1 |
Interannual-to-decadal Variation of Tropical-subtropical |
2 |
Exchange in the Pacific Ocean: Boundary Versus Interior |
3 |
Pathways |
4 |
|
5 |
Tong Lee, Ichiro Fukumori, Dimitris Menemenlis, Lee-Lueng Fu |
6 |
Jet Propulsion Laboratory, California Institute of Technology |
7 |
|
8 |
Tropical-subtropical exchange of water masses is considered |
9 |
to be important to interannual-to-decadal variability in the |
10 |
tropical Pacific. On average, the exchange is accomplished |
11 |
by a shallow (< 400 m) meridional circulation which connects |
12 |
the equatorial upwelling, poleward transport of surface |
13 |
Ekman flow, and equatorward transport of pycnocline waters |
14 |
originating from the subtropics. Pycnocline waters arrive |
15 |
at the tropics via western-boundary and interior pathways. |
16 |
It is well-established that, on average, the two pathways |
17 |
re-enforces the net exchange. However, less is known about |
18 |
their relative role on interannual-to-decadal time scales |
19 |
and the underlying physical processes. In this study, we |
20 |
address these issues using sea level measurements obtained |
21 |
by the TOPEX/Poseidon satellite altimter and circulation |
22 |
estimated by the Consortium for Estimating the Circulation |
23 |
and Climate of the Ocean (ECCO; http://www.ecco-group.org). |
24 |
Variation of pycnocline transport via the boundary is found |
25 |
to be (1) anti-correlated to and (2) smaller in magnitude |
26 |
than that of the interior. These features are attributed |
27 |
to the combined effect of two different forcing mechanisms: |
28 |
(1) the variation of local wind stress curl changes the |
29 |
strength of horizontal circulation and results in variation |
30 |
of boundary flow that is opposite in direction and |
31 |
comparable in magnitude to that of interior flow; (2) the |
32 |
variation of equatorial zonal wind stress which affects |
33 |
the strength of meridional circulation with net pycnocline |
34 |
flow opposing the surface Ekman flow. Due to the partial |
35 |
compensation by boundary flow, the convergence of |
36 |
pycnocline waters into the tropics is about 50% of that |
37 |
inferred from interior pycnocline flow alone. The net |
38 |
pycnocline transport reflects ENSO forcing on interannual |
39 |
time scales. There is less equatorward intrusion of |
40 |
pycnocline water in the 1990s than in the 1980s, |
41 |
consistent with recent observation. In the North Pacific, |
42 |
variability of interior pycnocline transport is larger |
43 |
than that through the boundary. This is different from the |
44 |
time mean where the interior transport is substantially |
45 |
smaller than that through the boundary. |
46 |
|