Thierry Delcroix

Mechanism of the Zonal Displacements of the Pacific Warm Pool: Implications for ENSO

The western equatorial Pacific warm pool is subject to strong east-west migrations on interannual time scales in phase with the Southern Oscillation Index. The dominance of surface zonal advection in this migration is demonstrated with four different current data sets and three ocean models. The eastward advection of warm and less saline water from the western Pacific together with the westward advection of cold and more saline water from the central-eastern Pacific induces a convergence of water masses at the eastern edge of the warm pool and a well-defined salinity front. The location of this convergence is zonally displaced in association with El Niño-La Niña wind-driven surface current variations. These advective processes and water-mass convergences have significant implications for understanding and simulating coupled ocean-atmosphere interactions associated with El Niño-Southern Oscillation (ENSO).

Seasonal and interannual variations of sea surface salinity in the tropical Pacific Ocean

Sea surface bucket measurements, obtained through a ship-of-opportunity program, are used to describe the sea surface salinity (SSS) field for the tropical Pacific during the period 1969–1988. Emphasis is placed upon the mean SSS distribution and the seasonal and interannual SSS variability occurring along four well-sampled shipping tracks. These tracks extend from New Zealand to Japan, from New Zealand to Hawaii, from Tahiti to California, and from Tahiti to Panama. They cross the equator at 155°E, 160°W, 140°W, and 100°W, respectively. Along each track, the mean SSS distribution is characterized by SSS minima which are 4°–6° further poleward than the axes of maximum precipitation associated with the Intertropical Convergence Zone (ITCZ) and South Pacific Convergence Zone (SPCZ). It is suggested that these SSS minima owe their existence mainly to heavy rainfall and poleward Ekman salt transport associated with the trade winds. The role of zonal salt advection was found negligible for these SSS minima. Except along the eastern track, maximum seasonal SSS variations are located in the ITCZ and SPCZ regions, with minimum SSS in September-October and March-April, respectively. On the basis of precipitation island stations, it is demonstrated that the maximum seasonal SSS variations are closely related to the rainfall regimes of the ITCZ and SPCZ (rainfall maximum 3 months before SSS minimum; rainfall amount sufficient to account for SSS changes). Along the eastern track, a strong annual SSS cycle is found from about 4°S (110°W) to the Panama coast (minimum SSS in February–March), reflecting the combined effects of rainfall, salt advection, and vertical mixing. Notable interannual SSS variability concerns only the western half of the tropical Pacific Ocean where El Nino-Southern Oscillation (ENSO) related SSS changes are strongly related to ENSO-related precipitation changes. During ENSO periods, the SSS field west of about 150°W is characterized by fresher-than-average SSS within about 8°N to 8°S, and conversely saltier-than-average SSS poleward of 8° latitudes. These modifications in the SSS field are thought to result mainly from an eastward displacement in the ascending branch of the Walker and Hadley cells which induces unusually high rainfall over the western and central equatorial Pacific region bordered on all sides by rainfall deficits. Reproducing the actual SSS changes at seasonal and interannual time scales would be a very stringent test for model capability.

Zonal displacement of the western equatorial Pacific fresh pool

The distribution of salt in the tropical Pacific Ocean and its variability are potentially important in better understanding the El Nino Southern Oscillation (ENSO) ocean-atmosphere coupled system. Complementary data sets (including sea surface salinity (SSS) derived from a ship-of-opportunity network, Geosat sea level and derived zonal surface currents, and outgoing longwave radiation-derived precipitation, evaporation, and cruise measurements along the 165°E longitude) are used to describe and understand SSS changes in the western equatorial Pacific over several ENSO cycles. It is first shown that the 1974–1989 zonal displacements of the eastern edge of the western equatorial Pacific “fresh pool” (SSS < 35), marked by a salinity front and closely related to the eastern edge of the warm pool, were dominated by interannual variations that are highly correlated with the Southern Oscillation Index (SOI). Owing to the availability of basin-wide Geosat surface current data, it is then shown that the eastward extension of the eastern edge of the fresh pool during the 1986–1987 El Nino, as well as the westward retreat during the 1988–1989 La Nina, were chiefly the result of zonal salt advection in the equatorial band. Although the equatorial upwelling extended toward the western Pacific during La Nina, its effectiveness in changing the near-surface salinity (and temperature) is found questionable. Bearing in mind the uncertainty in determining open-ocean precipitation, evaporation, and mixed-layer depth, it appears that the net freshwater flux probably played a minor role in the zonal displacements of the eastern edge of the fresh pool, as compared to zonal salt advection. The roles of first baroclinic Kelvin and first meridional mode Rossby waves are discussed considering that they account in a large part for zonal advection.

Evidence for stronger El Niño‐Southern Oscillation (ENSO) Events in a Mid‐Holocene massive coral

We present a 47-year-long record of sea surface temperature (SST) derived from Sr/Ca and U/Ca analysis of a massive Porites coral which grew at ∼ 4150 calendar years before present (B.P.) in Vanuatu (southwest tropical Pacific Ocean). Mean SST is similar in both the modern instrumental record and paleorecord, and both exhibit El Nino-Southern Oscillation (ENSO) frequency SST oscillations. However, several strong decadal-frequency cooling events and a marked modulation of the seasonal SST cycle, with power at both ENSO and decadal frequencies, are observed in the paleorecord, which are unprecedented in the modern record.

Precipitation and sea-surface salinity in the tropical Pacific Ocean

Monthly sea-surface salinity (SSS) and precipitation (P) in the tropical Pacific region are examined for the 1974–1989 period. The SSS data are derived mainly from water sample measurements obtained from a ship-of-opportunity program, and the rainfall data are derived from satellite observations of outgoing longwave radiation. The mean and standard deviation patterns of SSS and P exhibit good correspondence in the heavy-rainfall regions characterising the Intertropical Convergence Zone (ITCZ), the South Pacific Convergence Zone (SPCZ) and part of the western Pacific warm pool. An Empirical Orthogonal Function (EOF) analysis indicates two dominant modes of variation linking P and SSS changes, one mode at the seasonal timescale in both convergence zones, and the other at the ENSO timescale in the central-western equatorial Pacific (165°E–160°W) and in the SPCZ. The inferences derived from the EOF analysis are used in a simple linear regression model in order to try to specify P changes from known SSS changes. A comparison between hindcast and observed P changes suggests that, at seasonal and ENSO timescales, SSS changes could be used to infer the timing, but not the magnitude, of P in the central-western equatorial Pacific (165°E–160°W) and in the SPCZ mean area. The effects of evaporation, salt advection and mixed-layer depth on the results are discussed.

Observed surface oceanic and atmospheric variability in the tropical Pacific at seasonal and ENSO timescales: A tentative overview

Seasonal and El Nino-Southern Oscillation (ENSO)-related variations of sea surface temperature (SST) and salinity (SSS), 0/450-dbar dynamic height anomalies (η, an alias for sea level), zonal (τx) and meridional (τy) wind stress, wind stress curl (curl (τ)), and precipitation (P) are examined in the tropical Pacific during 1961–1995. In the equatorial band the El Nino (La Nina) events are chiefly concerned (1) in the east and center, with warmer (colder) than average SST and a η increase (decrease), and (2) in the west, with fresher (saltier) than average SSS, westerly (easterly) wind anomalies, above (below) average P limited to the east of about 150°E, and a η decrease (increase); Much smaller ENSO changes occur away from the equatorial band except in the convergence zones for SSS, P, arid τy changes and in two patches centered around 7°N and 7°S in the west for curl (τ). The ENSO-related η changes are schematically concerned with a zonal “seesaw” in phase with the Southern Oscillation Index (SOI) in the equatorial band and a meridional seesaw between the regions situated north and south of about 5°N, which lags by about 1 year behind the SOI. The double seesaws result in a longitudinal mean η rise (drop) within about 5°N–20°S up to the mature phase of El Nino (La Nina), and not just until its beginning, partly compensated by a longitudinal mean η drop (rise) within about 5°–20°N. Aside from its intrinsic substance, this paper offers a novel and concise observational basis for testing theoretical studies and model simulations.

The oceanic zone of convergence on the eastern edge of the Pacific warm pool : A synthesis of results and implications for El Niño-Southern Oscillation and biogeochemical phenomena

The eastern edge of the western Pacific warm pool corresponds to the separation between the warm, rainfall-induced low-salinity waters of the warm pool and the cold, high-salinity upwelled waters of the cold tongue in the central-eastern equatorial Pacific. Although not well defined in sea surface temperature (SST), this eastern edge is characterized by a sharp salinity front that is trapped to the equator. Several studies, using numerous in situ and satellite data and three classes of ocean models, indicate that this front is the result of the zonal convergence of the western and central Pacific water masses into the eastern edge of the warm pool. This occurs through the frequent encounter of the eastward jets in the warm pool and the westward South Equatorial Current in the cold tongue. The notable and alternate variations of these wind-driven zonal currents are trapped to the equator and are chiefly interannual in the vicinity of the edge. Consequently, the Eastern Warm Pool Convergence Zone (EWPCZ) is subject to eastward or westward displacements over several thousands of kilometers along the equatorial band, in synchrony with the warm phase (El Nino) and the cold phase (La Nina) of the El Nino-Southern Oscillation (ENSO) phenomenon. Zonal advection appears to be the predominant mechanism for the ENSO displacements of the eastern edge of the warm and fresh pool. The existence of the EWPCZ and its ENSO displacements have significant effects on the physics of the tropical Pacific and on related biogeochemical phenomena. The EWPCZ is important for the formation of the barrier layer in the isothermal layer of the warm pool. Its zonal displacements control SST in the central equatorial Pacific, which in turn drives the surface winds and atmospheric convection (and vice versa). Hence the central equatorial Pacific is a key region for ENSO coupled interactions. All these findings from several studies and additional analyses lead to a revision of the delayed action oscillator theory of ENSO. The existence of the EWPCZ and its zonal displacements are also reasons for the ENSO variations in production and exchange of CO2 with the atmosphere over the equatorial Pacific. The zone of one-dimensional convergence seems to congregate the worlds most important tuna fishery in the western equatorial Pacific, and its displacements are likely the reason for this fishery to move zonally over thousands of kilometers in phase with ENSO.

Equatorial Kelvin and Rossby waves evidenced in the Pacific Ocean through Geosat sea level and surface current anomalies

Equatorial Kelvin and Rossby waves are comprehensively demonstrated over most of the equatorial Pacific basin, through their signatures in sea level and zonal surface geostrophic current anomalies. This was made possible with altimeter data pertaining to the first year of the Geosat (Geodetic Satellite) 17-day exact repeat orbit (November 8, 1986, to November 8, 1987). To this end, along-track corrected Geosat sea level anomalies (SLAs), relative to the time period of interest, were first smoothed using nonlinear and linear filters. The original 17-day time step was then reduced by combining all ascending and descending tracks within 10° longitudinal bands. Finally, SLAs were gridded onto a regular grid, and low-pass filters were applied in latitude and time in order to smooth out remaining high-frequency noise. Anomalies of zonal surface geostrophic current were calculated using the first and second derivatives of the SLA meridional gradient, off and on the equator, respectively. Sea level and surface current anomalies are validated in the western equatorial Pacific with in situ data gathered during seven hydrographic cruises at 165°E, and through expendable bathythermograph and mooring measurements. Following their chronological appearance along the 165°E meridian, the major low-frequency SLAs and zonal surface current anomalies are described and explained in terms of the equatorial wave theory. An equatorial downwelling Kelvin wave, known to be the main oceanic signal of the 1986-1987 El Nino, is generated in December 1986, concomitant with a strong westerly wind anomaly occurring west of the dateline. The associated propagating equatorial SLAs correspond to an elevation of 15 cm. Independent estimates of this Kelvin wave phase speed are obtained through time-lag correlation matrix analysis (2.82 ± 0.96 m s−1) and the least squares fit of the SLA meridional structures to theoretical Kelvin wave shape (2.26 ± 1.02 m s−1). Both estimates indicate that the Kelvin wave has the characteristic of a first baroclinic mode. An equatorial upwelling Kelvin wave is then detectable in June 1987. It is characterized by a 10-cm sea level drop, propagating only from the western to the central equatorial Pacific. A first meridional mode (m = 1) equatorial upwelling Rossby wave crossing the entire Pacific basin from March 1987 (eastern part) to September 1987 (western part) shows up in SLAs and zonal surface current anomalies. Such a Rossby wave corresponds to propagating sea level drops which are extreme (−12 cm) at about 4°N and 4°S latitudes. The consequences on zonal surface geostrophic current are very important since, in the case of the upwelling, it dramatically decreases the three major surface currents (the North and South Equatorial Countercurrents, and South Equatorial Current) by an amplitude similar to their mean annual velocity values. The least squares fit of the Rossby wave SLA meridional structures to its theoretical m = 1 form cogently suggests the dominance of the first baroclinic mode (c = 2.59 ± 0.65 m s−1). This dominance is corroborated by an estimate of the Rossby wave phase speed (1.02 ± 0.37 m s−1), which roughly corresponds to the theoretical phase speed (c/2m + 1) of the m = 1 equatorial Rossby wave. It is suggested that the equatorial upwelling Rossby wave is mostly due to a reflection of an equatorial upwelling Kelvin wave generated in January 1987 near the dateline. Whether or not the overall propagating features are part of the 1986–1987 El Nino or belong to the “normal” seasonal cycle cannot be decided in the absence of longer altimeter sea level time series.

Near-Surface Salinity as Nature’s Rain Gauge to Detect Human Influence on the Tropical Water Cycle

AbstractChanges in the global water cycle are expected as a result of anthropogenic climate change, but large uncertainties exist in how these changes will be manifest regionally. This is especially the case over the tropical oceans, where observed estimates of precipitation and evaporation disagree considerably. An alternative approach is to examine changes in near-surface salinity. Datasets of observed tropical Pacific and Atlantic near-surface salinity combined with climate model simulations are used to assess the possible causes and significance of salinity changes over the late twentieth century. Two different detection methodologies are then applied to evaluate the extent to which observed large-scale changes in near-surface salinity can be attributed to anthropogenic climate change.Basin-averaged observed changes are shown to enhance salinity geographical contrasts between the two basins: the Pacific is getting fresher and the Atlantic saltier. While the observed Pacific and interbasin-averaged sal...

Variation of the western equatorial Pacific Ocean, 1986–1988

Twenty-one oceanographic sections made along 165°E during 1984–1988 provide a unique picture of the 1986–1987 El Nino and the subsequent La Nina in the western equatorial Pacific. The mean of six cruises from January 1984 through June 1986, a relatively normal period, provides a reference with which the later sections are compared. The net warm water transport across 165°E within 10° of the equator was small in this mean reference section: 7 × 106 m3 s−1 to the east. In December 1986, strong westerly winds at and to the west of 165°E increased the net eastward transport of warm water to 88 × 106 m3 s−1, and the 1986–1987 El Nino was underway. During the following 2 years the net transport varied widely and rapidly; the extrema were 56 × 106 m3 s−1 to the east and 58 × 106 m3 s−1 to the west. Changes in the stratification along 165°E were correspondingly large, reflecting both the geostrophic balance of the strong zonal currents and the changes in the volume of warm water in the western equatorial Pacific. The anomaly of warm water volume corresponded closely to the time integral of the warm water transport across 165°E. Local wind forcing and remotely forced waves were both important causes of the transport fluctuations. Winds, precipitation, and currents were all important factors determining the depth of the surface mixed layer and the thickness of the underlying barrier layer. The way in which these factors interact is a strong function of latitude.