HR: 13:40h
AN: OS32C-01 [PDF]
TI: A Generic Approach to Dynamical Interpretation of Geophysical Fluid Flow Processes
AU: * Liang, X
EM: liang@deas.harvard.edu
AF: Harvard University, Pierce Hall G2H
29 Oxford St, Cambridge, MA 02143 United States
AU: Robinson, A R
EM: robinson@pacific.deas.harvard.edu
AF: Harvard University, Pierce Hall G2H
29 Oxford St, Cambridge, MA 02143 United States
AB:
A real problem-oriented methodology, localized Multiscale Energy and Vorticity Analysis (MS-EVA), is developed to infer
fundamental processes from oceanic data for complex dynamics. In general, geofluid flow processes are locally structured and
windowed on scales (i.e., occurring on a range of scales). Many of the existing
GFD theories, however, are based on classical analysis tools
which are global in nature. A gap, therefore, exists between
these theories and real oceanic/atmospheric problems. The
development of MS-EVA attempts to bridge this gap. It begins
with the construction of a functional analysis tool, called
the multiscale window transform (MWT), which is local, self-similar, and windowed on scales. Equations for multiscale energy
and enstrophy are then derived and dynamical processes classified. The classification is made possible with the introduction
of the concept of perfect energy transfer. These concepts allow to establish rigorously an avenue to stability studies. For
example, it proves that barotropic and baroclinic instabilities are simply
represented by the interaction analyses of a certain group of perfect transfer terms.
The theory is validated with the Eady model and Kuo model, and the results are just as expected. The MS-EVA is then
employed to study the Iceland-Faeroe Frontal (IFF) variability, with a real-time dataset gathered during the 1993 NRV
Alliance cruise. The application starts with determining three scale windows, which characterize a
double-peak structure in either the time wavelet spectrum or the space wavelet spectrum. The resulting energetics reveal that
there is a clear baroclinic instability happening around the cold tongue intrusion observed in the middle of the study
domain. Moreover, mesoscale features in the process are found to grow in distinctly different modes at different time stages,
in the form of convective instability
or absolute instability. An interaction analysis shows that the energy released through these instabilities goes to the
sub-mesoscale window as well as the meso-scale window, but most of it stays in the meso-scale window, fueling the growth of
the intrusion.
DE: 3220 Nonlinear dynamics
DE: 3329 Mesoscale meteorology
DE: 4500 OCEANOGRAPHY: PHYSICAL
DE: 4520 Eddies and mesoscale processes
DE: 4568 Turbulence, diffusion, and mixing processes
SC: Ocean Sciences [OS]
MN: 2003 Fall Meeting