline staircases such as the C-SALT
staircase which have been observed extensively, particularly in the
tropics. The study of salt fingers could therefore improve our
understanding of processes in the ocean, and inform the design of
subgrid parameterisations in general circulation models.
We used the salt finger formation test case of Oezgoekmen et al (1998)
in order to validate ICOM. The formation of salt fingers is modelled by
solving the Navier-Stokes equations for a two-dimensional rectangular
area of Boussinesq fluid, beginning with two layers of water, the top
warm and salty and the bottom cold and fresh, with parameters chosen
to match the test case of Oezgoekmen et al (1998).
The positions of the interfaces between the fingering layer and the
mixed layers as well as the finger growth rate and the kinetic energy
are plotted against time. The results are compared with those of Oezgoekmen et al
(1998). We present results from structured meshes and preliminary
results using adaptive meshing. |
|
|
|
|
|
|