Geoscience Reference
In-Depth Information
29,5
29,5
29
29
28,5
28,5
28
28
27,5
27,5
27
27
26,5
26,5
26
26
25,5
25,5
29,5
29
28,5
28
27,5
27
26,5
26
25,5
Figure 17.7. Sequence of surface temperature in the standard geometry at 69, 73, and 77 revolutions.
ocean basins [ Bowden and Eden , 1968; Maxworthy and
Browand , 1974; Rayer et al. , 1998].
Not all oceanic flows are completely blocked in the
zonal direction. The Antarctic Circumpolar Current
(ACC), for example, is not blocked, but the flow has to
weave through the Drake Passage between the southern
tip of South America and the Antarctic continent. The
Drake Passage partially blocks the zonal flow. It is there-
fore straightforward to perform experiments with par-
tial barriers to understand better the intermediate case
between a free and fully blocked annulus flow.
In the experiment discussed here, the azimuthal flow is
blocked at the inner cylinder and at the bottom (see Figure
17.9). At the barrier, the gap width is reduced from 75 to
43 mm and the fluid depth is reduced from 135 to 95 mm.
More details can be found in the work of Harlander et al.
[2012a].
Figure 17.7 shows a sequence of surface temperature
images taken from an experiment in standard geometry
with = 4.6rpm and T = 2.8K. We see the baro-
clinic wave after 69, 73, and 77 revolutions. Obviously,
the wave is rather stable and rotates counterclockwise
(that is, prograde) within a prograde mean flow. In con-
trast, in Figure 17.8, we see a comparable experiment
( = 4.8rpm and T = 4.0K) but then with the
barrier mounted. The most obvious new feature is wave
breaking at the barrier and wave recovery downstream
of the barrier. That is, the baroclinic wave never satu-
rates but is invariably in a transient state. We can say
that the barrier leads to a mechanically induced baroclinic
life cycle. Life cycles play an important role for midlat-
itude atmospheric and oceanic flows where they occur
due to linear growth, nonlinear saturation, and dissipa-
tive decay of large-scale waves. The experiment with the
barrier opens the possibility to study baroclinic life cycles
in a controlled way.
The experiment we discuss shows a slow periodic
variation of the radial temperature difference with an
amplitude of 1K and a period of 26 min. The purpose
of this variation of boundary conditions is to make sure
that the wave breaking is due to the barrier and not
due to regime transitions that occur for certain Taylor
and Rossby numbers. We can exclude the latter from
the fact that the flow looks very similar for maximum
and minimum T . It is important to note that during
the experiment we observed the surface velocity and sur-
face temperature simultaneously. This allows for analyz-
ing coupled temperature and velocity anomalies and the
evaluation of the mean and turbulent surface heat flux.
The CEOF analysis already applied in Section 17.3.1 is
eminently suited for our data [ Pfeffer et al. , 1990]. The
CEOF analysis decomposes a propagating mode into a
 
Search WWH ::




Custom Search