U.S. Dept. of Commerce / NOAA / OAR / PMEL / Publications


Variability in the Eastern Equatorial Pacific Ocean During 1986-1988

Michael J. McPhaden and Stanley P. Hayes

NOAA/Pacific Marine Environmental Laboratory, Seattle, Washington

Journal of Geophysical Research, 95(C8), 13,195-13,208 (1990)
Not subject to U.S. copyright. Published in 1990 by the American Geophysical Union.

3. Hydrography

Plate 3 shows meridional temperature sections between 5°N and 5°S along 110°W from six approximately semiannual EPOCS cruises. The May-June 1986 section shows conditions during boreal spring near the onset of the ENSO. Minimum SSTs at the surface near 24°C are centered at 0.5°S, and there is a pronounced weakening of the equatorial thermocline associated with the presence of a geostrophically balanced Equatorial Undercurrent in the upper 100 m [e.g., Picaut et al., 1989]. By October-November 1986, surface temperatures poleward of about 2°N and 2°S have cooled by about 1°C, whereas SST near the equator is still near 24°C. The lack of significant seasonal cooling near the equator indicates the presence of anomalously warm water associated with the developing ENSO. Weakening of the thermocline near the equator is still evident in October-November, though the upper thermocline (as measured, for example, by the depth of the 20°C isotherm) has been depressed by about 50 m between about 2°N and 2°S.

Plate 3. Meridional CTD temperature sections along 110°W during 1986-1988. Contour interval is 1°C. Warmest temperatures are in magenta (>25°C); coldest temperatures are in blue (10°C). Intermediate temperatures are in orange (21°-25°C), green (16°-20°C), and aquamarine (11°-15°C).

Anomalous warming continues into 1987, with temperatures in May-June 1987 3°-4°C warmer in the upper 50 m than during May-June 1986. Moreover, no pronounced equatorial SST minimum appears in the May-June 1987 section, and the thermocline is 20-40 m deeper relative to the previous spring. From boreal spring to boreal fall 1987, significant seasonal cooling occurs south of 4°N, and an equatorial minimum in SST <24°C that is centered near 0.5°S develops. As compared to the fall of 1986, a sharp SST front can be found to the north of the equatorial SST minimum, and the thermocline near the equator is slightly shallower.

Both 1988 sections show an equatorial SST minimum that is several degrees Celsius colder than either of the previous 2 years. In particular, the difference between the May-June 1987 and June-July 1988 minima is 7°C in the center of the cold tongue near 0.5°S. A well-developed SST front is evident north of the equator in both spring and fall 1988. As compared to 1986-1987, the thermocline is much shallower, and there is less evidence for a weakening of the equatorial thermocline, especially in fall 1988.

Variations in thermal structure below the upper 150-200 m can also be seen in these sections. For example, deeper isotherms such as the 10°C isotherm undergo 50- to 100-m excursions, most notably near 5°N. It is probable that some of this variability is related to the 1986-1987 ENSO and the 1988 cold event, given the importance of low baroclinic mode equatorial waves in the dynamics of interannual variability [e.g., Busalacchi and Cane, 1985].

Plate 4 shows zonal sections of temperature along the equator during 1986-1988. The tendency for equatorial SST warming from 1986 to 1987 and cooling from 1987 to 1988 at 110°W is evident at all longitudes between 110°W and 140°W. Similarly, the interannual rise and fall of the thermocline at 110°W ( Plate 3) is evident to the west of 110°W. The thermocline in all cases is deeper near the western terminus of the sections than at the eastern terminus, though its slope is not monotonic from east to west. In 1988, for example, the thermocline undulates with an approximately 1000-km wavelength, which, though barely resolved with the 5° CTD zonal resolution, indicates a return of 20- to 30-day instability waves that are typically suppressed during ENSO [Philander et al., 1985]. These waves are characteristically associated with 100 cm s peak-to-peak meridional velocities [Halpern et al., 1988] whose zonal scale sets the pattern of temperature variations both in the thermocline and at the surface via meridional advection [e.g., Pullen et al., 1987].

Plate 4. Equatorial CTD temperature sections during 1986-1988. Contours are as in Plate 3.

Plate 5 shows the salinity sections corresponding to Plate 4. A prominent feature of these sections is the subsurface salinity maximum (>35 practical salinity units (psu)) in 1986-1987 near depths of 100 m. This maximum is typically located in or just above the core of the Equatorial Undercurrent in the central and eastern Pacific [McPhaden, 1985; Mangum et al., 1986] and has been interpreted in terms of eastward advection of high-salinity southern hemisphere subtropical waters by the Undercurrent [Tsuchiya, 1968; Lukas, 1985]. High-salinity water penetrates further to the east in boreal spring 1986 and 1987 consistent with the seasonal cycle of zonal flow in the Undercurrent (Plate 2, lower panel; see also Halpern [1987a]). However, there is a clear freshening of the thermocline from 1987 to 1988 by 0.2-0.4 psu across 110°-140°W. The appearance of relatively fresh water in the eastern Pacific in 1988 coincides with a relatively shallow thermocline along the equator (Plate 4) and weak isothermal spreading near the equator at 110°W. Direct velocity measurements discussed below (Figures 4 and 5) show 1988 to be a period of weaker than normal eastward flow in the Undercurrent at 110°W.

Plate 5. Equatorial CTD salinity sections during 1986-1988. Contour interval is 0.1 psu. Saltiest water is in magenta (35.3 psu); freshest water is in blue ( 34.6 psu). Intermediate salinities are in orange (35.1-35.2 psu), green (34.9-35.0 psu), and aquamarine (34.7-34.8 psu).

Plate 5 also shows alternating pockets of high and low salinity near the surface with peak-to-peak amplitudes of about 0.3 psu at about 500-km intervals. They are most apparent in 1988, and like the thermocline and SST undulations on similar zonal scales (Plate 4), they are associated with the presence of energetic instability waves. The oscillating salinity pattern is set by meridional advection of a salinity front which separates relatively fresh water (>35 psu) north of the equator from relatively saline water south of the equator [McPhaden et al., 1990b]. Salinity variations below about 200 m, on the other hand, show little variability (<0.2 psu) over the 3-year sequence in Plate 5.


Go back to previous section or forward to next section

PMEL Outstanding Papers

PMEL Publications Search

PMEL Homepage