METHODOLOGY
The principal data set studied is the XBT profile compilation produced by Donoso et al (1994), which consists of 36185 profiles (Fig. 1) in the region east of 120°W between 25°S-30°N taken during the period 1980-1996, most of which extend to at least 300m. Since the profiles are concentrated north of the equator, the present analysis studies the region from 3°N to 20°N, namely the "warm pool" of the northeastern tropical Pacific (Fig. 2, top). The profiles were gridded to a 1° by 1° by 10m depth average annual cycle grid (12 average months), using a gaussian weighting scheme (Kessler and McCreary 1993) with scales of 2° longitude, 1° latitude and 2 months. There were an average of about 5 profiles within the e!folding region of influence for each 1° by 1° by 1 month grid box, but these tended to be concentrated east of 110°W and especially along the coastal shipping routes (Fig. 1). For the calculation of geostrophic currents, salinity was estimated via a mean T-S relation found from the Levitus et al (1994) fields, and dynamic heights were calculated relative to 300m depth, which is below most of the isotherm slopes except in the northern parts of the region.

Thermocline depths (represented by the 20°C isotherm) were compared with a simple Rossby wave model forced by the average annual cycle of FSU winds over the period 1961-99 (Stricherz et al 1992). This model allows the thermocline to be influenced by wind stress curl to its east, and also by Rossby waves coming off the coast.

RESULTS: Mean
The mean thermocline (Fig. 2, bottom) shows the eastern ends of the ridges and troughs that are well-known from the central Pacific (the classical ridge-trough nomenclature refers to dynamic height features (Wyrtki and Kilonsky 1984), not thermocline depth): the Countercurrent Ridge (CCR) near 5°N separates the South Equatorial Current (SEC) from the North Equatorial Countercurrent (NECC), the Countercurrent Trough (CCT) near 8°-10°N separates the NECC from the North Equatorial Current (NEC), and the North Equatorial Current Ridge (NECR) near 13°-15°N defines the center of the northern subtropical gyre dipping well south of its mid-basin position. There is little apparent connection between the thermocline features and the SST pattern (Fig. 2, top). Although the thermocline topography is at first glance similar to that in mid-basin, it is noteworthy that both the CCT and NECR weaken greatly near 110°W and then finish in relatively tight eddy-like features at their eastern ends (the terminus of the CCT is the Costa Rica dome). The 110°W ridge-trough weakening is associated with a significant recirculation of both the NECC and the NEC: about half the NECC turns northwestward into the NEC west of 110°W (rather than continuing to the coast), and about half the NEC turns southeastward around the Costa Rica dome (rather than continuing west). About 2 Sv of the flow around the Costa Rica dome appears to enter the equatorial current system, providing another route from the subtropical gyres to the equator.

These unusual features of the circulation can be understood in terms of the distinctive aspects of the winds. While the mid-basin wind field is characterized by trade winds blowing westward and into the ITCZ, in the east the pattern is rather more complicated. An easterly jet flows through Central America via the lowlands of Nicaragua, the low point of the American Cordillera. Unlike the synoptic-timescale Tehuantepec jets (Trasvina et al 1995; McCreary et al 1989), the Nicaraguan jet is a much more regular flow of the Caribbean trades across the lowlands. The result is a easterly surface jet about 400 km wide, well to the east of the main trade system (Fig. 3, top). This produces positive curl on its southern flank, and negative curl on its north (Fig. 3, bottom). These curl features produce the counter-rotating eddies at the east ends of the CCT and CCR (Fig. 2, bottom). When these curls are integrated westwards to obtain the Sverdrup circulation, the integration finds the sign of the curl changing as the Nicaraguan jet merges into the ITCZ (Fig. 3, bottom), and the result is the weakening of all three ridges and troughs at 110°W seen in Fig. 2, bottom), which in turn causes both the NECC and NEC to weaken at 110°W. The result is that the NECC turns around well west of the coast, and there is substantial flow into the NEC from the north well west of the coast, while the circulation further east becomes partly isolated from that in mid-basin.

There are important distinctions between the above-thermocline circulation described above and that in the thermostad layer (between 100 and about 300m, below about 14°C). In the lower layer, there is a single ridge near 8°N. The northern subsurface countercurrent (Tsuchiya jet) flows east under the SEC, then turns around the ridge to the north to flow west as the deep NEC. The northward flow (which is under the Costa Rica dome) is a symptom of water rising into the upwelling region of the dome in accord with Sverdrup theory (Johnson and McPhaden 1999).

RESULTS: Annual cycle
The annual cycle of thermocline depth in the central Pacific is characterized by out of phase variability between the CCR and CCT, with consequent strong annual fluctuations of the NECC and SEC (Wyrtki 1974; Kessler and Taft 1987). Both currents are strong in Oct-Dec and weak in Mar-Jun. Several authors have shown that this can be well accounted for through simple Ekman pumping and linear, first-baroclinic mode physics. This type of annual cycle is seen west of 105°W (Fig. 4). By contrast, the amplitude of the annual cycle decreases to the east, first south of 7°N, and by 90°W (at the Costa Rica dome), there is little annual thermocline depth variation (Fig. 4 and Fig. 5. top left), despite the fact that the wind stress curl variations are similar all the way to the east. Although the small amplitude in this region might seem to be a natural consequence of the fact that Rossby waves travel west, so the Rossby solution is an integral of wind forcing from the east and therefore might be expected to be weak near the coast, this is not the reason for the small amplitude in this case. The amplitude indicated by a simple Rossby model based on the winds is still relatively large in this region (Fig. 5, top right). However, there is a strong annual cycle of thermocline depth along the coast south of Panama, and this boundary condition is phased so as to oppose the annual wind forcing (Fig. 5, bottom right), with the result that the total Rossby solution demonstrates the weak amplitude observed (Fig. 5, bottom left). It is not known what causes the large coastal annual cycle in the Gulf of Panama, but it does not appear to be a simple result of arriving equatorial Kelvin waves.

FUTURE WORK
Future work in this area is taking two directions. First, a more sophisticated use of the data, including the available salinity and chemical constituent observations, is being used in a formal inversion to produce transport estimates that are as consistent as possible with all available information. In particular, this technique can identify regions of upwelling and downwelling that can only be qualitatively inferred from the present analysis.

Second, ion conjunction with the inverse modeling, the analysis will be extended to the region south of the equator, especially for the upwelling along the Peru coast. This upwelling is potentially a contributor to the formation of the southeast Pacific stratus decks that have been identified as a major feature of the air-sea flux balance of the entire region.

REFERENCES
Donoso, M.C., J.E. Harris and D.B. Enfield, 1994: Upper ocean thermal structure of the eastern tropical Pacific. NOAA Technical Report ERL-450-AOML.

Johnson, G.C. and M.J. McPhaden, 1999: Interior pycnocline flow from the subtropical to the the equatorial Pacific. J.Phys.Oceanogr., 29, 3073-3089.

Kessler, W.S. and B.A. Taft, 1987: Dynamic heights and zonal gesotrophic transports in the central tropical Pacific during 1979-1984. J.Phys.Oceanogr., 17, 97-122.

Kessler, W.S. and J.P. McCreary, 1993: The annual wind-driven Rossby wave in the subthermocline equatorial Pacific. J.Phys.Oceanogr., 23, 1192-1207.

Levitus, S., R. Burgett and T. P. Boyer, 1994: World Ocean Atlas 1994, Vol. 3: Salinity. NOAA/NESDIS/NODC, 99pp.

McCreary, J. P., H. S. Lee and D. B. Enfield, 1989: The response of the coastal ocean to strong offshore winds, with application to the Gulfs of Tehuantepec and Papagayo. J.Mar.Res., 47, 81-109.

Stricherz, J., J.J. O’Brien and D.M. Legler, 1992: Atlas of Florida State University tropical Pacific winds for TOGA. Florida State University. 275pp.

Trasviña, A., E. D. barton, H. S. Velez, P. M. Kosro and R.L. Smith, 1995: Offshore wind forcing in the Gulf of Tehuantepec, Mexico: the asymmetric circulation. J.Geophys.Res., 100, 20649-20663.

Wyrtki, K. 1974: Sea level and the seasonal fluctuations of the equatorial currents in the western Pacific Ocean. J.Phys.Oceanogr., 4, 90-103.

Wyrtki, K. and B.J. Kilonsky, 1984: Mean water and current structure during the Hawaii-to-Tahiti Shuttle Experiment. J.Phys.Oceanogr., 14, 242-254.

Dr. William S. Kessler
kessler@pmel.noaa.gov

Institutions:
NOAA / PMEL / OCRD
7600 Sand Point Way NE
Seattle WA 98115 USA

Return to CLIVAR PACS Home Page