2006 Wang Fiedler

Progress in Oceanography
Progress in Oceanography 69 (2006) 239266 www.elsevier.com/locate/pocean
ENSO variability and the eastern tropical Pacic: A review

Chunzai Wang
a
a,*
, Paul C. Fiedler
NOAA Atlantic Oceanographic and Meteorological Laboratory, Physical Oceanography Division, 4301 Rickenbacker Causeway, Miami, FL 33149, USA b NOAA Southwest Fisheries Science Center, La Jolla, CA, USA Available online 19 May 2006
Abstract El Nin o-Southern Oscillation (ENSO) encompasses variability in both the eastern and western tropical Pacic. During the warm phase of ENSO, the eastern tropical Pacic is characterized by equatorial positive sea surface temperature (SST) and negative sea level pressure (SLP) anomalies, while the western tropical Pacic is marked by o-equatorial negative SST and positive SLP anomalies. Corresponding to this distribution are equatorial westerly wind anomalies in the central Pacic and equatorial easterly wind anomalies in the far western Pacic. Occurrence of ENSO has been explained as either a self-sustained, naturally oscillatory mode of the coupled oceanatmosphere system or a stable mode triggered by stochastic forcing. Whatever the case, ENSO involves the positive oceanatmosphere feedback hypothesized by Bjerknes. After an El Nin o reaches its mature phase, negative feedbacks are required to terminate growth of the mature El Nin o anomalies in the central and eastern Pacic. Four requisite negative feedbacks have been proposed: reected Kelvin waves at the ocean western boundary, a discharge process due to Sverdrup transport, western Pacic wind-forced Kelvin waves, and anomalous zonal advections. These negative feedbacks may work together for terminating El Nin o, with their relative importance being time-dependent. ENSO variability is most pronounced along the equator and the coast of Ecuador and Peru. However, the eastern tropical Pacic also includes a warm pool north of the equator where important variability occurs. Seasonally, ocean advection seems to play an important role for SST variations of the eastern Pacic warm pool. Interannual variability in the eastern Pacic warm pool may be largely due to a direct oceanic connection with the ENSO variability at the equator. Variations in temperature, stratication, insolation, and productivity associated with ENSO have implications for phytoplankton productivity and for sh, birds, and other organisms in the region. Long-term changes in ENSO variability may be occurring and are briey discussed. This paper is part of a comprehensive review of the oceanography of the eastern tropical Pacic. Published by Elsevier Ltd.
1. Introduction Our rst knowledge of El Nin o came from Peruvian geographers, who at the end of the 19th century were interested in the unusual climate aberrations that occurred along the Peru coast in the odd year (Eguiguren,
*
Corresponding author. E-mail address: [email protected] (C. Wang).
0079-6611/$ - see front matter. Published by Elsevier Ltd. doi:10.1016/j.pocean.2006.03.004
240
C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266
1894). They took note of what a knowledgeable ship captain said about the shermen in northern Peru, who typically saw a switch from cold to tropical ocean conditions around Christmas of every year and attributed this to a southward warm El Nin o current. This name was a reference to the annual celebration of the birth of the Christ child, who is much more prominent than Santa Claus/Saint Nicholas in Latin American traditions of the Christmas season. The geographers noted that in some years the onset of warm conditions was stronger than usual and was accompanied by unusual oceanic and climatic phenomena. Starting with the arrival of foreign-based scientic expeditions o Peru in the early 20th century, the concept gradually spread through the worlds scientic community that El Nin o referred to the unusual events (Murphy, 1926; Lobell, 1942). The annual occurrence was forgotten. It was separately noted by Sir Gilbert Walker in the 1920s and 1930s that notable climate anomalies occur around the world every few years, associated with what he called the Southern Oscillation (SO) (Walker, 1923, 1924, 1928; Walker and Bliss, 1932). The SO is a large interannual uctuation in tropical sea level pressure (SLP) between the Western and Eastern Hemispheres (SO index is dened as SLP anomaly dierence between Tahiti and Darwin). It was not until the 1960s that scientists came to realize that the warming o Peru is only part of an ocean-wide perturbation that extends westward along the equator out to the date line. Berlage (1957, 1966) recognized the linkage between the SO and episodic warmings of sea surface temperature (SST) along the coast of Southern Ecuador and Northern Peru, known locally as El Nin o. El Nin o became associated with unusually strong warmings that occur every two to seven years in concert with basin-scale tropical Pacic Ocean anomalies. About the same time, the noted meteorologist Jacob Bjerknes proposed that El Nin o was just the oceanic expression of a large-scale interaction between the ocean and the atmosphere. Using observed data in the context of earlier studies dating back to those of Walker (1924), Bjerknes (1966, 1969) provided evidence that the long-term persistence of climate anomalies associated with the Walkers SO (Walker and Bliss, 1932) is closely associated with slowly evolving SST anomalies in the equatorial eastern and central Pacic. Bjerknes recognized the importance of oceanatmosphere interaction over the eastern tropical Pacic. He hypothesized that a positive oceanatmosphere feedback involving the Walker circulation is responsible for the SST warming observed in the equatorial eastern and central Pacic. In his seminal paper, he stated (Bjerknes, 1969, p. 170): A decrease of the equatorial easterlies weakens the equatorial upwelling, thereby the eastern equatorial Pacic becomes warmer and supplies heat also to the atmosphere above it. This lessens the eastwest temperature contrast within the Walker Circulation and makes that circulation slow down. This positive oceanatmosphere feedback or coupled oceanatmosphere instability leads the equatorial Pacic to a never-ending warm state. During that time, Bjerknes did not know what causes a turnabout from a warm phase to a cold phase (Bjerknes, 1969, p. 170): There is thus ample reason for a never-ending succession of alternating trends by airsea interaction in the equatorial belt, but just how the turnabout between trends takes place is not quite clear. The positive oceanatmosphere feedback of Bjerknes (1969) has inuenced later studies. The essence of Bjerknes hypothesis still stands as the basis of present day work. Oceanographers and meteorologists began to combine their eorts to expand and rene the Bjerknes hypothesis by systematically studying the El Nin o and the Southern Oscillation together in what we now call El Nin o-Southern Oscillation, or ENSO. After Bjerknes published his hypothesis, ENSO was not intensively studied until the 1980s. The intense warm episode of the 19821983 El Nin o, which was not recognized until it was well developed, galvanized the tropical climate research community to understand ENSO and ultimately predict ENSO. The 19821983 El Nin o was not consistent with the buildup of sea level in the western Pacic by stronger than normal trade winds prior to 1982, presumed to be a necessary precursor of El Nin o (Wyrtki, 1975). Also, there was no warming o the west coast of South America in early 1982, considered to be part of the normal sequence of events characterized the evolution of El Nin o (Rasmusson and Carpenter, 1982). This motivated the 10year international TOGA (Tropical Ocean-Global Atmosphere) program (198594) to study and predict ENSO. One outcome was to build the ENSO observing system that includes the TAO/TRITON array of moored buoys (Hayes et al., 1991; McPhaden et al., 1998), an island tide-gauge network, surface drifters, the volunteer ship program, and various satellite observations. TOGA also supported analytical and diag-
241
nostic studies of the ENSO phenomenon (Wallace et al., 1998), and the development of a sequence of coupled oceanatmosphere models to study and predict ENSO (Philander, 1990; Neelin et al., 1998; Wang and Picaut, 2004). Since TOGA, our understanding of ENSO has been greatly advanced by focusing on interaction between the tropical Pacic Ocean and atmosphere. This paper will provide a brief review of ENSO observations and of our present understanding of ENSO, with a focus on patterns and processes in the eastern tropical Pacic Ocean. ENSO variability in the eastern tropical Pacic is centered along the equator, but is closely related to variability of the tropical Western Hemisphere warm pool (WHWP), which has been dened as the region covered by water warmer than 28.5 C (Wang and Eneld, 2001, 2003). The WHWP is comprised of the eastern north Pacic west of Central America; the Intra-Americas Sea (IAS), i.e., the Gulf of Mexico and the Caribbean; and the western tropical North Atlantic. The WHWP is the second-largest tropical warm pool on Earth. Unlike the western Pacic warm pool in the Eastern Hemisphere, which straddles the equator, the WHWP is entirely north of the equator. The WHWP has a large seasonal cycle and the interannual uctuations of its area are comparable to the annual variation, although it does not undergo large anomalous zonal excursions such as occur in the western Pacic. The WHWP is a critical component of the boreal summer climate of the Caribbean and surrounding land areas. From an oceanographic point of view, the WHWP can be separated into two parts by the Central American landmass: the eastern north Pacic warm pool and the Atlantic warm pool. To the atmosphere, the WHWP is a monolithic heat source that annually migrates and changes in size (Wang, 2002b), with little regard for the narrow landmass of Central America. We nevertheless recognize that WHWP development may involve oceanographic processes that are fundamentally dierent between the two oceans (Wang and Eneld, 2003). This paper will focus on review of seasonal and interannual variations of the eastern Pacic component of the WHWP, because of the eastern Pacic warm pool being part of the eastern tropical Pacic that is the subject of this review volume. ENSO variability and the eastern Pacic warm pool are related to eastern tropical Pacic interdecadal variability reviewed by Mestas-Nun ez and Miller (2006), the ocean circulation of the eastern tropical Pacic by Kessler (2006), atmospheric forcing of the eastern tropical Pacic by Amador et al. (2006), and hydrography of the eastern tropical Pacic by Fiedler and Talley (2006). ENSO variability is associated with biological and ecological variability in the eastern tropical Pacic. We herein also briey review biological and ecological eects of ENSO. The paper is organized as follows. Section 2 briey describes major observed features of ENSO. Section 3 reviews our present understanding of ENSO. Section 4 presents seasonal and interannual variations of the eastern Pacic warm pool. Section 5 briey reviews ENSO biological and ecological variability. Section 6 discusses changes in ENSO variability. Finally, Section 7 provides a summary. 2. Observations of ENSO ENSO variability has been documented in the written record over hundreds of years (e.g., Quinn et al., 1987; Eneld, 1989). It is evident in paleoclimatic records, with slight changes in amplitude or frequency, over thousands of years (Diaz and Markgraf, 1992, 2000). For example, Rodbell et al. (1999) showed that the frequency of ENSO variability increased progressively over the period from about 70005000 years ago, and archaeological evidence suggests that El Nin o events were either absent or very dierent from today for several millennia prior to that time (Sandweiss et al., 2001). This paper will not review ENSO variability based on paleoclimatic records. Instead, we will focus on ENSO variability from modern observational data. Numerous observational studies had been published by the early 1980s describing the structure and evolution of ENSO (e.g., Trenberth, 1976; Weare et al., 1976; Quinn et al., 1978; Van Loon and Madden, 1981; Pazan and Meyers, 1982; Wooster and Guillen, 1974; Ramage and Hori, 1981; Weare, 1982; Wyrtki, 1975; Rasmusson and Carpenter, 1982). During the TOGA decade (19851994), more observational papers provided a much improved description and understanding of ENSO. McPhaden et al. (1998) and Wallace et al. (1998) have provided a comprehensive review of ENSO variability and structure from an observational point of view for the TOGA decade. In this section, we only briey describe major observed features of ENSO variability. The SST monthly anomalies in the Nin o3 region (5 S5 N, 150 W90 W) are often used to index ENSO variability. Nin o3 includes the eastern equatorial Pacic, west of the Galapagos, but no part of the eastern
242
North Pacic warm pool. From 19502003, there were eight signicant El Nin o warm events (19571958, 19651966, 19721973, 19821983, 19861987, 19911992, 19971998; and 20022003) and eight recognized La Nin a cold events (19551956, 19641965, 19701971, 19731974, 19751976, 19841985, 19881989, and 19992000) (see Fig. 1a and b). The maximum (minimum) Nin o3 SST anomalies for each warm (cold) event occur during the calendar months from November to January, except for the 19861987 El Nin o event which has double peaks with the major one in the boreal summer. This indicates a robust tendency for the mature phase of El Nin o to occur toward the end of the calendar year (e.g., Rasmusson and Carpenter, 1982; Fig. 1a); the peak phase of La Nin a also occurs in the boreal winter (see Fig. 1b). That is, ENSO is phase-locked to the seasonal cycle. Because of this ENSO phase-locking to the seasonal cycle, oceanographers and meteorologists usually calculate and derive ENSO composites for better understanding of the evolving nature of ENSO. The horizontal patterns of the tropical Pacic sea surface temperature (SST), sea level pressure (SLP), surface wind, and outgoing longwave radiation (OLR) anomalies during the mature phase of El Nin o are shown in Fig. 2 (Wang et al., 1999), by compositing COADS data (Comprehensive OceanAtmosphere Data Set;
Nino3 SST Anomalies

3.0
(a): El Nino
2.0
Nino3 SSTA ( C)
1.0
0.0
57-58 El Nino 65-66 El Nino 72-73 El Nino 82-83 El Nino 86-87 El Nino 91-92 El Nino 97-98 El Nino 02-03 El Nino
-1.0
-2.0 J 2.0 M M J
(b): La Nina
55-56 La Nina 64-65 La Nina 70-71 La Nina 73-74 La Nina 75-76 La Nina 84-85 La Nina 88-89 La Nina 99-00 La Nina
1.0
Nino3 SSTA ( C)
0.0
-1.0
-2.0 J M M J S N J M M J S N
Fig. 1. SST anomalies in the Nin o3 region (5 S5 N, 150 W90 W) from January of the ENSO development year to November the following year for (a) eight El Nin o warm events and (b) eight La Nin a cold events during 19502003. The data are from the NCEP monthly mean SST.
243
Woodru et al., 1987) and NCEP OLR data. During the El Nin o warm phase of ENSO, warm SST and low SLP anomalies are found in the eastern equatorial Pacic, and low OLR anomalies are in the central equatorial Pacic. Associated with the distributions of SST, SLP, and OLR anomalies, zonal wind anomalies are westerly in the central equatorial Pacic, indicating weakened easterly trade winds. During the mature and decay phases, the region of maximum westerly wind anomalies in the central/eastern Pacic is shifted to the south of the equator which may facilitate the El Nin o decay (Harrison and Vecchi, 1999) since it causes a relaxation of westerly anomalies on the equator. ENSO is a basin-scale phenomenon, so it also shows western Pacic patterns in addition to eastern Pacic patterns. During the mature phase of El Nin o, when the warmest SST anomalies are in the eastern equatorial Pacic, the coldest SST anomalies are located to the north and south of the equator in the western Pacic, instead of on the equator. Since atmospheric convection over the western Pacic warm pool shifts into the central equatorial Pacic during the warm phase of ENSO, the region of the lowest OLR anomalies is located to the west of the warmest SST anomalies. Similar to the zonal oset of SST and OLR anomalies in the equatorial eastern and central Pacic, in the western Pacic the o-equatorial region of highest OLR anomalies is positioned west of the o-equatorial region of coldest SST anomalies. The o-equatorial western Pacic cold SST anomalies are also accompanied by o-equatorial western Pacic high SLP anomalies. As a result of o-equatorial high SLP anomalies, easterly wind anomalies appear in the far equatorial western Pacic, as shown in Fig. 2c. Thus, during the mature phase of El Nin o, the equatorial eastern Pacic shows warm SST and low SLP anomalies, and the equatorial central Pacic shows low OLR anomalies, while the o-equatorial western Pacic shows cold SST and high SLP anomalies, and the o-equatorial far western Pacic shows high OLR anomalies. Associated with these SST, SLP, and OLR anomaly patterns are equatorial westerly wind anomalies in the central Pacic and equatorial easterly wind anomalies in the far western Pacic. The nearly out-of-phase behavior between the eastern and western tropical Pacic is also observed during the mature phase of La Nin a, but with anomalies of opposite sign
El Nino Composites
30N 20N 10N
(a): SST Anomalies
30N 20N 10N 0 10S 20S
(c): Wind Anomalies
2.8 m/s
Latitude
0 10S 20S 30S 120E 30N 20N
140E
160E
180
160W
140W
120W
100W
80W
30S 120E 30N 20N 10N 0 10S 20S 30S 120E
140E
160E
180
160W
140W 120W
100W
80W
(b): SLP Anomalies
(d): OLR Anomalies
Latitude
10N 0 10S 20S 30S 120E
140E
160E
180
160W
140W
120W
100W
80W
140E
160E
180
160W
140W 120W
100W
80W
Longitude
Longitude
Fig. 2. El Nin o composites of (a) sea surface temperature (SST) anomalies (C), (b) sea level pressure (SLP) anomalies (mb), (c) surface wind anomalies (m s1), and (d) out-going longwave radiation (OLR) anomalies (W m2).
244
(Fig. 3). Figs. 13 also show asymmetry between El Nin o and La Nin a, with anomalies of El Nin o larger than those of La Nin a. The evolution of El Nin o and La Nin a can be seen in the SST, zonal wind, and 20 C isotherm depth (a proxy for thermocline depth) anomalies of the TAO/TRITON array data along the equator from January 1986 to August 2003 (Fig. 4). The TAO/TRITON moored data show warm events occurring in 19861987, 19911992, 19971998, and 20022003, and cold events in 19881989, 19951996, and 19992000. There is a close relationship between zonal wind anomalies in the western Pacic and thermocline depth anomalies in the central and eastern Pacic (e.g., Kessler et al., 1995). Fig. 4 shows that zonal wind uctuations in the equatorial western Pacic correspond to 20 C isotherm depth signals propagating eastward across the basin at Kelvin wave-like speeds. Remote forcing (from the western Pacic) is clearly important for thermocline and SST anomalies in the central and eastern Pacic (e.g., Kessler and McPhaden, 1995). The importance of the western Pacic can also be seen by comparing the Nin o3 SST anomalies with the indices of the 850-mb zonal wind anomalies in the equatorial western Pacic (5 S5 N, 120 E170 E) and in the equatorial eastern Pacic (5 S5 N, 150 W100 W) from the NCEP-NCAR reanalysis eld, as shown in Fig. 5 (Wang, 2002a). The maximum correlation of 0.56 occurs when the western Pacic zonal wind anomalies lead the Nin o3 SST anomalies by four months, whereas the maximum correlation between the eastern Pacic zonal wind anomalies and the Nin o3 SST anomalies is 0.67 at zero month lag. The correlation relations suggest that the eastern equatorial Pacic is a location of strong oceanatmosphere interaction (Battisti and Hirst, 1989; Cane et al., 1990) and that the western Pacic is an important region for initiating and terminating El Nin o (e.g., McCreary, 1976; Busalacchi and OBrien, 1981; Philander, 1981; Tang and Weisberg, 1984; Philander, 1985; Wang et al., 1999; McPhaden and Yu, 1999; Wang and Weisberg, 2000; Boulanger and Menkes, 2001; Wang, 2001a; Picaut et al., 2002; Boulanger et al., 2003). In addition, the salinity-stratied barrier layer in the western Pacic is hypothesized to inuence the development of El Nin o (Maes et al., 2002).
La Nina Composites
30N 20N 10N
(a): SST Anomalies
30N 20N 10N 0 10S 20S
(c): Wind Anomalies
1.8 m/s
Latitude
0 10S 20S 30S 120E 30N 20N
140E
160E
180
160W
140W
120W
100W
80W
30S 120E 30N 20N 10N 0 10S 20S 30S 120E
140E
160E
180
160W
140W 120W
100W
80W
(b): SLP Anomalies
(d): OLR Anomalies
Latitude
10N 0 10S 20S 30S 120E
140E
160E
180
160W
140W
120W
100W
80W
140E
160E
180
160W
140W 120W
100W
80W
Longitude
Longitude
Fig. 3. La Nin a composites of (a) sea surface temperature (SST) anomalies (C), (b) sea level pressure (SLP) anomalies (mb), (c) surface wind anomalies (m s1), and (d) out-going longwave radiation (OLR) anomalies (W m2).
245
Fig. 4. Time-longitude sections of monthly SST, zonal wind, and 20 C isotherm depth anomalies between 2S to 2 N from January 1986 to August 2003. The data are provided by TAO/TRITON project oce.
Zonal Wind Anomalies

5.0
(a): 850-mb zonal wind anomalies in the west

Nino3 SSTA UA =0.56 (lag: 4 months)
3.0
Nino3 SSTA
UA (m/s)
2.5 0.0 -2.5 -5.0 1950
1.5 0.0 -1.5 -3.0 2000
1955
1960
1965
1970
1975
1980
1985
1990
1995
7.0
(b): 850-mb zonal wind anomalies in the east

Nino3 SSTA UA =0.67 (zero lag)
3.0
Nino3 SSTA
UA (m/s)
3.5 0.0 -3.5 -7.0 1950
1.5 0.0 -1.5 -3.0 2000
1955
1960
1965
1970
1975
1980
1985
1990
1995
Year
Fig. 5. Comparison of the Nin o3 SST anomalies with (a) 850-mb zonal wind anomalies in the equatorial western Pacic (5 S5 N, 120 E170 E), (b) 850-mb zonal wind anomalies in the equatorial eastern Pacic (5 S-5 N, 150 W100 W). SST data are from the NCEP SST and winds are from the NCEP-NCAR reanalysis. All of the time series are three-month running means. The c represents correlation coecient.
By studying the El Nin o events between 1949 and 1976, Rasmusson and Carpenter (1982) showed that the SST anomalies along the South American coast reached peak warming in the boreal spring of the El Nin o year. However, El Nin o events have evolved dierently after 1977 (e.g., Wang, 1995; Trenberth and Stepaniak, 2001). The coastal warmings for the El Nin o events between 1977 and 1996 occurred in the boreal spring subsequent to the El Nin o year rather than in the boreal spring of the El Nin o year (Wang, 1995). The 199798 El Nin o developed in both the central Pacic and the South American coast during the spring of
246
1997 (Wang and Weisberg, 2000). Fig. 6 shows dierent origins and development for the El Nin os between 195076, the El Nin os between 1977 and 1996, the 199798 El Nin o, and the 20022003 El Nin o (Wang and Picaut, 2004). The coastal warming over South America appeared in MarchMay for El Nin os before 1976, but not for the El Nin os between 1977 and 1996. For the 19971998 El Nin o, the initial warming occurred in both the equatorial central Pacic and along the South American coast. The 20022003 El Nin o started and remained in the equatorial central Pacic. Why El Nin os originated in varied ways in the last ve decades is not understood yet, although it may be related to high- or low-frequency variability. ENSO is an interannual phenomenon that superimposes on the seasonal variability. A rough partitioning of global monthly SST variance is illustrated in Fig. 7 (also see Delcroix, 1993; Fiedler, 2002a). Seasonal (1 12 months) variability is high in the eastern boundary currents, the equatorial cold tongue, and at high latitudes. Seasonal variability is low in the equatorial Pacic west of the equatorial cold tongue, in the Gulf of Panama, and in a band north of the cold tongue corresponding to the thermal equator (Fiedler, 2002b; Fiedler and Talley, 2006). Interannual (ENSO) timescale (110 years) variability is high along the equator, and along coastal Ecuador, Peru, and Baja California. Interannual variability is low in the subtropical gyres and in the eastern Pacic warm pool. Interannual variance exceeds seasonal variance between 10 S and 10 N west of 110 W, and also east of the Galapagos. Interannual variability is low in the Atlantic side of the Western Hemisphere warm pool (WHWP), both in absolute magnitude and in relation to seasonal variability.
Different Origin and Development of El Ninos

0
0.3
30N
(a): El Ninos between 1950 and 1976

0.3
0.
30N
0.0
(c): The 1997/98 El Nino

-0.6
-0.6
0.3
0.6
0.0
-0.3
-0.3
-0.3
0.0
-0.3
0.
-0.6
-0.
0.0
0.0
0.3
0.0
0.9
15N
0. 0
15N
0.3
0.
0.0
3
-0.3
-0.3
0.0
0.0
0.0
0.0
0.0
0.
0.9
0
-0. 3
0.0
0.3
0.6
0
0.9
0.6
0.0
0.3
0.6
-0.3
0.9
-0.3
0.0
0.
0.0
0.0
-0.3
0.6
-0.6
0.0
15S
-0.3
15S
-0 .3
0.3
0.0
0.3
-0.3
0.6
-0.3
0.9
0. 6
30S 120E 30N
0.0
150E
0.0
-0.3
180
0.0
150W
120W
90W
0.3
60W
0.0
30S 120E 30N
150E
0.3
0.3
180
1. 2
150W
120W
0.0
90W
0.6
1.2
0.9
0.3
1.2
60W
(b): El Ninos between 1976 and 1996

0.3
0.0
(d): The 2002/03 El Nino

0.0
-0.3
0.0
-0 .3
0.0
0.6
0.3
15N
0.3
-0.3
-0.3
0.0
0.3
15N
0.0
0.0
-0.3
0.3
0.6
0.0
0.0
-0.3 0.0
-0.9
0.3
0.0
-0.3
0. 9
-0
0.0
.3
0.0
0.6
0.0
-0.3
0.0
0.3
0.0
0.0
0.0
0
-0 .3
-0.6
0.9
0.3
0.0
0
0.3
0.0
0.0
0.3
0.9
-0.
-0.6
0.3
.6 -0
15S
15S
0.9
0.6
0.9
0.0 0.3
-0.3
-0
.3
0.6
0.6
0. 3
30S 120E
150E
180
150W
120W
90W
60W
30S 120E
0.3
0.6
150E
180
150W
120W
-0 .3
90W
0.0
0.3
-0.3
0.3
-0.9
0.3
0.6
0.3
0.6
-0.9 -0.3
-0.3
0.3
-0.3
Fig. 6. SST anomaly composites showing dierent origin and development of El Nin os in the last ve decades. (a) El Nin os between 1950 and 1976, (b) El Nin os between 1977 and 1996, (c) the 19971998 El Nin o, and (d) the 20022003 El Nin o. The composites are calculated by averaging the SST anomalies during the development phase of El Nin o (MarchMay of the El Nin o year). Since the 20022003 El Nin o starts earlier, its composite is used the SST anomalies of December 2001 to February 2002.
0. 0
0.0
60W
247
(a) Seasonal SST Variability

60N
3
4
30N
3
1
1.5
0. 5
1.5
1
0.7
0.5
0.2
1. 5
1
2
0. 3 1. 5
2
0.5
1
0.7
1.5
2
30S
1.5
60S
(b) ENSO SST Variability

60N
0.7
0.7
0.5
30N
0.5
0
5 0.
0. 7
1 0.5
0. 5
0. 7
0.5
30S
60S
0.5
(c) ENSO/Seasonal
60N
30N
0. 1
0.1
0.2
0.2
0.5
0
0. 2
2
1
0.2
0.5
30S
0. 1
0.1
0. 1
0. 1
0.1
0.2
0.2
60S
30E
60E
90E
120E
150E
180
150W
120W
90W
60W
30W
30E
Fig. 7. Monthly global SST variability (19502002) from GISST data set (UK Met Oce Global sea Ice Coverage and Sea Surface Temperature v2.3b, Parker et al., 1995). (a) Intra-annual or seasonal variability, estimated as sqrt{Var (SST) Var (SSTA)}, (b) ENSO variability, estimated as sqrt{Var (SSTA)} or the standard deviation of monthly SST anomalies, (c) Ratio of interannual (ENSO) to seasonal variance. Variability on scales greater than 10 years was removed by subtracting a 10-year running mean from the monthly time series (SST).
248
3. Mechanisms of ENSO The eastern tropical Pacic is a region that can both involve local oceanatmosphere interaction and be remotely aected by processes in the western Pacic, because the absence of a Coriolis eect causes the equatorial ocean to act as a waveguide (Gill, 1982). Bjerknes (1969) rst hypothesized that interaction between the atmosphere and the equatorial eastern Pacic Ocean causes El Nin o. In Bjerknes view an initial positive SST anomaly in the equatorial eastern Pacic reduces the eastwest SST gradient and hence the strength of the Walker circulation, resulting in weaker trade winds around the equator. The weaker trade winds in turn drive the ocean circulation changes that further reinforce SST anomaly. This positive oceanatmosphere feedback leads the equatorial Pacic to a never-ending warm state. A negative feedback is needed to turn the coupled oceanatmosphere system around. However, during that time, it was unknown what causes a turnabout from a warm phase to a cold phase, which is commonly referred to as La Nin a (Philander, 1985, 1990). Wyrtki (1975) noticed that the local southeast winds o Peru are not abnormally weak during times of El Nin o (they can even be stronger, which are upwelling-favorable), but the thermocline deepens and SST warms along the South American coast. This led him to search for other ways to explain El Nin o. Using the equatorial wind and sea level data, Wyrtki (1975) proposed a buildup mechanism for El Nin o. He suggested that prior to El Nin o, the trade winds along the equator strengthened, and there was a buildup in heat content and a consequent rise in sea level in the western Pacic warm pool. A trigger for El Nin o is a rapid collapse of the easterly trade wind, which remained unexplained since Wyrtki (1975) was not addressing the coupled problem. When the easterly trade wind rapidly decreases, the accumulated warm water in the western Pacic would collapse and surge eastward in the form of Kelvin waves to initiate a warm El Nin o event. He concluded that El Nin o warming along the South American coast is not the result of local change in surface wind forcing, but rather reects a remote response to a rapid decrease in the easterly trade wind. Wyrtkis mechanism addressed the question of the onset of the coastal El Nin o warming over South America and emphasized that an El Nin o is forced by an abrupt change in zonal wind stress, diering from Bjerknes (1969) hypothesis of a slowly evolving warming due to oceanatmosphere interaction. Wyrtkis idea was supported by simple model studies of Hurlburt et al. (1976), McCreary (1976), and Busalacchi and OBrien (1981). However, Wyrtkis buildup mechanism is not consistent with the development of the 19821983 El Nin o that showed no rise in sea level in the western Pacic and no intensication of the easterly trade winds before 1982, and no initial warming o the west coast of South America in 1982 (e.g., Fig. 6). Failure to recognize the 19821983 El Nin o (the strongest in over a hundred years) until it was well developed motivated the scientic community to intensively study ENSO. Since the 1980s, our understanding of ENSO has been greatly advanced (e.g., see ENSO reviews of Philander, 1990; Neelin et al., 1998; Wang and Picaut, 2004). The theoretical explanations of El Nin o can be summarized as two views. First, El Nin o is one phase of a self-sustained, naturally oscillatory mode of the coupled oceanatmosphere system. Second, El Nin o is a stable (damped) mode triggered by stochastic forcing (e.g., random noise). Whatever the case, El Nin o begins with warm SST anomalies in the equatorial central and eastern Pacic. After an El Nin o reaches its mature phase, negative feedbacks are required to terminate growth of the mature El Nin o anomalies in the central and eastern Pacic. That is, negative feedbacks are needed to turn the coupled system from a warm phase to a cold phase. In search of necessary negative feedbacks for the coupled system, four conceptual ENSO oscillator models have been proposed: the delayed oscillator (Suarez and Schopf, 1988; Graham and White, 1988; Battisti and Hirst, 1989; Cane et al., 1990), the recharge oscillator (Jin, 1997a,b), the western Pacic oscillator (Weisberg and Wang, 1997b; Wang et al., 1999), and the advective-reective oscillator (Picaut et al., 1997). These oscillator models respectively emphasized the negative feedbacks of reected Kelvin waves at the ocean western boundary, a discharge process due to Sverdrup transport, western Pacic windforced Kelvin waves, and anomalous zonal advection. These negative feedbacks may work together for terminating El Nin o warming, as suggested by the unied oscillator (Wang, 2001a). 3.1. The delayed oscillator A mechanism for the oscillatory nature of ENSO was originally proposed by McCreary (1983), based on the reection of subtropical oceanic upwelling Rossby waves at the western boundary. McCreary (1983)
249
explored shallow water ocean dynamics coupled to wind stress patterns that are changed by a discontinuous switch depending on thermocline depth, and he demonstrated how oceanic Rossby waves might be involved in generating the low-frequency oscillations associated with ENSO. Suarez and Schopf (1988) introduced the conceptual delayed oscillator as a candidate mechanism for ENSO, by considering the eects of equatorially trapped oceanic wave propagation. Based on the coupled oceanatmosphere model of Zebiak and Cane (1987), Battisti and Hirst (1989) formulated and derived a version of the Suarez and Schopf (1988) conceptual delayed oscillator model and argued that this delayed oscillator model could account for important aspects of the numerical model of Zebiak and Cane (1987). Graham and White (1988) presented observational evidence of o-equatorial Rossby waves and their reection at the western boundary and then empirically constructed a conceptual oscillator model for ENSO. As shown in McCreary and Anderson (1991), the conceptual equations of the Graham and White model can be reduced to a single equation that has similar form to the delayed oscillator (also see the comments of Neelin et al. (1998)). The schematic diagram of the delayed oscillator is shown in Fig. 8. The positive oceanatmosphere feedback occurs in the equatorial eastern Pacic (e.g., in the Nin o3 region), leading the Nin o3 SST anomaly to a warm state. The delayed negative feedback is by free Rossby waves generated in the eastern Pacic coupling region that propagate to and reect from the western boundary, returning as Kelvin waves to reverse the Nin o3 SST anomalies in the eastern Pacic coupling region. The delayed oscillator assumes that the western Pacic is an inactive region for airsea interaction and that ocean eastern boundary wave reection is unimportant, emphasizing the importance of wave reection at the western ocean boundary. The role of the eastern tropical Pacic in this model is initiation of an El Nin o by airsea coupling. 3.2. The recharge oscillator Wyrtki (1975) rst suggested a buildup in the western Pacic of warm water as a necessary precondition to the development of El Nin o. This concept was later modied to involve the entire tropical Pacic Ocean between 15 S and 15 N (Wyrtki, 1985). Prior to El Nin o upper ocean heat content or warm water volume over the entire tropical Pacic tends to build up (or charge) gradually, and during El Nin o warm water is ushed toward (or discharged to) higher latitudes. After the discharge, the tropical Pacic becomes cold (La Nin a) and then warm water slowly builds up again (recharge) before occurrence of next El Nin o. The recharge and discharge processes have also been examined by Cane et al. (1986), Zebiak (1989), Miller and Cheney (1990), and Springer et al. (1990). Holland and Mitchum (2003) showed that interannual volume redistribution within the tropical Pacic is much greater than volume loss for the tropical Pacic as a whole.
The Delayed Oscillator
Upwelling Rossby Waves Equator
Nino3 Region
Upwelling Kelvin Waves
Fig. 8. Schematic diagram of the delayed oscillator for ENSO. The delayed oscillator model has a positive feedback and a negative feedback, assuming that the western Pacic is inactive for airsea interaction and wave reection in the ocean eastern boundary is unimportant. The positive feedback is represented by local oceanatmosphere coupling in the equatorial eastern Pacic (for example, in the Nin o3 region). The delayed negative feedback is represented by free Rossby waves generated in the eastern Pacic coupling region that propagate to and reect from the western boundary, returning as Kelvin waves to reverse the anomalies in the eastern Pacic coupling region. Thus, the coupled oceanatmosphere system oscillates on interannual time scales.
250
The concept of the recharge and discharge processes is further emphasized by Jin (1997a,b). Based on the coupled model of Zebiak and Cane (1987), Jin (1997a,b) formulated and derived the recharge oscillator model. During the warm phase of ENSO, the divergence of Sverdrup transport associated with equatorial central Pacic westerly wind anomalies and equatorial eastern Pacic warm SST anomalies results in the discharge of equatorial heat content (Fig. 9). The discharge of equatorial heat content leads to a transition phase in which the entire equatorial Pacic thermocline depth is anomalously shallow due to the discharge of equatorial heat content. This anomalous shallow thermocline at the transition phase allows anomalous cold waters to be pumped into the surface layer by climatological upwelling, leading to the cold phase. The converse occurs during the cold phase of ENSO. It is the rechargedischarge process that makes the coupled oceanatmosphere system oscillate on interannual time scales. The role of the eastern tropical Pacic in this model is primarily as one extreme of the tropical Pacic redistribution of heat and mass. 3.3. The western Pacic oscillator Observations show that ENSO displays both eastern and western Pacic interannual anomaly patterns (e.g., Rasmusson and Carpenter, 1982; Mayer and Weisberg, 1998; Wang et al., 1999; Wang and Weisberg, 2000; Figs. 2 and 3). During the warm phase of ENSO, warm SST anomalies in the equatorial eastern Pacic are accompanied by cold SST and shallow thermocline depth anomalies in the o-equatorial western Pacic. Also, while the zonal wind anomalies over the equatorial central Pacic are westerly, those over the equatorial western Pacic are easterly. Consistent with these observations, Weisberg and Wang (1997b) developed a conceptual western Pacic oscillator model for ENSO. This model emphasizes the role of the western Pacic in ENSO that has been overlooked in the delayed oscillator. In particular, o-equatorial SST anomalies (and o-
Depth anomaly
a
Sverdrup Transpor t SST (+)

a
II
a ~ 0
SST ~ 0
a
EQ Depth anomaly
EQ
IV
a ~ 0
Depth anomaly SSTa ~ 0
III
a
EQ Depth anomaly
Sverdrup Transpor t SSTa( )
EQ
Fig. 9. Schematic diagram of the recharge oscillator for ENSO. The four phases of the recharge oscillation are: (I) the warm phase, (II) the warm to cold transition phase, (III) the cold phase, and (IV) the cold to warm transition phase. During the warm phase of ENSO, the divergence of Sverdrup transport associated with equatorial central Pacic westerly wind anomalies and equatorial eastern Pacic warm SST anomalies results in the discharge of equatorial heat content. The discharge of equatorial heat content leads to a transition phase in which the entire equatorial Pacic thermocline depth is anomalously shallow due to the discharge of equatorial heat content. This anomalous shallow thermocline in the transition phase allows anomalously cold waters to be pumped into the surface layer by climatological upwelling and then leads to the cold phase. The converse occurs during the cold phase of ENSO.
251
equatorial anomalous anticyclones) in the western Pacic induce equatorial western Pacic wind anomalies that aect the evolution of ENSO. Arguing from the vantage point of a Gill (1980) atmosphere, condensation heating due to convection in the equatorial central Pacic (Deser and Wallace, 1990; Zebiak, 1990; Weisberg and Wang, 1997a) induces a pair of o-equatorial cyclones with westerly wind anomalies on the equator (Fig. 10). These equatorial westerly wind anomalies act to deepen the thermocline and increase SST in the equatorial eastern Pacic, thereby providing a positive feedback for anomaly growth. On the other hand, the o-equatorial cyclones raise the thermocline there via Ekman pumping. Thus, a shallow o-equatorial thermocline anomaly expands over the western Pacic leading to a decrease in SST and an increase in SLP in the o-equatorial western Pacic (e.g., Wang et al., 1999; Wang, 2000). During the mature phase of El Nin o, the o-equatorial anomalous anticyclones initiate equatorial easterly wind anomalies in the western Pacic. These equatorial easterly wind anomalies cause upwelling and cooling that proceed eastward as a forced ocean response providing a negative feedback, allowing the coupled oceanatmosphere system to oscillate. The role of the western Pacic windforced Kelvin waves in terminating ENSO variability in the eastern Pacic has been demonstrated by McPhaden and Yu (1999), Delcroix et al. (2000), Boulanger and Menkes (2001), Vialard et al. (2001), Picaut et al. (2002), and Boulanger et al. (2003). As in other oscillator models, the eastern tropical Pacic plays a role in this model of ENSO by supporting a positive oceanatmosphere feedback. 3.4. The advective-reective oscillator Picaut et al. (1996) found that zonal displacement of the oceanic convergence zone at the eastern edge of the western Pacic warm pool is in phase with the Southern Oscillation Index. Based on this nding and the study of Picaut and Delcroix (1995) regarding wave reection, Picaut et al. (1997) proposed a conceptual advectivereective oscillator model for ENSO (Fig. 11). In this conceptual model, they emphasize a positive feedback of zonal currents that advect the western Pacic warm pool toward the east during El Nin o. Three negative feedbacks tending to push the warm pool back to its original position of the western Pacic are: anomalous zonal current associated with wave reection at the western boundary, anomalous zonal current associated with
The Western Pacific Oscillator
Nino6 h H
L Nino4
Nino5 2
C H
Nino3 T
Equator
170W
Fig. 10. Schematic diagram of the western Pacic oscillator for ENSO. This oscillator emphasizes the role of the western Pacic in ENSO which has been overlooked in the delayed oscillator. Arguing from the vantage point of a Gill atmosphere, condensation heating in the central Pacic induces a pair of o-equatorial cyclones with westerly wind anomalies in the Nin o4 region. The Nin o4 westerly wind anomalies act to deepen the thermocline and increase SST in the Nin o3 region, thereby providing a positive feedback for anomaly growth. On the other hand, the o-equatorial cyclones raise the thermocline there via Ekman pumping. Thus, a shallow o-equatorial thermocline anomaly expands over the western Pacic leading to a decrease in SST and an increase in SLP in the Nin o6 region. During the mature phase of El Nin o, the Nin o6 high SLP initiates equatorial easterly wind anomalies in the Nin o5 region. The Nin o5 easterly wind anomalies cause upwelling and cooling that proceed eastward as a forced ocean response providing a negative feedback for the coupled ocean atmosphere system to oscillate on interannual time scales.
252
The Advective-Reflective Oscillator
Kup
Rd
Rup
Kd
Fig. 11. Schematic diagram of the advective-reective oscillator for ENSO. In (a), the thick line represents the eastern edge of the western Pacic warm pool. In (b), the dashed line shows observational mean zonal current between 2 N and 2 S, and the thick line is an idealized zonal current. This oscillator emphasizes a positive feedback of zonal currents that advect the western Pacic warm pool toward the east during El Nin o. Three negative feedbacks tending to push the warm pool back to the western Pacic are: anomalous zonal current associated with wave reection at the western boundary; anomalous zonal current associated with wave reection at the eastern boundary; and mean zonal current converging at the eastern edge of the warm pool. Equatorial westerly wind anomalies in the central Pacic produce upwelling Rossby (downwelling Kelvin) waves that propagate westward (eastward). The upwelling Rossby (downwelling Kelvin) waves reect to upwelling Kelvin (downwelling Rossby) waves after they reach the western (eastern) boundary. Since both the upwelling Kelvin and downwelling Rossby waves have westward zonal currents, they tend to push the warm pool back to its original position of the western Pacic. These negative feedbacks along with the negative feedback of the mean zonal current cause the coupled oceanatmosphere system to oscillate.
wave reection at the eastern boundary, and mean zonal current converging at the eastern edge of the warm pool. During the warm phase of ENSO, equatorial westerly wind anomalies in the central Pacic produce upwelling Rossby and downwelling Kelvin waves that propagate westward and eastward, respectively. The westward propagating upwelling Rossby waves reect as upwelling Kelvin waves after they reach the western boundary, whereas the eastward propagating downwelling Kelvin waves reect as downwelling Rossby waves at the eastern boundary. Since both the upwelling Kelvin and downwelling Rossby waves have westward zonal currents, they tend to push the warm pool back to its original position in the western Pacic. These negative feedbacks along with the negative feedback of the mean zonal current, force the coupled oceanatmosphere system to oscillate. The role of the eastern tropical Pacic in this model is reection of downwelling Rossby waves as one component of this negative feedback. 3.5. The unied oscillator With several dierent conceptual oscillator models capable of producing ENSO-like oscillations, more than one may operate in nature. Motivated by existence of the above oscillator models, Wang (2001a,b) formulated
253
and derived a unied ENSO oscillator model from the dynamics and thermodynamics of the coupled ocean atmosphere system that is similar to the Zebiak and Cane (1987) coupled model. Since ENSO is observed to show both eastern and western Pacic anomaly patterns, this oscillator model is formulated and constructed to consider SST anomalies in the equatorial eastern Pacic, zonal wind stress anomalies in the equatorial central Pacic, thermocline depth anomalies in the o-equatorial western Pacic, and zonal wind stress anomalies in the equatorial western Pacic. This model can oscillate on interannual time scales. The unied oscillator includes the physics of all oscillator models discussed above (Fig. 12). All of the above ENSO oscillator models are special cases of the unied oscillator. As suggested by the unied oscillator, ENSO is a multi-mechanism phenomenon (see Picaut et al. (2002) for observations of dierent ENSO mechanisms) and the relative importance of dierent mechanisms is time-dependent. 3.6. A stable mode triggered by stochastic forcing Another view of ENSO is that El Nin os are a series of discrete warm events punctuating periods of neutral or cold conditions (La Nin as). That is, ENSO can be characterized as a stable (or damped) mode triggered by stochastic atmospheric forcing (e.g., Lau, 1985; Penland and Sardeshmuhk, 1995; Neelin et al., 1998; Moore and Kleeman, 1999; Thompson and Battisti, 2001; Dijkstra and Burgers, 2002; Philander and Fedorov, 2003; Kessler, 2003; Zavala-Garay et al., 2003). This hypothesis proposes that disturbances external to the coupled system are the source of random forcing that drives ENSO. An attractive feature of this hypothesis is that it oers a natural explanation in terms of noise to the irregular behavior of ENSO variability. Since this view of ENSO requires the presence of atmospheric noise, it easily explains why each El Nin o is distinct and El Nin o is so dicult to predict (e.g., Landsea and Kna, 2000; Philander and Fedorov, 2003). The external atmospheric forcing may include the Madden-Julian oscillation and westerly wind bursts (see the review of Lengaigne et al. (2004)), and may even involve explosive volcanism (Adams et al., 2003) although this remains a controversial hypothesis. No matter whether El Nin o is a self-sustained cycle or a stable mode triggered by stochastic forcing, El Nin o begins with warm SST anomalies in the equatorial central and eastern Pacic. After an El Nin o reaches
Fig. 12. Schematic diagram of the unied oscillator for ENSO. Bjerknes positive oceanatmosphere feedback leads the equatorial central/ eastern Pacic to a warm state (El Nin o). Four negative feedbacks, required to turn the warm state around, are (1) reection of Kelvin waves at the ocean western boundary, (2) discharge due to Sverdrup transport, (3) western Pacic wind-forced Kelvin waves, and (4) reection of Rossby waves at the ocean eastern boundary. These negative feedbacks correspond to the delayed oscillator, the recharge oscillator, the western Pacic oscillator, and the advective-reective oscillator. The unied oscillator suggests that all of the four negative feedbacks may work together in terminating El Nin o warming. The four ENSO oscillators are special cases of the unied oscillator.
254
its mature phase, negative feedbacks are required to terminate growth of the mature El Nin o anomalies in the central and eastern Pacic. In other words, the negative feedbacks of the delayed oscillator, the recharge oscillator, the western Pacic oscillator, and the advective-reective oscillator may be still valid for demise of an El Nin o, even if El Nin o is regarded as a stable mode triggered by stochastic forcing. As discussed by Mantua and Battisti (1994), a sequence of independent warm events can still be consistent with delayed oscillator physics, since the termination of an individual El Nin o event still requires negative feedback that can be provided by wave reection at the western boundary. 4. The eastern Pacic warm pool As stated in Section 1, the Western Hemisphere warm pool (WHWP) is dened as the region covered by water warmer than 28.5 C on both the Pacic and Atlantic sides of Central America (Wang and Eneld, 2001, 2003). These are temperatures that have a signicant impact on organized tropical convection (e.g., Graham and Barnett, 1987). The choice of 28.5 C is based not only on limiting the WHWP to a closed region, but also on the fact that the depth of the 28.5 C isotherm is closest to the average mixed layer depth in the WHWP (Wang and Eneld, 2003). The WHWP is separated into two parts by the Central American landmass: the eastern north Pacic warm pool and the Intra-Americas Sea (IAS) warm pool (see Fig. 13). We review here seasonal and interannual variations of the eastern Pacic warm pool. Kessler (2002) and Kessler (2006) provided the ocean circulation patterns in the eastern tropical Pacic. Xie et al. (2005) used high-resolution satellite observations to examine oceanatmosphere interaction over the eastern Pacic warm pool. Amador et al. (2006) reviewed the atmospheric forcing of the eastern tropical Pacic. 4.1. Seasonal cycle The warm pool starts to develop in the eastern north Pacic during the boreal early spring (Fig. 13). By May, the warm pool is well developed in the eastern Pacic, but begins to shrink shoreward and spread along the coast where tropical storms frequently develop in the early summer. This becomes the core region of warm waters, which, in conjunction with strengthening of the intertropical convergence zone (ITCZ), triggers the onset of the Mexican and North American summer monsoons. June is a transition period when the warm pool on the Atlantic side of Central America starts to develop, while the warm pool in the eastern Pacic decays. By July, water warmer than 28.5 C is well developed in the Gulf of Mexico, and the WHWP covers the Gulf of Mexico and to less extent the eastern Pacic. In August, the warm water in the Gulf of Mexico reaches its maximum with a large area covered by water warmer than 29.5 C. By September, the warm pool has expanded south into the Caribbean and eastward into the western tropical North Atlantic, while the water in the Gulf of Mexico has cooled. The WHWP decays quickly after October. The seasonal cycle of the eastern Pacic warm pool is clearly shown in Fig. 14a, which is calculated as the mean NCEP SST (Smith et al., 1996) over the area enclosed by the 28.5 C isotherm. There is no water warmer than 28.5 C in the eastern Pacic during January and February, and the seasonal cycle shows double peaks. The SST in the eastern Pacic warm pool displays two maximum values in May and August. This pattern seems to relate to the midsummer drought discussed by Magan a et al. (1999) (also see Amador et al., 2006). Magan a et al. (1999) showed double peaks in the annual cycle of precipitation over the southern part of Mexico and Central America with maxima during June and September, separated by the so-called the midsummer drought during July/August. Fig. 14a shows that the eastern Pacic warm pool SST leads double peaks of precipitation over the southern part of Mexico and Central America by one month, and that the area of the eastern Pacic warm pool is approximately in phase with the precipitation. Using the NCEP-NCAR reanalysis, the NCEP SST, and the Levitus mixed layer depth data, Fig. 14b shows surface net heat ux and heat storage tendency (qCPhoT/ot) over the eastern Pacic warm pool. The Levitus data show that the mixed layer depth across the warm pool is on average about 25 m (Wang and Eneld, 2003), so we choose h = 25 m for our calculation. The penetration of shortwave radiation follows the recent work of Murtugudde et al. (2002): QP = 0.47QS exp (h/17). Although the heat storage tendency is approximately in phase with the net heat ux, the net heat ux is larger than the heat storage tendency
255
Western Hemisphere Warm Pool (WHWP)

30N
(a): March
18.5
22. 5
24.5
22.5
30N
(e): July
.5 24 .5 26
24.5
26 .5
28.5
20N
22.5 24.5
26.5
20N
26 .5
5 26.
28.5
10N
26.5
10N
26.5
24.5
26
28.5
.5
0 120W 30N
100W
80W
60W
22.5
40W
0 120W 30N
100W
80W
60W
40W
(b): April
24 .5
20.5
(f): August
26 24.5 .5
26.5
24.5
26.5
20N
24.5
5 26.
20N
28.5
26.5
28.5
10N
28.5
10N
24.5
28.5 26.5
0 120W 30N
26.5
100W
18.5
80W
60W
40W
0 120W 30N
22.5
100W
80W
60W
40W
(c): May
26. 5
24.5
(g): September
.5
24 .5
20N
26.
20N
5 26.
28.5
28.5
26
22.5
28
.5
10N
28.5
26.5
10N
24.5
24.5
26.5
0 120W 30N
100W
80W
60W
40W
0 120W 30N
22.5
100W
80W
60W
26. 5
40W
(d): June
20.5
(h): October
24 .5
26
.5
28.5
20N
24.
26.5
20N
26
.5
. 28
28.5
5
28.5
10N
28.5
10N
26.5
26.5
24.5
0 120W
24.5
100W
80W
60W
40W
0 120W
22.5
100W
80W
60W
40W
Fig. 13. Seasonal distributions of SST for the tropical Western Hemisphere warm pool (WHWP): (a) March, (b) April, (c) May, (d) June, (e) July, (f) August, (g) September, and (h) October. The shading and dark contour represent water warmer than 28.5 C. The data are from the NCEP SST.
whether or not heat ux includes the penetration of shortwave radiation. This suggests that processes other than surface ux are needed to cool SST in the eastern Pacic warm pool, such as ocean advection. Notice that the eastern Pacic warm pool is dierent from the IAS warm pool, where surface heat uxes are primarily responsible for seasonal warming (Wang and Eneld, 2003).
256
Eastern Pacific Warm Pool

29.4 29.3 2.0
(a): SST
SST Area
2.5
Area (10 km )
29.2
SST ( C)
29.1 29.0 28.9
1.5 1.0 0.5
6 2
28.8 28.7 J F M A M J J A S O N D 0.0
Month
75 50
(b): HS tendency and net heat flux
Flux (W/m )
25 0 -25
Tendency Net Net (penetration)
-50 J F M A M J J A S O N D
Month
Fig. 14. (a) Seasonal variations of SST over the eastern Pacic warm pool (larger than 28.5 C) and the area enclosed by water warmer than 28.5 C; (b) heat storage (HS) tendency (qCPhoT/ot) of the eastern Pacic warm pool, surface net heat ux (shortwave radiation minus latent heat ux, longwave radiation, and sensible heat ux), and surface net heat ux after considering penetration of shortwave radiation. The data are from the NCEP SST, the NCEP-NCAR reanalysis eld, and the Levitus data.
The eastern Pacic and IAS warm pools are a heat source of summer Hadley circulation and play a key role in the transition from the South American to the North American Monsoon. Fig. 15 shows the boreal summer (July) climatologies of tropospheric circulation patterns. The center of upper tropospheric divergence associated with middle tropospheric ascent is near the region of the WHWP. Two subsiding limbs of the overturning are located over the subtropical western South Atlantic and eastern South Pacic (as manifested by 200-mb divergence and 500-mb descending motion). The Hadley circulation between 110 W70 W involves air rising around 10 N15 N, diverging southward in the upper troposphere, descending in the subtropical South Pacic, then crossing the equator at the surface and returning to the Northern Hemisphere convergent region. The atmospheric circulation patterns are consistent with the results of Bosilovich and Schubert (2002), and Hu and Feng (2001) showing that warm pool is a source of moisture for North America. 4.2. Interannual variability The eastern Pacic warm pool index is calculated for the anomalies of the area enclosed by the 28.5 C isotherm, as shown in Fig. 16a. During the 50-year period since 1950 (from 1950 to 1999), large warm pools occur in 195758, 1969, 1972, 19821983, 1987, 1990199119921993, and 19971998, all corresponding to El Nin o events (Fig. 16b). The maximum correlation between the eastern Pacic warm pool area anomalies and the Nin o3 SST anomalies is 0.75 at zero lag. The horizontal structure of the eastern Pacic warm pool during the warm events can be examined by comparing with the climatological warm pool. Since the largest climatological eastern Pacic warm pool occurs in May, we composite the structure of the eastern Pacic warm pool based on the May SST values of 1957, 1969, 1972, 1983, 1987, 1992, and 1997. The result is shown in Fig. 17. As expected, Fig. 17 shows that the size of the
257
Climatologies (July)
30N
(a): vel. potential and div. wind (200 mb)
20N
10N
10S
20S 120W 30N
100W
5.1 m/s
80W
60W
40W
20W
(b): vertical velocity (500 mb)

-2
-2
-2
-2
20N
0
6
0
-2
2
6
10N
2
0
-2
2
0
0
-2
-2
-4
10S
-4
-4
-4
-4
20S 120W 100mb 250mb 400mb 550mb 700mb 850mb 1000mb 20S
100W
80W
60W
40W
20W
(c): Hadley circulation (110W-70W)
10S
3.0 m/s (10 mb/s)
-4
10N
20N
30N
Fig. 15. The boreal summer (July) climatologies of tropospheric circulation patterns. (a) 200-mb velocity potential (106 m2/s) and divergent wind (m/s), (b) 500-mb vertical velocity (104 mb/s), and (c) meridional-vertical circulation by averaging divergent wind and vertical velocity between 110 W70 W. The vertical velocity is taken as the negative of the pressure vertical velocity in the reanalysis, i.e., positive values indicate an upward movement of air parcels. Positive values are shaded.
eastern Pacic warm pool for the interannual warm events is much larger than the climatological warm pool (dark contour). All of these suggest that interannual variability of the eastern Pacic warm pool is clearly related to the El Nin o events. What mechanism controls interannual variability of the eastern Pacic warm pool? Like the seasonal cycle, the mechanisms for controlling interannual variability of the eastern Pacic and IAS warm pools seem to be dierent. Wang (2002b) and Wang and Eneld (2003) showed that during winters preceding large IAS warm pools, there is a strong weakening of the Hadley circulation that serves as a tropospheric bridge for
258
Interannual Variability
150.0
(a): Index of eastern Pacific warm pool
Area Anomaly (%)
120.0 90.0 60.0 30.0 0.0 -30.0 -60.0 1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
3.0
(b): Nino3 SST anomalies
SST Anomaly ( C)
2.0 1.0 0.0 -1.0 -2.0 1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year
Fig. 16. (a) Eastern Pacic warm pool area anomalies (%) for SST warmer than 28.5 C, and (b) SST anomalies in the Nin o3 region (5 S 5 N, 150 W90 W). The eastern Pacic warm pool area anomalies (in unit of percentage) are calculated as anomalies of area (P28.5 C) divided by the May climatological warm pool area.
Warm Events of Eastern Pacific Warm Pool

20N
22.5
24.5
15N
26.5
Latitude
10N
28 .5
5N
28.5
28.5
0 140W
130W
120W
110W
100W
90W
80W
70W
Longitude
Fig. 17. SST composite for the eastern Pacic warm pool interannual warm events. The composites are calculated by averaging the May SST values over the warm years of 1957, 1969, 1972, 1983, 1987, 1992, and 1997. The shading represents water warmer than 28.5 C. The dark contour is climatological May eastern Pacic warm pool (SST warmer than 28.5 C).
transferring Pacic El Nin o eects to the Atlantic sector and inducing warming of IAS warm pool. After the ENSO-related warming, a positive oceanatmosphere feedback operating through longwave radiation is responsible for maintaining or amplifying SST anomaly growth in the region of the IAS warm pool.
259
Associated with the warm SST anomalies is a decrease in SLP anomalies and an anomalous increase in atmospheric convection and cloudiness. The increase in convective activity and cloudiness results in less longwave radiation loss from the sea surface, which then reinforces SST anomalies. However, this mechanism does not seem to operate in the eastern Pacic warm pool. We calculated correlations among variables in the eastern Pacic warm pool. The eastern Pacic warm pool longwave radiation anomalies are not signicantly correlated with the eastern Pacic warm pool SST anomalies, and the eastern Pacic warm pool SLP anomalies are not correlated with the eastern Pacic warm pool cloud anomalies either. All of these suggest that the positive oceanatmosphere feedback operating through longwave radiation and associated cloudiness is not responsible for the warmings of the eastern Pacic warm pool on interannual scales. Notice that the eastern Pacic warm pool shortwave radiation and latent heat ux anomalies are not signicantly correlated with the eastern Pacic warm pool SST anomalies. Interannually, the eastern Pacic warm pool may be due to a direct oceanic connection to the equator, associated with extreme ENSO phases. 5. ENSO biological and ecological variability ENSO-related changes in winds, insolation, hydrography and circulation in the eastern tropical Pacic are of sucient magnitude and duration to aect organisms, populations and ecosystems. The species and communities of the region have evolved to persist through the quasi-regular disturbances imposed by ENSO events. Thus, typical or even exceptional events should not result in long-term, fundamental changes (Paine et al., 1998). Biological eects of recent El Nin o events in the region have been documented and explained primarily for phytoplankton and commercial sh stocks. For example, Barber and Chavez (1986) summarized the eects of the 19821983 El Nin o in the eastern equatorial and Peru upwelling systems: deepening of the thermocline, and thus the nutricline, resulted in decreased primary production that ultimately aected sur nchez et al. (2000) summarized vival, reproduction, and distribution of higher trophic level organisms. Sa eects of the 19971998 El Nin o on the coastal marine ecosystem o Peru: SST rose by up to 9 C, the thermocline, nutricline and oxycline deepened, nitrate in surface waters decreased, zooplankton volumes decreased, exotic warm-water dinoagellates and copepods increased, and changes in distribution of sh, cetaceans and seabirds were observed. In the eastern equatorial Pacic, sampling by ships of opportunity showed that the 19821983 El Nin o caused a deepening of the thermocline, and a reduction in chlorophyll and copepod abundance (Dessier and Donguy, 1987). Eects of ENSO on phytoplankton and zooplankton in the eastern tropical Pacic are lamo and Fa ndez-A further reviewed in Pennington et al. (2006) and Ferna rber-Lorda (2006), respectively. El Nin o eects on sh in coastal upwelling systems are well-known. The 1972 El Nin o resulted in a recruitment failure for Peruvian anchoveta that, along with overshing, led to a collapse of the worlds largest shery (Clark, 1977). The 19821983 El Nin o had a variety of eects on commercial sh stocks in Peru: hake moved down the continental slope to stay in cooler deep water, shrimp and sardines moved southward so that catches in some areas decreased and in other areas increased, jack mackerel moved inshore in search of euphausiid prey and were subject to high predation mortality there, scallop abundance increased due to enhanced reproductive success in warmer water, and the anchoveta population crashed due to reduced food availability for adults and larvae. Many of these stocks recovered rapidly beginning in late 1983 (Barber and Chavez, 1986). There have been few reports of ENSO eects on animals other than commercially exploited sh in the eastern tropical Pacic, except for the well-known mass mortalities of guano-producing seabirds on the coast of Peru (Wooster, 1960). Seabird populations experienced breeding failures, mass mortalities, and migrations in search of food throughout the tropical Pacic in 19821983 (Ainley et al., 1986), although a few species were not aected. Body weight of Galapagos penguins increased during the 1971 La Nin a and decreased during 1972 El Nin o, indicating short-term response to food availability, but the population suered 77% mortality during the 19821983 El Nin o and had not recovered by 1997 (Dee Boersma, 1998). Blue-footed booby reproductive attempts failed and breeding colonies were abandoned during the 19861987 El Nin o, apparently in response to reduced availability of sardines, but several other species were not aected (Anderson, 1989). Piscivorous seabirds in coastal Peru have consistently experienced adult mortality and decreased reproductive success during El Nin o events; these are short-term population eects, resulting from reduced availability of
260
anchoveta (Tovar et al., 1987; Crawford and Jahncke, 1999). All fteen species of seabirds nesting on the Galapagos Islands stopped breeding or experienced reduced reproductive success during the 19821983 El Nin o, but resumed breeding the following year (Valle et al., 1987). Pelagic seabird surveys have shown changes in the relative abundance of less common species, but not the dominant species, during El Nin o and La Nin a events in the eastern equatorial Pacic (Ribic et al., 1992). Such changes may be explained by shifts in distribution between the equatorial and subtropical water masses covered by these surveys (Ballance et al., 2006). In general, seabirds that forage in equatorial and coastal upwelling areas of the eastern tropical Pacic suer reproductive failure and mortality due to food shortage during El Nin o events. However, other species that forage in areas less aected by El Nin o, for example the warm pool, may be relatively unaected (Ballance et al., 2006). Mortality and other population eects of El Nin o on marine mammals have been observed in coastal ecosystems in the eastern tropical Pacic. Manzanilla (1989) observed a 1983 El Nin o mark in the teeth of mature female Peruvian dusky dolphins and suggested that the mark resulted from low foraging success for the preferred prey, anchoveta, which became unavailable during the 19821983 El Nin o. Consistent with this inference, no such marks were observed in other dolphin species that consume other prey. Galapagos pinnipeds suered great mortality in 19821983, especially in younger year classes, and reduced pup production due to reduced food availability (Trillmich and Limberger, 1985). Peruvian pinnipeds were aected by the reduced availability of anchoveta in 1983 (Majluf and Reyes, 1989) and again in 1998 (Soto et al., 2003). Ramirez (1986) observed diet changes and reduced feeding success of Brydes whales o Peru during 19821983 El Nin o. ENSO eects on oceanic cetaceans are reviewed in Ballance et al. (2006). Biological eects of El Nin o, such as temporary range shifts, may be benign in the sense that normal or average conditions are quickly restored. Eects of the opposite, La Nin a, phase of the ENSO cycle are less well documented, probably because La Nin a events do not bear the negative reputation of El Nin o events. It has been argued that ENSO variability might be responsible for the high production of the Peru coastal upwelling system by disrupting biological controls such as predation and opening loopholes for the success of fecund, small pelagic shes (Bakun and Broad, 2003). ENSO variability is also subject to low-frequency modulation. See Mantua and Hare (2002) and Miller et al. (2003) for recent reviews of Pacic decadal variability and its eects on ocean ecosystems. 6. Changes in ENSO variability Changes in the characteristics, or modulation, of ENSO variability over the past one to two centuries have been described in instrumental and proxy records from the tropical Pacic. Mestas-Nun ez and Eneld (2001) found that the late 1970s climate shift that warmed the eastern equatorial Pacic (Nin o3 region) by about 0.5 C was also characterized by increased interannual variance through the 1980s and 1990s. An 1893 1994 coral record from Clipperton Atoll (within the eastern Pacic warm pool area) shows both ENSO and decadal-scale variability closely related to instrumental Southern Oscillation Index (SOI) and Pacic Decadal Oscillation (PDO) records from recent years (Linsley et al., 2000). Reduced ENSO variability is evident between 1925 and the mid-1940s. An and Wang (2000) found that the dominant period of SST variability in the central and eastern equatorial Pacic (Nin o-3.4 region) shifted from 3.3 years during 19671973 to 4.2 years during 19801993. Setoh et al. (1999) found a similar increase in the period of ENSO in the equatorial Pacic during the late 1970s, but also found slight changes in the spatial pattern of the ENSO signal. Spectral analysis of a composite record of ENSO events since 622 A.D. showed that the period of ENSO variability has varied, within a range of 1.510 years, in cycles of 90, 50, and 23 years (Anderson, 1992). Philander and Fedorov (2003) suggest that there are multiple modes of ENSO, all weakly damped oscillations sustained by random disturbances, depending on the background state of the tropical Pacic Oceanatmosphere system. The irregularity of ENSO is obvious when the magnitude and spacing of individual events are considered in an instrumental time series (cf. Fig. 16b). Wunsch (1999) warns, however, that before concluding that one is seeing evidence for trends, shifts in the mean, or changes in oscillation periods, one must rule out the purely random uctuations expected from stationary time series. Solow and Huppert (2003), for example, found no signicant nonstationarity in 18762002 records of Darwin sea level pressure and Nin o3 SST. Rajagopalan et al. (1997) show that the anomalous prolonged El Nin o of 19901995 was within the natural variability
261
of the stationary time series of the preceding century. An as-yet unpublished study by Mitchell and Wallace1 shows that spatial and temporal patterns of ENSO-related anomalies in global wind, temperature and precipitation elds have changed very little since the late 19th century. Fedorov and Philander (2000) argued that apparent changes in the characteristics of ENSO variability may simply reect decadal-scale changes in the background state (climatology) against which El Nin o and La Nin a are measured. Characteristics and mechanisms of scales of variability longer than ENSO are reviewed in Wang and Picaut (2004) and Mestas-Nun ez and Miller (2006). 7. Summary ENSO shows an interannual variability in both the eastern and western tropical Pacic. During the warm phase of ENSO, warm SST and low SLP anomalies in the equatorial eastern Pacic and low OLR anomalies in the equatorial central Pacic are accompanied by cold SST and high SLP anomalies in the o-equatorial western Pacic and high OLR anomalies in the o-equatorial western Pacic. The o-equatorial anomalous anticyclones in the western Pacic initiate and produce equatorial easterly wind anomalies over the western Pacic. So, while the zonal wind anomalies over the equatorial central Pacic are westerly, those over the equatorial western Pacic are easterly. ENSO is observed to be phase-locked to the seasonal cycle. That is, the maximum (minimum) SST anomalies in the equatorial eastern and central Pacic for each El Nin o warm (La Nin a cold) event occur during the boreal winter. In spite of this, the El Nin o warm events over the last 50 years have evolved dierently. For the El Nin o events between 1949 and 1976, anomalous surface warming occurs rst o the South American coast, and then progresses westward along the equator into the eastern and central Pacic. The initial warming of the El Nin o events between 1976 and 1996 occurred in the equatorial central Pacic. The 19971998 El Nin o developed in both the equatorial central Pacic and the South American coast during the spring of 1997. The 20022003 El Nin o started in the equatorial central Pacic. However, a common feature for all El Nin o events since 1950 is that westerly wind anomalies in the equatorial western Pacic always led the Nin o3 SST anomalies by about four months, suggesting that the western Pacic is an important region for ENSO variability in the eastern tropical Pacic. ENSO has been viewed as a self-sustained, naturally oscillatory mode or a stable mode triggered by stochastic forcing. For both views, ENSO involves the positive oceanatmosphere feedback over the eastern tropical Pacic hypothesized by Bjerknes in the 1960s. After an El Nin o reaches its peak, a negative feedback is required for terminating a continued growth of El Nin o. Dierent negative feedbacks have been proposed since the 1980s associated with a delayed oscillator, a recharge oscillator, a western Pacic oscillator, and an advective-reective oscillator. The delayed oscillator assumes that wave reection at the ocean western boundary provides a negative feedback for the coupled system oscillation. The recharge oscillator argues that discharge and recharge of equatorial heat content cause the coupled system to oscillate. The western Pacic oscillator emphasizes equatorial wind in the western Pacic that provides a negative feedback for the coupled system. The advective-reective oscillator emphasizes the importance of zonal advection associated with wave reection at both the western and eastern boundaries and of the mean zonal current. As suggested by the unied oscillator, all of these negative feedbacks work together to terminate El Nin os, and their relative importance is time-dependent. The eastern Pacic warm pool, dened by surface water warmer than 28.5 C, is in the region of the eastern tropical Pacic north of the equator. The seasonal SST of the eastern Pacic warm pool shows double maximum values in May and August that lead the double peaks of precipitation over the southern part of Mexico and Central America by one month. The precipitation peaks are separated by the midsummer drought. Seasonally, ocean advection seems to play an important role for SST variations. Interannually, the eastern Pacic warm pool may be remotely forced by ENSO variability along the equator. ENSO variability has been shown to aect phytoplankton, invertebrates, sh, birds, pinnipeds and cetaceans. Almost all observations of such environmental eects have been on coastal or island populations. While
Mitchell, T.P., and Wallace, J.M. The instrumental record of ENSO: 18402000. http://jisao.washington.edu/wallace/ enso_chronology.pdf.
262
it is true that ENSO eects are more extreme in highly productive coastal environments, other factors may be important. Perhaps such populations are less adaptable or opportunistic. Certainly, these populations are more readily accessible for long-term study. El Nin o events often cause changes in distribution of species as the distributions of preferred water masses and prey changes. Population eects are observed on local breeding grounds, but recovery usually occurs rapidly when the El Nin o event is over. Populations of marine animals in the eastern tropical Pacic have undoubtedly evolved life history strategies to deal with ENSOrelated variability in their environment. Acknowledgements CW thanks David Eneld for discussions of historical background of El Nin o. We thank anonymous reviewers comments and Charles Millers editorial comments. Fig. 4 was provided by TAO project oce. This work was supported by a grant from National Oceanic and Atmospheric Administration (NOAA) Oce of Global Programs and by the base funding of NOAA Atlantic Oceanographic and Meteorological Laboratory (AOML). The ndings and conclusions in this report are those of the author(s) and do not necessarily represent the views of the funding agency. This is a contribution to the scientic agenda of the Eastern Pacic Consortium of the InterAmerican Institute for Global Change Research. Support was also provided by the Protected Resources Division of NOAA Fisheries, Southwest Fisheries Science Center. References
Adams, J.B., Mann, M.E., Ammann, C.M., 2003. Proxy evidence for an El Nin o-like response to volcanic forcing. Nature 426, 274278. Ainley, D.G., Carter, H.R., Anderson, D.W., Briggs, K.T., Coulter, M.C., Cruz, F., Cruz, J.B., Valle, C.A., Fefer, S.I., Hatch, S.A., Schreiber, E.A., Schreiber, R.W., Smith, N.G., 1986. Eects of the 19821983 El Nin o-Southern Oscillation on Pacic Ocean bird populations. In: Onelet, H. (Ed.), Proceedings XIX International Ornithological Congress. National Museum of Natural Sciences, Ottawa, pp. 17471756. Amador, J.A., Alfaro, E.J., Lizano, O.G., Magan a, V.O., 2006. Atmospheric forcing of the eastern tropical Pacic: A review. Progress in Oceanography 69 (24), 101142. An, S.-I., Wang, B., 2000. Interdecadal change of the structure of the ENSO mode and its impact on the ENSO frequency. Journal of Climate 13, 20442055. pagos to the 198687 El Nin Anderson, D.J., 1989. Dierential responses of boobies and other seabirds in the Gala o-Southern Oscillation event. Marine Ecology Progress Series 52, 209216. ` o events. In: Diaz, H.F., Markgraf, V. (Eds.), El Nin `o Anderson, R.Y., 1992. Long-term changes in the frequency of occurrence of El Nin and the Southern Oscillation: Multiscale Variability and Global and Regional Impacts. Cambridge University Press, Cambridge, pp. 193200. Bakun, A., Broad, K., 2003. Environmental loopholes and sh population dynamics: comparative pattern recognition with focus on El Nin o eects in the Pacic. Fisheries Oceanography 12, 458473. Ballance, L.T., Pitman, R.L., Fiedler, P.C., 2006. Oceanographic inuences on seabirds and cetaceans in the eastern tropical Pacic: A review. Progress in Oceanography 69 (24), 360390. Barber, R.T., Chavez, F.P., 1986. Ocean variability in relation to living resources during the 198283 El Nin o. Nature 319, 279285. Battisti, D.S., Hirst, A.C., 1989. Interannual variability in the tropical atmosphere-ocean model: inuence of the basic state, ocean geometry and nonlinearity. Journal of the Atmospheric Sciences 45, 16871712. Berlage, H.P., 1957. Fluctuations in the general atmospheric circulation of more than one year, their nature and prognostic value. Koninklijk Nederlands Meteorologisch Instituut, Mededelingen en verhandelingen 69, 1152. Berlage, H.P., 1966. The Southern Oscillation and world weather. Koninklijk Nederlands Meteorologisch Instituut, Mededelingen en verhandelingen 88, 1152. Bjerknes, J., 1966. A possible response of the atmospheric Hadley circulation to equatorial anomalies of ocean temperature. Tellus 18, 820829. Bjerknes, J., 1969. Atmospheric teleconnections from the equatorial Pacic. Monthly Weather Review 97, 163172. Bosilovich, M.G., Schubert, S.D., 2002. Water vapor tracers as diagnostics of the regional hydrologic cycle. Journal of Hydrometeorology 3, 149165. Boulanger, J.P., Menkes, C., 2001. The trident Pacic model. Part 2: role of long equatorial wave reection on sea surface temperature anomalies during the 19931998 TOPEX/POSEIDON period. Climate Dynamics 17, 175186. Boulanger, J.P., Cravatte, S., Menkes, C., 2003. Reected and locally wind forced interannual Kelvin waves in the western Pacic Ocean. Journal of Geophysical Research 108 (C10), 3311. doi:10.1029/2002JC001760. Busalacchi, A., OBrien, J.J., 1981. Interannual variability of the equatorial Pacic in the 1960s. Journal of Geophysical Research 86, 1090110907. Cane, M.A., Zebiak, S.E., Dolan, S.C., 1986. Experimental forecasts of El Nin o. Nature 321, 827832.
263
Cane, M.A., Matthias, M., Zebiak, S.E., 1990. A study of self-excited oscillations of the tropical oceanatmosphere system. Part I: Linear analysis. Journal of the Atmospheric Sciences 47, 15621577. Clark, W.G., 1977. The lessons of the Peruvian anchoveta shery. California Cooperative Oceanic Fisheries Investigations Reports 19, 57 63. Crawford, R.J.M., Jahncke, J., 1999. Comparison of trends in abundance of guano-producing seabirds in Peru and southern Africa. South African Journal of Marine Science 21, 145156. pagos penguin: impacts of El Nin Dee Boersma, P., 1998. Population trends of the Gala o and La Nin a. Condor 100, 245253. Delcroix, T., 1993. Seasonal and interannual variability of sea surface temperature in the tropical Pacic, 19611991. Deep-Sea Research I 40, 22172228. Delcroix, T., Dewitte, B., duPenhoat, Y., Masia, F., Picaut, J., 2000. Equatorial waves and warm pool displacements during the 19921998 El Nin o Southern Oscillation events: Observation and modeling. Journal of Geophysical Research 105, 2604526062. Deser, C., Wallace, J.M., 1990. Large-scale atmospheric circulation features of warm and cold episodes in the tropical Pacic. Journal of Climate 3, 12541281. Dessier, A., Donguy, J.R., 1987. Response to El Nin o signals of the epiplanktonic copepod populations in the eastern tropical Pacic. Journal of Geophysical Research 92 (C13), 1439314403. Diaz, H.F., Markgraf, F., 1992. El Nin o: Historical and Paleoclimatic Aspects of the Southern Oscillation. Cambridge University Press, Cambridge. Diaz, H.F., Markgraf, F., 2000. El Nin o and Southern Oscillation: Multiscale variability and Global and Regional Impacts. Cambridge University Press, Cambridge. Dijkstra, H.A., Burgers, G., 2002. Fluid Dynamics of El Nin o variability. Annual Review of Fluid Mechanics 34, 531558. Eguiguren, D.V., 1894. Las lluivas de Piura. Boletin de la Sociedad Geograca de Lima 4, 241258. Eneld, D.B., 1989. El Nin o, past and present. Reviews of Geophysics 27, 159187. Fedorov, A.V., Philander, S.G., 2000. Is El Nin o changing? Science 288, 19972002. lamo, M.A., Fa ndez-A Ferna rber-Lorda, J., 2006. Zooplankton and the oceanography of the eastern tropical Pacic: A review. Progress in Oceanography 69 (24), 318359. Fiedler, P.C., 2002a. Environmental change in the eastern tropical Pacic Ocean: Review of ENSO and decadal variability. Marine Ecology Progress Series 244, 265283. Fiedler, P.C., 2002b. The annual cycle and biological eects of the Costa Rica Dome. Deep-Sea Research 49, 321338. Fiedler, P.C., Talley, L.D., 2006. Hydrography of the eastern tropical Pacic: A review. Progress in Oceanography 69 (24), 143180. Gill, A.E., 1980. Some simple solutions for heat-induced tropical circulation. Quarterly Journal of Royal Meteorological Society 106, 447 462. Gill, A.E., 1982. AtmosphereOcean Dynamics. Academic Press, Orlando. Graham, N.E., Barnett, T.P., 1987. Sea surface temperature, surface wind divergence, and convection over tropical oceans. Science 238, 657659. Graham, N.E., White, W.B., 1988. The El Nin o cycle: A natural oscillator of the Pacic Oceanatmosphere system. Science 24, 12931302. Harrison, D.E., Vecchi, G.A., 1999. On the termination of El Nin o. Geophysical Research Letters 26, 15931596. Hayes, S.P., Mangum, L., Picaut, J., Sumi, J.A., Takeuchi, K., 1991. TOGA-TAO: A moored array for real-time measurements in the tropical Pacic Ocean. Bulletin of the American Meteorological Society 72, 339347. Holland, C.L., Mitchum, G.T., 2003. Interannual volume variability in the tropical Pacic. Journal of Geophysical Research 108, 3369. doi:10.1029/2003JC001835. Hu, Q., Feng, S., 2001. Climate role of the southerly ow from the Gulf of Mexico in interannual variations in summer rainfall in the Central United States. Journal of Climate 14, 31563170. Hurlburt, H.E., Kindle, J.C., OBrien, J.J., 1976. A numerical simulation of the onset of El Nin o. Journal of Physical Oceanography 6, 621631. Jin, F.F., 1997a. An equatorial ocean recharge paradigm for ENSO. Part I: Conceptual model. Journal of the Atmospheric Sciences 54, 811829. Jin, F.F., 1997b. An equatorial ocean recharge paradigm for ENSO. Part II: A stripped-down coupled model. Journal of the Atmospheric Sciences 54, 830847. Kessler, W.S., 2002. Mean three-dimensional circulation in the Northeast tropical Pacic. Journal of Physical Oceanography 32, 2457 2471. Kessler, W.S., 2003. Is ENSO a cycle or a series of events? Geophysical Research Letters 29 (23), 2125. doi:10.1029/2002GL015924. Kessler, W.S., 2006. The circulation of the eastern tropical Pacic: A review. Progress in Oceanography 69 (24), 181217. Kessler, W.S., McPhaden, M.J., 1995. Oceanic equatorial waves and the 199193 El Nin o. Journal of Climate 8, 17571774. Kessler, W.S., McPhaden, M.J., Weickmann, K.M., 1995. Forcing of intraseasonal Kelvin waves in the equatorial Pacic. Journal of Geophysical Research 100, 1061310631. Landsea, C.W., Kna, J.A., 2000. How much skill was there in forecasting the very strong 19971998 El Nin o? Bulletin of the American Meteorological Society 81, 21072119. Lau, K.M., 1985. Elements of a stochastic-dynamical theory of the long-term variability of the El Nin o-Southern Oscillation. Journal of the Atmospheric Sciences 42, 15521558. Lengaigne, M., Boulanger, J., Menkes, C., Delecluse, P., Slingo, J., 2004. Westerly wind events and their inuence on the coupled ocean atmosphere system: A review. In: Wang, C., Xie, S.-P., Carton, J. (Eds.), Earths Climate: The OceanAtmosphere Interaction. American Geophysical Union, pp. 4969.
264
Linsley, B.K., Ren, L., Dunbar, R.B., Howe, S.S., 2000. El Nin o Southern Oscillation (ENSO) and decadal-scale climate variability at 10 N in the eastern Pacic from 1893 to 1994: A coral-based reconstruction from Clipperton Atoll. Paleoceanography 15, 322335. Lobell, M.J., 1942. Some observations on the Peruvian coastal current. EOS Transactions AGU 23, 332336. Maes, C., Picaut, J., Belamari, S., 2002. Barrier layer and onset of El Nin o in a Pacic coupled model. Geophysical Research Letters 29, 2206. doi:10.1029/2002GL016029. Magan a, V., Amador, J.A., Medina, S., 1999. The midsummer drought over Mexico and central America. Journal of Climate 12, 1577 1588. Majluf, P., Reyes, J.C., 1989. The marine mammals of Peru: a review. ICLARM Conference Proceedings 18, 344363. Mantua, N.J., Battisti, D.S., 1994. Evidence for the delayed oscillator mechanism for ENSO: The observed oceanic Kelvin mode in the far western Pacic. Journal of Physical Oceanography 24, 691699. Mantua, N.J., Hare, S.R., 2002. The Pacic decadal oscillation. Journal of Oceanography 58, 3544. Manzanilla, S.R., 1989. The 19821983 El Nin o event recorded in dentinal growth layers in teeth of Peruvian dusky dolphins (Lagenorhynchus obscurus). Canadian Journal of Zoology 67, 21202125. Mayer, D.A., Weisberg, R.H., 1998. El Nin o-Southern Oscillation-related oceanatmosphere coupling in the western equatorial Pacic. Journal of Geophysical Research 103, 1863518648. McCreary, J.P., 1976. Eastern tropical ocean response to changing wind systems: with applications to El Nin o. Journal of Physical Oceanography 6, 632645. McCreary, J.P., 1983. A model of tropical oceanatmosphere interaction. Monthly Weather Review 111, 370387. McCreary, J.P., Anderson, D.L.T., 1991. An overview of coupled oceanatmosphere models of El Nin o and the Southern Oscillation. Journal of Geophysical Research 96, 31253150. McPhaden, M.J., Yu, X., 1999. Equatorial waves and the 199798 El Nin o. Geophysical Research Letters 26, 29612964. McPhaden, M.J. et al., 1998. The tropical ocean-global atmosphere observing system: A decade of progress. Journal of Geophysical Research 103, 1416914240. Mestas-Nun ez, A.M., Eneld, D.B., 2001. Eastern equatorial Pacic SST variability: ENSO and non-ENSO components and their climatic associations. Journal of Climate 14, 391402. Mestas-Nun ez, A.M., Miller, A., 2006. Interdecadal variability and climate change in the eastern tropical Pacic: A review. Progress in Oceanography 69 (24), 267284. Miller, G.T., Cheney, R., 1990. Large-scale meridional transport in the tropical Pacic during the 198687 El Nin o from Geosat. Journal of Geophysical Research 95, 1790517919. Miller, A.J., Alexander, M.A., Boer, G.J., Chai, F., Denman, K., Erickson, D.J., et al., 2003. Potential feedbacks between Pacic Ocean ecosystems and interdecadal climate variations. Bulletin of American Meteorological Society 84, 617633. Moore, A.M., Kleeman, R., 1999. Stochastic forcing of the ENSO by the intraseasonal oscillation. Journal of Climate 12, 1199 1220. Murphy, R.C., 1926. Oceanic and climatic phenomena along the west coast of South America during 1925. Geographical Review 16, 25 54. Murtugudde, R., Beauchamp, J., McClain, C.R., Lewis, M., Busalacchi, A.J., 2002. Eects of penetrative radiation on the upper tropical ocean circulation. Journal of Climate 15, 470486. Neelin, J.D. et al., 1998. ENSO theory. Journal of Geophysical Research 103, 1426214290. Paine, R.T., Tegner, M.J., Johnson, E.A., 1998. Compounded perturbations yield ecological surprises. Ecosystems 1, 535545. Parker, D.E., Folland, C.K., Jackson, M., 1995. Marine surface temperature: observed variations and data requirements. Climatic Change 31, 559600. Pazan, S., Meyers, G., 1982. Interannual uctuations in the tropical Pacic wind eld and the Southern Oscillation. Monthly Weather Review 110, 587600. Penland, C., Sardeshmuhk, P., 1995. The optimal growth of the tropical sea surface temperature anomalies. Journal of Climate 8, 1999 2024. Pennington, J.T., Mahoney, K.L., Kuwahara, V.S., Kolber, D.D., Calienes, R., Chavez, F.P., 2006. Primary production in the eastern tropical Pacic: a review. Progress in Oceanography 69, 285317. Philander, S.G., 1981. The response of equatorial oceans to a relaxation of the trade winds. Journal of Physical Oceanography 11, 176 189. Philander, S.G., 1985. El Nin o and La Nin a. Journal of the Atmospheric Sciences 42, 26522662. Philander, S.G., 1990. El Nin o, La Nin a, and the Southern Oscillation. Academic Press, London, 289 pp. Philander, S.G., Fedorov, A., 2003. Is El Nin o sporadic or cyclic? Annual Review of Earth and Planetary Sciences 31, 579594. Picaut, J., Delcroix, T., 1995. Equatorial wave sequence associated with the warm pool displacement during the 19861989 El Nin o and La Nin a. Journal of Geophysical Research 100, 1839818408. Picaut, J., Ioualalen, M., Menkes, C., Delcroix, T., McPhaden, M.J., 1996. Mechanism of the zonal displacements of the Pacic warm pool: Implications for ENSO. Science 274, 14861489. Picaut, J., Masia, F., du Penhoat, Y., 1997. An advective-reective conceptual model for the oscillatory nature of the ENSO. Science 277, 663666. Picaut, J., Hackert, E., Busalacchi, A.J., Murtugudde, R., Lagerloef, G.S.E., 2002. Mechanisms of the 19971998 El Nin o-La Nin a, as inferred from space-based observations. Journal of Geophysical Research 107. doi:10.1029/2001JC000850. Quinn, W., Zopf, D.O., Short, K.S., Yang, R.W.T., 1978. Historical trends and statistics of the Southern Oscillation, El Nin o, and Indonesian droughts. Fishery Bulletin 76, 663678.
265
Quinn, W., Neal, V.T., Mayolo, S.E.A., 1987. El Nin o occurrences over the past four and a half centuries. Journal of Geophysical Research 92, 1444914461. Rajagopalan, B., Lall, U., Cane, M.A., 1997. Anomalous ENSO occurrences: an alternate view. Journal of Climate 10, 23512357. Ramage, C.S., Hori, A.M., 1981. Meteorological aspects of El Nin o. Monthly Weather Review 109, 18271835. Ramirez, P., 1986. Distribucion y alimentacion de la ballena Bryde durante al fenomeno El Nin o 19821983. Boletin ERFEN 17, 2027. Rasmusson, E.M., Carpenter, T.H., 1982. Variations in tropical sea surface temperature and surface wind elds associated with the Southern Oscillation/El Nin o. Monthly Weather Review 110, 354384. Ribic, C.A., Ainley, D.G., Spear, L.B., 1992. Eects of El Nin o and La Nin a on seabird assemblages in the equatorial Pacic. Marine Ecology Progress Series 80, 109124. Rodbell, D.T., Seltzer, G.O., Anderson, D.M., Abbott, M.B., Eneld, D.B., Newman, J.H., 1999. A high-resolution $15000-year record of El Nin o driven alluviation in southwestern Ecuador. Science 283, 516520. nchez, G., Calienes, R., Zuta, S., 2000. The 199798 El Nin Sa o and its eects on the coastal marine ecosystem o Peru. California Cooperative Oceanic Fisheries Investigations Reports 41, 6286. Sandweiss, D.H., Maasch, K.A., Burger, R.L., Richardson, J.B.I., Rollins, H.B., Clement, A., 2001. Variation in Holocene El Nin o frequencies: climate records and cultural consequences in ancient Peru. Geology 29, 603606. Setoh, T., Imawaki, S., Ostrovskii, A., Umatani, S.-I., 1999. Interdecadal variations of ENSO signals and annual cycles revealed by wavelet analysis. Journal of Oceanography 55, 385394. Smith, T.M., Reynolds, R.W., Livezey, R.E., Stokes, D.C., 1996. Reconstruction of historical sea surface temperature using empirical orthogonal functions. Journal of Climate 9, 14031420. Solow, A.R., Huppert, A., 2003. On non-stationarity of ENSO. Geophysical Research Letters 30, 1910. doi:10.1029/2003GL018225. Soto, K., Trites, A.W., Arias-Schreiber, M., 2003. The eects of prey availability on pup mortality and the timing of birth of South American sea lions (Otaria avescens) in Peru. In: 15th Biennial Conference on the Biology of marine Mammals, Society for Marine Mammalogy, 1419 December 2003. Springer, S.R., McPhaden, M.J., Busalacchi, A.J., 1990. Oceanic heat content variability in the tropical Pacic during the 1982/1983 El Nin o. Journal of Geophysical Research 95, 2208922101. Suarez, M.J., Schopf, P.S., 1988. A delayed action oscillator for ENSO. Journal of the Atmospheric Sciences 45, 32833287. Tang, T.Y., Weisberg, R.H., 1984. On the equatorial Pacic response to the 1982/1983 El Nin o-Southern Oscillation event. Journal of Marine Research 42, 809829. Thompson, C.J., Battisti, D.S., 2001. A linear stochastic dynamical model of ENSO. Part II: Analysis. Journal of Climate 14, 445 466. n, V., Cabrera, D., 1987. Reproduction and population levels of Peruvian guano birds, 1980 to 1986. Journal of Tovar, H., Guille Geophysical Research 92 (C13), 1444514448. Trenberth, K.E., 1976. Spatial and temporal variations in the Southern Oscillation. Quarterly Journal of Royal Meteorological Society 102, 639653. Trenberth, K.E., Stepaniak, D.P., 2001. Indices of El Nin o evolution. Journal of Climate 14, 16971701. Trillmich, F., Limberger, D., 1985. Drastic eects of El Nin o on Galapagos pinnipeds. Oecologia 67, 1922. Valle, C.A., Cruz, F., Cruz, J.B., Merlen, G., Coulter, M.C., 1987. The impact of the 19821983 El Nin o-Southern Oscillation on seabirds in the Galapagos Islands, Ecuador. Journal of Geophysical Research 92 (C13), 1443714444. Van Loon, H., Madden, R.A., 1981. The Southern Oscillation, I, Global associations with pressure and temperature in northern winter. Monthly Weather Review 109, 11501172. Vialard, J., Menkes, C., Boulanger, J.P., Deleccluse, P., Guilyardi, E., McPhaden, M.J., Madec, G., 2001. Oceanic mechanisms driving the SST during the 199798 El Nin o. Journal of Physical Oceanography 31, 16491675. Walker, G.T., 1923. Correlation in seasonal variations of weather VIII: A preliminary study of world weather. Memoirs of the Indian Meteorological Department 24, 75131. Walker, G.T., 1924. Correlation in seasonal variations of weather IX: A further study of world weather. Memoirs of the Indian Meteorological Department 24, 275332. Walker, G.T., 1928. World weather III. Memoirs of the Royal Meteorological Society 2, 97106. Walker, G.T., Bliss, E., 1932. World Weather V. Memoirs of the Royal Meteorological Society 4, 5384. Wallace, J.M. et al., 1998. On the structure and evolution of ENSO-related climate variability in the tropical Pacic: Lessons from TOGA. Journal of Geophysical Research 103, 1424114259. Wang, B., 1995. Interdecadal changes in El Nin o onset in the last four decades. Journal of Climate 8, 267285. Wang, C., 2000. On the atmospheric responses to tropical Pacic heating during the mature phase of El Nin o. Journal of the Atmospheric Sciences 57, 37673781. Wang, C., 2001a. A unied oscillator model for the El Nin o-Southern Oscillation. Journal of Climate 14, 98115. Wang, C., 2001b. On the ENSO mechanisms. Advances in Atmospheric Sciences 18, 674691. Wang, C., 2002a. Atmospheric circulation cells associated with the El Nin o-Southern Oscillation. Journal of Climate 15, 399419. Wang, C., 2002b. Atlantic climate variability and its associated atmospheric circulation cells. Journal of Climate 15, 15161536. Wang, C., Eneld, D.B., 2001. The tropical Western Hemisphere warm pool. Geophysical Research Letters 28, 16351638. Wang, C., Eneld, D.B., 2003. A further study of the tropical Western Hemisphere warm pool. Journal of Climate 16, 14761493. Wang, C., Picaut, J., 2004. Understanding ENSO physics A review. In: Wang, C., Xie, S.-P., Carton, J. (Eds.), Earths Climate: The OceanAtmosphere Interaction. American Geophysical Union, pp. 2148. Wang, C., Weisberg, R.H., 2000. The 199798 El Nin o evolution relative to previous El Nin o events. Journal of Climate 13, 488501.
266
Wang, C., Weisberg, R.H., Virmani, J.I., 1999. Western Pacic interannual variability associated with the El Nin o-Southern Oscillation. Journal of Geophysical Research 104, 51315149. Weare, B.C., 1982. El Nin o and tropical Pacic Ocean temperature. Journal of Physical Oceanography 12, 1727. Weare, B.C., Navato, A.R., Newell, R.E., 1976. Empirical orthogonal function analysis of Pacic sea surface temperatures. Journal of Physical Oceanography 6, 671678. Weisberg, R.H., Wang, C., 1997a. Slow variability in the equatorial west-central Pacic in relation to ENSO. Journal of Climate 10, 1998 2017. Weisberg, R.H., Wang, C., 1997b. A western Pacic oscillator paradigm for the El Nin o-Southern Oscillation. Geophysical Research Letters 24, 779782. Woodru, S.D., Slutz, R.J., Jenne, R.L., Steurer, P.M., 1987. A comprehensive oceanatmosphere data set. Bulletin of the American Meteorological Society 68, 12391250. Wooster, W.S., 1960. El Nin o. California Cooperative Oceanic Fisheries Investigations Reports 7, 4345. Wooster, W.S., Guillen, O., 1974. Characteristics of El Nin o in 1972. Journal of Marine Research 32, 378404. Wunsch, C., 1999. The interpretation of short climate records, with comments on the north Atlantic and southern oscillations. Bulletin of the American Meteorological Society 80, 245255. Wyrtki, K., 1975. El Nin o The dynamic response of the equatorial Pacic Ocean to atmospheric forcing. Journal of Physical Oceanography 5, 572584. Wyrtki, K., 1985. Water displacements in the Pacic and genesis of El Nin o cycles. Journal of Geophysical Research 90, 71297132. Xie, S.-P., Xu, H., Kessler, W.S., Nonaka, M., 2005. Airsea interaction over the eastern Pacic warm pool: Gap winds, the thermocline dome, and atmospheric convection. Journal of Climate 18, 525. Zavala-Garay, J., Moore, A.M., Perez, C.L., Kleeman, R., 2003. The response of a coupled model of ENSO to observed estimates of stochastic forcing. Journal of Climate 16, 28272842. Zebiak, S.E., 1989. Ocean heat content variability and El Nin o cycles. Journal of Physical Oceanography 19, 475486. Zebiak, S.E., 1990. Diagnostic studies of Pacic surface winds. Journal of Climate 3, 10161031. Zebiak, S.E., Cane, M.A., 1987. A model El Nin o-Southern Oscillation. Monthly Weather Review 115, 22622278.

2006 Wang Fiedler

Uploaded by

Copyright:

Available Formats

2006 Wang Fiedler

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2006 Wang Fiedler

Uploaded by

Copyright:

Available Formats

Progress in Oceanography

Progress in Oceanography 69 (2006) 239266 www.elsevier.com/locate/pocean

ENSO variability and the eastern tropical Pacic: A review

Corresponding author. E-mail address: [email protected] (C. Wang).

0079-6611/$ - see front matter. Published by Elsevier Ltd. doi:10.1016/j.pocean.2006.03.004

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

Nino3 SST Anomalies

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

(a): SST Anomalies

30N 20N 10N 0 10S 20S

(c): Wind Anomalies

0 10S 20S 30S 120E 30N 20N

30S 120E 30N 20N 10N 0 10S 20S 30S 120E

(b): SLP Anomalies

(d): OLR Anomalies

10N 0 10S 20S 30S 120E

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

(a): SST Anomalies

30N 20N 10N 0 10S 20S

(c): Wind Anomalies

0 10S 20S 30S 120E 30N 20N

30S 120E 30N 20N 10N 0 10S 20S 30S 120E

(b): SLP Anomalies

(d): OLR Anomalies

10N 0 10S 20S 30S 120E

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

Zonal Wind Anomalies

(a): 850-mb zonal wind anomalies in the west

2.5 0.0 -2.5 -5.0 1950

1.5 0.0 -1.5 -3.0 2000

(b): 850-mb zonal wind anomalies in the east

3.5 0.0 -3.5 -7.0 1950

1.5 0.0 -1.5 -3.0 2000

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

Different Origin and Development of El Ninos

(a): El Ninos between 1950 and 1976

(c): The 1997/98 El Nino

30S 120E 30N

30S 120E 30N

(b): El Ninos between 1976 and 1996

(d): The 2002/03 El Nino

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

(a) Seasonal SST Variability

(b) ENSO SST Variability

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

The Delayed Oscillator

Upwelling Rossby Waves Equator

Upwelling Kelvin Waves

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

Sverdrup Transpor t SST (+)

Sverdrup Transpor t SSTa( )

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

The Western Pacific Oscillator

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

The Advective-Reflective Oscillator

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266

C. Wang, P.C. Fiedler / Progress in Oceanography 69 (2006) 239266