Low-frequency variability of currents in the deepwater eastern Gulf of Mexico
Vertical structure of the low frequency horizontal currents at the northern edge of the Loop Current during eddy shedding events is observed using concurrent hydrographic, moored, and satellite altimetry data from 2005. Dynamic modes are calculated at three deep (~3000 m), full water-column moorings...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1969.1/ETD-TAMU-3030 http://hdl.handle.net/1969.1/ETD-TAMU-3030 |
| Summary: | Vertical structure of the low frequency horizontal currents at the northern edge of
the Loop Current during eddy shedding events is observed using concurrent
hydrographic, moored, and satellite altimetry data from 2005. Dynamic modes are
calculated at three deep (~3000 m), full water-column moorings in the eastern Gulf of
Mexico. Time-series of the barotropic and first two baroclinic modes are found using a
least squares minimization that fits theoretically derived modes to observed moored
velocity data.
EOF analyses show that the majority of observed variance is explained by a
surface-trapped mode that is highly coherent with the temporal amplitudes of the first
baroclinic mode, and a lower, but significant percentage of variance is captured in
bottom-intensified modes. Amplitudes of the second empirical mode indicate that
currents are more coherent in the ocean interior approaching the Loop Current, as more
variance is explained by this mode at the southernmost mooring near the Loop Current.
A dynamic mode decomposition of the horizontal currents reveals that the
barotropic and first baroclinic modes exhibit low frequency variability and eddy time scales of 10 – 40 days. Second baroclinic mode amplitudes show higher frequency
variability and shorter time scales. A model utility test for the least squares fit of
modeled to observed velocity shows that the second baroclinic mode is useful to the
statistical model during 50 – 85 % of the mooring deployment, and is particularly
necessary to the model when cyclonic features are present in the study area. The
importance of the second baroclinic mode to the model increases significantly closer to
the Loop Current.
High-speed currents associated with the Loop Current and anticyclones stimulate
a strong first baroclinic response, but the second baroclinic mode amplitudes are found
to be similar in magnitude to the first baroclinic mode amplitudes at times. This happens
episodically and could be an indication of higher order dynamics related to frontal eddies
or Loop Current eddy shedding. |
|---|