East Pacific Warm Pool

The east Pacific warm pool extends roughly from the ITCZ, $5^{\circ} - 10^{\circ} \mbox{ N}$, northward to the Mexican coast. The warm pool is most likely the result of local thermodynamic and mixing processes as well as wind curl stresses. Mean winds are light, but strong bursts reach the warm pool from gaps in the mountains of Central America and from intensifying easterly waves and tropical depressions.

The dynamics of the east Pacific warm pool are therefore likely to have similar elements to those of the west Pacific warm pool. Solar heating of near-surface waters and abundant rainfall should produce complex structures limiting the depth of the mixed layer during light winds. During light winds the surface layer probably has a daily cycle of convectively-driven homogenization and restratification similar to that in the west, but with different amplitudes and depth ranges reflecting local conditions. In particular, because the eastern warm pool is north of the equator, ``daily deep cycles'' of strong mixing below the surface layer are unlikely. The deep cycle is a daily extension of turbulence into the highly-sheared, strongly-stratified water below the surface layer. The turbulence is strongest at night and is clearly driven by convection in the mixed layer. Just how remains controversial. On the equator the deep cycles are produced by the daily convective cycle above a thermocline strongly sheared between the westward surface drift and the eastward undercurrent. The turbulent heat flux at the base of the layer is a major component of the layer's heat balance and is modulated by anything changing the mean shear, e. g., Kelvin waves. In view of the experience of TOGA/COARE, EPIC must pay particular attention to the shallow thermohaline structure and the depth-dependent absorption of solar radiation while determining the budgets of heat, freshwater, and momentum.

  

Figure 7: Comparison between $(95^\circ \mbox{ W} , 8^\circ \mbox{ N})$(right panel) and $(156^\circ \mbox{ E} , 0^\circ \mbox{ N})$ (left panel) of wind speed air temperature, SST, and depth of the $20^\circ \mbox{ C}$ contour in the ocean. The time interval is the northern summer of 1996. Notice the difference in SST scales between the two panels. Data are from NOAA/PMEL's TAO moorings at these locations.

There are also likely to be considerable differences between the forcing of west and east Pacific warm pools. The thermocline is much shallower in the east, limiting the maximum potential thickness of the mixed layer. Although there is significant intraseasonal variability in the east, the dominant atmospheric variability has time scales of several days to two weeks and is associated with the passage of easterly waves and developing tropical storms. By contrast, the strongest time scales in the west are the diurnal and intraseasonal oscillations. Episodic deepenings lasting a few days can be expected during high winds in traveling storms and bursts extending westward from gaps in the mountains of Central America. The frequency and structure of these events should differ considerably from the westerly bursts and typhoons over the western warm pool. The response will also differ, owing to the much shallower thermocline and lower mean shear. This is seen in Figure 7, which compares wind speed, air temperature, SST, and thermocline depth between the summertime northeastern Pacific with the equatorial western Pacific. Notice how the SST variations on the time scale of a few days are much larger in the east than in the west, in spite of only slightly stronger average windspeeds. These variations, which are large enough to affect the character of the overlying convection, are clearly correlated with windspeed history. This confirms the tightly coupled nature of the ocean-atmosphere system in this region.

A spectrum of upper ocean processes should be excited during the passage of easterly waves and tropical disturbances, such as near-inertial motions within the oceanic mixed layer and in the thermocline (Smyth, Hebert, and Moum, 1996a,b). Entrainment mixing is presumably due primarily to vertical shear of the near-inertial currents, and to a lesser extent surface-generated turbulence within the mixed layer. Both mechanisms modulate the mixed layer thermodynamics and influence surface fluxes across the air-sea interface.

Because the western warm pool is more symmetric about the equator, COARE mixed layer studies were centered at $1.75^{\circ} \mbox{ S}$ and thereby imbedded in the equatorial current system (Wijesekera and Gregg, 1996). The eastern warm pool is approximately centered between the much weaker North Equatorial Current and the North Equatorial Counter Current. Mean shears below the mixed layer are therefore expected to be much less than in the west.

When episodic strong winds deepen the western warm pool, much of the deepening occurs into a profile already close to shear instability owing to the strong mean shear between the eastward undercurrent and the westward surface drift. In the eastern warm pool the background shear is most likely produced by low-frequency internal waves and we have no reason to expect that the region immediately below the mixed layer is usually close to shear instability. Consequently, the regime should differ from that studied during TOGA/COARE and from that at higher latitudes, where the inertial frequency is much higher.

Gregg, et al. (1996) found that internal waves at low latitudes in the central and western Pacific were more energetic than those at mid latitudes but produced weak mixing. The finescale shear spectra differed substantially from the Garrett and Munk model and there is no model to predict what to expect at other low-latitude sites. However, at $6.5^{\circ} \mbox{ S}$ in the Banda Sea, within the Indonesian archipelago, shear wavenumber spectra resembled those found earlier in the low-latitude Pacific, and shear, strain, and turbulent mixing were strongly near-inertial (Alford and Gregg, 2000). Therefore, EPIC2001 mixed layer studies should include understanding the mixing regime of the upper thermocline and need to last for at least two inertial periods at each site.

Upper ocean heat and mass budgets are also influenced by horizontal advection (Jacob et al., 2000). TAO moorings along 95 $^{\circ} \mbox{ W}$ provide long-term measurements of the upper ocean conditions that are needed in understanding the role of meridional advection by low-frequency flows. However, zonal advection of temperature gradients may also be important in the mixed layer and upper thermocline response in the east Pacific depending on the magnitude and direction of the applied wind stress. It is important to understand the effects of spatial variability of the upper ocean currents and current shears on ocean mixed layer thermodynamics.

The ocean's momentum response to atmospheric forcing is classified into two regimes: the directly-forced or near-field, and the wake or far-field. In the near-field, the wind stress significantly increases the ocean mixed layer currents, which generates vertical current shear (Sanford et al., 1987) and hence entrainment mixing through shear instabilities. When the rotation of the wind stress vector of the moving disturbance is in phase with background inertial motions in the mixed layer (which represent the low-frequency end of the internal wave spectrum), a resonant interaction increases upper ocean current shear substantially that deepens the ocean mixed layer (Price 1981). The spatial extent of the forced current and its shear field needs to be understood through the mixed layer and into the thermocline.

The far-field response to an easterly wave may be characterized as two-dimensional with strong near-inertial motions (Kundu and Thompson, 1985) whereas the response to a moving cyclone is a three-dimensional wake (Shay et al., 1998). In both cases, the wavelength of the response will be proportional to the speed of translation, and the local inertial period (about $70 \mbox{ h}$ at $10^{\circ} \mbox{ N}$). While there has been a substantial amount of research focused on near-inertial response at midlatitudes for extratropical and tropical storms, little is known about this response within $10^{\circ}$ of the equator where inertial periods are significantly larger and project onto longer-time scale variability.

The accompanying divergence of upper ocean near-inertial currents also causes upwelling of cooler water. The upwelling of the cooler, stratified water will tend to weaken turbulent mixing processes by keeping the Richardson numbers above criticality. Given shallow thermoclines of $50 \mbox{ m}$ to $75 \mbox{ m}$ and reduced gravities of order $0.05 \mbox{ m} \mbox{ s}^{-2}$, how the mixing events are modulated by the internal waves (and near-inertial processess) and the upwelling processes is critical to the SST, sea surface salinity, and the ocean mixed layer response in the warm pool.