Western Equatorial Pacific Warm Pool Research Articles

Estimates of Pliocene Tropical Pacific Temperature Sensitivity to Radiative Greenhouse Gas Forcing

AbstractThe Western Equatorial Pacific (WEP) warm pool, with surface temperatures >28 °C and a deep thermocline, is an important source of latent and sensible heat for the global climate system. Because the tropics are not sensitive to ice‐albedo feedbacks, the WEP's response to radiative forcing can be used to constrain a minimum estimate of Earth system sensitivity. Climate modeling of pCO2‐radiative warming projections shows little change in WEP variability; here we use temperature distributions of individual surface and subsurface dwelling fossil foraminifera to evaluate past variability and possible radiative and dynamic climate forcing over the Plio‐Pleistocene. We investigate WEP warm pool variability within paired glacial‐interglacial (G‐IG) intervals for four times: the Holocene‐Last Glacial Maximum, ~2, ~3, and ~4 Ma. Our results show that these surface and subsurface temperature distributions are similar for all G‐IG pairs, indicating no change in variability, even as pCO2‐radiative forcing and other boundary conditions changed on G‐IG timescales. Plio‐Pleistocene sea surface temperature (SST) distributions are similar to those from the Holocene, indicating WEP SSTs respond to pCO2‐radiative forcing and associated feedbacks. In contrast, Plio‐Pleistocene subsurface temperature distributions suggest subsurface temperatures respond to changes in thermocline temperature and depth. We estimate tropical temperature sensitivity for the mid‐Pliocene (~3 Ma) using our individual foraminifera SST data set and a previously published high‐resolution boron isotope‐based pCO2 reconstruction. We find tropical temperature sensitivity was equal to, or less than, that of the Late Pleistocene.

El Niño Physics and El Niño Predictability

Much of the year-to-year climate variability on the Earth is associated with El Niño and the Southern Oscillation (ENSO). This variability is generated primarily by a coupled ocean-atmosphere instability near the eastern edge of the western equatorial Pacific warm pool. Here, I discuss the physics of this variability, including its phase locking to the seasonal cycle. ENSO growth typically occurs from April/May to November, and by July the perturbation is usually strong enough that it persists to the beginning of the following year, when ENSO events usually end. Consequently, predicting ENSO is easy from July to February but is more challenging across the April/May transition to the next event. I discuss precursors of this transition and recent results from dynamical and statistical models used for ENSO forecasting.

Isolation of sea surface salinity maps on various timescales in the tropical Pacific Ocean

To isolate sea surface salinity (SSS) maps on seasonal, ENSO, decadal, and long-term trend timescales in the tropical Pacific Ocean, the ensemble empirical mode decomposition (EEMD) is applied on an SSS data set covering 1950–2009, concerning three key regions including the western equatorial Pacific Warm Pool (WP), South Pacific Convergence Zone (SPCZ), and Inter Tropical Convergence Zone (ITCZ); then a self-organizing map is performed on the intrinsic mode function maps decomposed by the EEMD, concerning the whole basin. The ENSO and decadal signals concern mainly the western Pacific, in contrast to the seasonal signal mostly notable in the east. (1) The modulated annual cycle has smaller (larger) amplitudes during El Nino (La Nina) years and later appearance of minima during El Nino years; one unique annual cycle is observed at the northwestern edge of the ITCZ lagging the well-known ITCZ cycle by ~3 months. (2) The pronounced 1999–2001 SSS-related La Nina event in the SPCZ was reinforced twice by the decadal shift in the 1990s; the eastern Pacific and central Pacific ENSO-related SSS features are compared. (3) The contrasted anomalies between the western equatorial and non-equatorial regions were pronounced during 1977–1996, whereas they were less pronounced during 1971–1976, 2005–2008, and a roughly opposite pattern appeared with strong and abrupt decrease shift prevailing in large areas over the southwestern basin during 1997–2004. (4) The freshening at the western equator and the saltening located east of the SPCZ SSS front together amplify the geographical SSS contrasts exhibited by the WP and SPCZ SSS fronts.

Mid-Piacenzian sea surface temperature record from ODP Site 1115 in the western equatorial Pacific

Planktic foraminifer assemblages and alkenone unsaturation ratios have been analyzed for the mid-Piacenzian (3.3 to 2.9 Ma) section of Ocean Drilling Program (ODP) Site 1115B, located in the western equatorial Pacific off the coast of New Guinea. Cold and warm season sea surface temperature (SST) estimates were determined using a modern analog technique. ODP Site 1115 is located just south of the transition between the planktic foraminifer tropical and subtropical faunal provinces and approximates the southern boundary of the western equatorial Pacific (WEP) warm pool. Comparison of the faunal and alkenone SST estimates (presented here) with an existing nannofossil climate proxy shows similar trends. Results of this analysis show increased seasonal variability during the middle of the sampled section (3.22 to 3.10 Ma), suggesting a possible northward migration of both the subtropical faunal province and the southern boundary of the WEP warm pool.

Faunal re-evaluation of Mid-Pliocene conditions in the western equatorial Pacific

Mid-Pliocene low-latitude Pacific faunal (planktic foraminifer) sea surface temperature (SST) estimates are normally based upon the Modern Analog Technique (MAT). In the Eastern equatorial Pacific (EEP), where upwelling of cool water predominates, MAT can be used to discern both cooling and warming in Neogene records. SST today is ~30°C in the western equatorial Pacific (WEP) warm pool, the upper limit of the modern calibration data, and past warming above that level is difficult to assess using faunal methods. Mid-Pliocene fossil samples from the WEP have been analyzed using several variations of MAT with different outcomes and associated levels of confidence. While SST above ~30°C in the WEP during the mid-Pliocene cannot be ruled out due to the limitations of the method, temperatures this warm seem unlikely. In addition to the mid-Pliocene, planktic foraminifer assemblages from the coretop, last glacial maximum, last interglacial and the penultimate glacial (Marine Isotope Stage 6) show striking similarity to each other which suggests little to no change in the region between times of global climate extremes. There is generally good agreement between the Mg/Ca paleothermometer and MAT derived faunal SST estimates. Both suggest stability of the WEP warm pool.

An examination of summer precipitation over Asia based on TRMM/TMI

A 6-year dataset of summer monthly mean precipitation derived from Tropical Precipitation Measurement Mission (TRMM)-Microwave Imager (TMI) was used to delineate the spatial distribution patterns of precipitation throughout Asian areas, which indicates that there are three rainfall centers located at the northern coast of the Bay of Bengal, the South China Sea and the western equatorial Pacific Warm Pool, respectively. Based upon the analysis of horizontal distribution, the capability of TMI for characterizing terrestrial and maritime precipitation has been evaluated and compared with Global Precipitation Climatology Project (GPCP) dataset. It was found that TMI and GPCP are well consistent with each other, while a few significant differences occur at several regions over land. By investigating rainfall estimates over six specific locations in Asia, a systematic underestimation of TMI was demonstrated, which could be explained by the inherent deficiency within TMI terrestrial algorithm relying on scattering signal from ice particles in a precipitation system. A further analysis shows that the highly inhomogeneous distribution of rain gauges employed by GPCP contributes a great deal to the significant discrepancy between GPCP and TMI, especially over regions surrounding the Tibetan Plateau where rain gauges are quite scarce.

The Mid-Pleistocene Transition in the Tropical Pacific

A sea surface temperature (SST) record based on planktonic foraminiferal magnesium/calcium ratios from a site in the western equatorial Pacific warm pool reveals that glacial-interglacial oscillations in SST shifted from a period of 41,000 to 100,000 years at the mid-Pleistocene transition, 950,000 years before the present. SST changes at both periodicities were synchronous with eastern Pacific cold-tongue SSTs but preceded changes in continental ice volume. The timing and nature of tropical Pacific SST changes over the mid-Pleistocene transition implicate a shift in the periodicity of radiative forcing by atmospheric carbon dioxide as the cause of the switch in climate periodicities at this time.

The frontal area at the eastern edge of the western equatorial Pacific warm pool in April 2001

The western part of the equatorial Pacific Ocean is characterized, on average, by a relatively warm, fresh, and oligotrophic near‐surface water mass, the warm pool. On its eastern side the warm pool joins cooler, saltier, and richer waters from the equatorial upwelling. Previous works have shown that the boundary between these two water masses is highly variable in zonal location and structure, depending on large‐scale climatic conditions and/or local atmospheric forcing. This paper reports results from an oceanographic cruise carried out in April 2001, dedicated to the study of that boundary. By combining physical and biogeochemical data, the frontal area could unambiguously be found close to 157.5°E, in the far‐western equatorial Pacific, in accordance with La Niña climatic conditions. Some characteristics of the frontal zone were similar to the scarce previous observations: a sharp pCO2 step of about 40 ppm and a variation of NO3 concentration from about 1 μM to complete depletion. Other features were not previously reported, particularly the unexpected lack of a noticeable surface salinity front. Additional data from the Tropical Atmosphere‐Ocean/Triangle Trans‐Ocean Buoy Network (TAO/TRITON) moorings array provide a tentative explanation for that lack of salinity gradient: It appears that at the time of our observations the eastern part of the warm pool presented anomalously high surface salinity, about 0.5 above normal. The local evaporation‐precipitation budget did not explain this high salinity. Rather, it was associated with an anomalous northward extension of the South Pacific high‐salinity tongue in the upper thermocline.

Modeling the evolution of a fresh sea surface anomaly produced by tropical rainfall

The evolution of a rain‐produced, fresh surface anomaly observed in the western equatorial Pacific warm pool was modeled by use of the Princeton Ocean Model with Mellor‐Yamada turbulent closure. The simulation was forced by a wind stress of 0.12 N m−2 and surface cooling of 225 W m−2. Following spin‐up, 34‐mm average rainfall was applied over a small portion of the domain. When rainfall ceased, momentum was trapped in a 1‐m, fresh surface layer, and turbulent mixing was inhibited below the layer. Seven hours later, the surface density anomaly was reduced by an order of magnitude, and the surface velocity anomaly in the direction of the wind was 0.15 m s−1. The simulations show upwelling at the upwind edge of the anomaly and downwelling at the downwind edge with maxima of 15 and 35 m d−1, respectively. The momentum balance for the velocity component in the direction of the wind indicated that local acceleration, horizontal advection, and vertical diffusion were in approximate balance except at the downwind edge of the anomaly, where pressure gradient terms were also important. The budget of turbulent kinetic energy showed that shear production plus buoyancy production (or destruction) was approximately balanced by dissipation in much of the anomaly; advection and vertical turbulent transport terms were significant at some depths at the edges. The strength of the surface density front at the downwind edge of the anomaly tended to increase with time relative to the front at the upwind edge.

Mixed layer depth variability over the global ocean

The spatial and monthly variability of the climatological mixed layer depth (MLD) for the global ocean is examined using the recently developed Naval Research Laboratory (NRL) Ocean Mixed Layer Depth (NMLD) climatologies. The MLD fields are constructed using the subsurface temperature and salinity data from the World Ocean Atlas 1994 [Levitus et al., 1994; Levitus and Boyer, 1994]. To minimize the limitations of these global data in the MLD determination, a simple mixing scheme is introduced to form a stable water column. Using these new data sets, global MLD characteristics are produced on the basis of an optimal definition that employs a density‐based criterion having a fixed temperature difference of ΔT = 0.8°C and variable salinity. Strong seasonality of MLD is found in the subtropical Pacific Ocean and at high latitudes, as well as a very deep mixed layer in the North Atlantic Ocean in winter and a very shallow mixed layer in the Antarctic in all months. Using the climatological monthly MLD and isothermal layer depth (ILD) fields from the NMLD climatologies, an annual mean ΔT field is presented, providing criteria for determining an ILD that is approximately equivalent to the optimal MLD. This enables MLD to be determined in cases where salinity data are not available. The validity of the correspondence between ILD and MLD is demonstrated using daily averaged subsurface temperature and salinity from two moorings: a Tropical Atmosphere Ocean array mooring in the western equatorial Pacific warm pool, where salinity stratification is important, and a Woods Hole Oceanographic Institute (WHOI) mooring in the Arabian Sea, where strongly reversing seasonal monsoon winds prevail. In the western equatorial Pacific warm pool the use of ILD criterion with an annual mean ΔT value of 0.3°C yields comparable results with the optimal MLD, while large ΔT values yield an overestimated MLD. An analysis of ILD and MLD in the WHOI mooring show that use of an incorrect ΔT criterion for the ILD may underestimate or overestimate the optimal MLD. Finally, use of the spatial annual mean ΔT values constructed from the NMLD climatologies can be used to estimate the optimal MLD from only subsurface temperature data via an equivalent ILD for any location over the global ocean.

Salinity barrier layer and onset of El Niño in a Pacific coupled model

The importance of the barrier layer during the onset of El Niño is investigated using a coupled ocean‐atmosphere general circulation model. Sensitivity experiments are done by removing or keeping the salinity stratification in the upper layer of the western equatorial Pacific warm pool. The barrier layer favors the maintenance and displacement of the warm pool into the central Pacific by isolating the mixed layer from the entrainment cooling at depth and by confining the response of westerly wind events (WWEs) to a shallow mixed layer. The increased zonal fetch of WWEs through the coupling with sea surface temperature (SST) enhances downwelling equatorial Kelvin waves and thus leads to El Niño. In the absence of salinity stratification, slightly cooler SST and a reduced eastward displacement of the warm pool result in a reduced El Niño or a return to the mean seasonal cycle of the model. The possibility that the barrier layer affects the onset of El Niño pleads for a careful consideration of the salinity stratification in climate forecasts.

Sharp frontal interfaces in the near-surface layer of the tropical ocean

The small-scale structure of oceanic fronts contains important information about horizontal and vertical exchange of properties in the upper ocean. The data obtained in the western equatorial Pacific warm pool during Tropical Ocean–Global Atmosphere Coupled Ocean–Atmosphere Response Experiment (TOGA COARE) and Tropical Ocean Climate Study (TOCS) suggest that the sharp frontal interfaces occasionally observed in the upper layer of the tropical ocean may be associated with gravity currents. These gravity currents originate from the surface pools of relatively low-density (fresh and/or warm) water produced by convective rainfalls and spatially variable diurnal warming. It has been shown that frontal interfaces of less than 100 m width may interact with wind stress via the mechanism of Stommel's “overturning gate.” The anisotropy of sharp frontal interfaces with respect to the wind stress direction can be predicted with a simple nonlinear model including both dissipation and dispersion effects. This study elucidates the role of fronts in the dynamics of the tropical ocean and provides important details to the description of how the temperature–salinity relationship and barrier layer in the warm pool areas are formed.

Simulation of the Mixed Layer in the Western Equatorial Pacific Warm Pool

The upper ocean in the western equatorial Pacific warm pool during TOGA-COARE IMET IOP was simulated using a one-dimensional turbulence closure ocean mixed-layer model, which considered recent observations, such as the remarkable enhancement of turbulent kinetic energy near the ocean surface. The shoaling/deepening of the mixed layer and warming/cooling subsurface water in the model were in reasonable agreement with the observations. There was a significant improvement in simulating the cooling trend of the sea surface temperature under a westerly wind burst with heavy rainfall over previous simulations using bulk mixed-layer models. By contrast the simulated sea surface salinity (SSS) departed significantly from the observed SSS, especially during a westerly burst and the subsequent restratification period, which might be due to 3-D control processes, such as downwelling/upwelling or advection.

Using an Ocean Model to Examine ENSO Dynamics

Abstract An equatorial ocean model with the Kelvin and first six Rossby waves and n vertical modes is developed to simulate the equatorial sea surface temperature anomalies (SSTAs) and, more importantly, zonal movement of the western equatorial Pacific warm pool associated with an ENSO episode. When forced by observed wind stress, the model reproduces well the SSTAs in both the central and eastern equatorial Pacific. The two gravest baroclinic modes are needed to model the equatorial thermocline depth, zonal current, and SSTAs. The SSTA near the equatorial South American coast is mostly due to upwelling and that in the eastern Pacific from about 150°–100°W is mainly due to both zonal advection and upwelling. The SSTA in the equatorial Pacific near the edge of the warm pool (160°E–160°W) mostly results from the advection of mean SST by the anomalous zonal current and Newtonian damping. Model calculations show that the growing ENSO disturbance, which propagates eastward from the western equatorial Pacific d...

Seasonal change in foraminiferal production in the western equatorial Pacific warm pool: evidence from sediment trap experiments

Seasonal changes in the flux of individual planktonic foraminifera in the deep ocean have been studied in the western equatorial Pacific warm pool using material collected by sediment traps over a period of 1 yr (1991-1992). Mean total foraminiferal fluxes (TFFs) were 171 and 97 shells m(-2) day(-1) at Sites N1 (2degrees59.8'N, 135degrees01.5'E) and N2 (4degrees07.5'N, 136degrees16.6'E), respectively. The seasonal change in TFFs was much higher than for organic matter (OM) and carbonate fluxes. The major species of planktonic foraminifera were Pulleniatina obliquiloculata, Neogloboquadrina dutertrei, Globigerinoides sacculifer, Globigerinella aequilateralis and Globigerinoides ruber. Symbiont-bearing species and N. dutertrei were predominant in the western equatorial Pacific warm pool. Comparison between the warm pool and upwelling (Deep-Sea Research II 43 (1996) 1257, Mar. Micropaleontology 33 (1998) 157) indicates that the foraminiferal assemblage trapped at the western equatorial Pacific warm pool was comparable with the South Equatorial and subtropical assemblages during a plankton tow study conducted at 140degreesW in 1992. It suggests that fairly oligotrophic conditions prevailed in the western equatorial Pacific warm pool in association with terrestrial nutrient inputs or temporary mixing by wind bursts. The OM flux was more positively correlated with the < 63 Jma carbonate fraction, mainly composed of coccoliths (r = 0.87) and biogenic opal, which suggests that OM fluxes were more affected by phytoplankton productivity. Although food availability is generally important for foraminiferal production, TFFs were poorly correlated with OM fluxes at Sites N1 and N2, which is partly attributed to the significant influence of symbiont photosynthesis. The delta(18)O values of P. obliquiloculata and N. dutertrei showed maxima in association with high OM fluxes during Period 2 (from August to October) and Period 4 (from late December to April) at Site N1, reflecting thickening of the mixed layer, following wind bursts. G. sacculifer showed a lunar reproduction cycle, when, between July and October, carbonate production was more predominant than biogenic opal production. A similar profile was evident for G. ruber, Orbulina universa and G. aequilateralis. These four species are generally predominant at low latitudes. Therefore, the lunar reproduction cycle may significantly affect the fluctuation in carbonate production in the equatorial ocean. (C) 2002 Elsevier Science Ltd. All rights reserved.

Distributions and variations in the partial pressure of CO2 in surface waters (pCO2w) of the central and western equatorial Pacific during the 1997/1998 El Niño event

Measurements of partial pressure of CO 2 in surface waters ( p CO 2 w ) and overlying air ( p CO 2 a ) were made in the central and western equatorial Pacific from October 1997 to February 1998 within the period of the 1997/1998 El Niño , which was reported to be the strongest El Niño event on record. The distribution of the p CO 2 w showed a pattern driven by the eastward movement of western Pacific warm pool and thermodynamic effects (temperature and salinity), which was different from those of the moderate 1986/1987 El Niño and non-El Niño periods. Due to the eastward movement of the warm pool with sea surface temperature (SST) higher than 28.5 °C and sea surface salinity (SSS) lower than 34.5, the p CO 2 w between 180° and 163°W (347–364 μatm) was almost equal to that of the air (351 μatm). Between 143°E and 180°, the p CO 2 w tended to increase toward the west (387 μatm at 0°, 144°E in December 1997) along with the SST and SSS. West of 143°E in January 1998, a steep change in p CO 2 w ranging from 320 to 365 μatm occurred while retaining high SST (>28.5 °C) and SSS (>34.5). This was caused by the advection of surface water from the southern low latitudes that had been affected by biological activity (New Guinea Coastal Current). From December 1997 to January/February 1998, the SSS was usually higher than 34.5 west of 180°, which was significantly high compared to the western equatorial Pacific warm pool. This was probably due to the decrease of the net fresh water input for the western equatorial Pacific and/or the northward migration of surface water from the Southern Hemisphere . The CO 2 outflux from the central and western equatorial Pacific (5.5°S–5.5°N, 139.5°E–159.5°W) was estimated to be 0.027 Pg-C/year in December 1997 and 0.038 Pg-C/year in January/February 1998. This presents a significant decrease from the CO 2 outflux during the non-El Niño periods (0.34 Pg-C/year in January/February 1989, 0.11 Pg-C/year in September/November 1990) and a slight one from the moderate 1986/1987 El Niño period (0.055 Pg-C/year in January/February 1987). Following the El Niño–Southern Oscillation phenomena, CO 2 outflux varied largely in the central equatorial Pacific and little in the western equatorial Pacific.

The Heat Balance in the Western Equatorial Pacific Warm Pool during the Westerly Wind Bursts: A Case Study

The responses of sea surface temperature (SST) in the western equatorial Pacific warm pool to the westerly wind bursts (WWBs) play an important role in the relationship between WWB and ENSO. By using data collected from eight buoys of TOGA (Tropical Ocean-Global Atmosphere)- COARE (Coupled Ocean-Atmosphere Response Experiment), the heat balances of the upper ocean in the western equatorial Pacific around 0°, 156°E during two WWB events were calculated according to Stevenson and Niiler’s (1983) method. In both events, SST increased before and after the WWBs, while decreased within the WWBs. The SST amplitudes approximated to I °C. Although sometimes the horizontal heat advections may become the biggest term in the heat balance, the variation of SST was dominated by the surface heat flux. On the other aspect, some different features of the two events are also revealed. The two cases have different variation of mixed layer depth. The depth of mixed layer is almost double in the first case (35 m to 70 m), which is caused by Ekman convergence, while only !Om increments due to entrainment in the second one. There are also differences in the currents structure. The different variations of thermal and currents structure in the mixing layers accounted for the different variation of the heat balance during the two events, especially the advection and residue terms. The seasonal variation of SST in this area is also investigated simply. The first WWB event happened just during the seasonal transition. So we considered that it is a normal season transition rather than a so-called anomaly. That also suggested that the seasonal distinction of the WWB is worthy of more attention in the researches of its relationship to ENSO.

Heat Balance in the Pacific Warm Pool Atmosphere during TOGA COARE and CEPEX

The atmosphere above the western equatorial Pacific warm pool (WP) is an important source for the dynamic and thermodynamic forcing of the atmospheric general circulation. This study uses a high-resolution reanalysis and several observational datasets including Global Precipitation Climatology Project precipitation, Tropical Ocean Global Atmosphere (TOGA) Tropical Atmosphere Ocean moored buoys, and Earth Radiation Budget Experiment, TOGA Coupled Ocean‐Atmosphere Response Experiment (COARE), and Central Equatorial Pacific Experiment (CEPEX) radiation data to examine the details of the dynamical processes that lead to this net positive forcing. The period chosen is the period of two field experiments: TOGA COARE and CEPEX during December 1992‐March 1993. The four months used in the study were sufficient to establish that the warm pool atmosphere (WPA) was close to a state of radiative‐convective‐dynamic equilibrium. The analysis suggests that the large-scale circulation imports about 200 W m22 of sensible heat and about 140 W m22 of latent energy into the WPA mainly through the low-level mass convergence and exports about 420 W m22 potential energy mainly through the upper-level mass divergence. Thus the net effect of the large-scale dynamics is to export about 80 W m22 energy out of the WPA and cool the WPA by about 0.8 K day21. The dynamic cooling in addition to the radiative cooling of about 0.4 K day21 or 40 W m22 leads to a net radiative‐dynamic cooling of about 1.2 K day21 or 120 W m22, which should be balanced by convective heating of the same magnitude. The WPA radiative cooling is only about 0.4 K day21, which is considerably smaller than previously cited values in the Tropics. This difference is largely due to the cloud radiative forcing (CRF), about 70 W m 22, associated with the deep convective cirrus clouds in the WPA, which compensates the larger clear sky radiative cooling. Thus moist convection heats the WPA, not only through the direct convective heating, that is, the vertical eddy sensible heat and latent energy transport, but also through the indirect convective heating, that is, the CRF of deep convective clouds. The CRF of the deep convective clouds has a dipole structure, in other words, strong heating of the atmosphere through convergence of longwave radiation and a comparable cooling of the surface through the reduction of shortwave radiation at the surface. As a result, the deep convective clouds enhance the required atmospheric heat transport and reduce the required oceanic heat transport significantly in the WP. A more detailed understanding of these convective processes is required to improve our understanding of the heat transport by the large-scale circulation in the Tropics.

ENSO prediction using an ENSO trigger and a proxy for Western Equatorial Pacific Warm Pool Movement

The zonal windstress anomaly in the far‐western equatorial Pacific (130–160°E) is a precursor to El Niño and La Niña episodes. A linear combination of this windstress with the El Niño index NINO3.4 can be used to predict ENSO successfully. Cross‐verified prediction results for NINO3.4 compare favorably with those of leading ENSO prediction models. The model is improved slightly if the time trend in sea surface temperature (SST) near the eastern edge of the western Pacific warm pool is included in the linear combination. Physically this trend is related to zonal equatorial ocean flow which advects the warm pool. The trend therefore aids prediction since it is a precursor to warm pool position and hence ENSO variability.

A Simple Warm-Pool Displacement ENSO Model

Abstract The authors derive analytical and numerical solutions to a simple model of the zonal displacement of the western equatorial Pacific warm pool. The model is based on a single variable x(t), the anomalous displacement of the eastern edge (28.5°C isotherm) of the warm pool. Although the physics differs, this model has a similar mathematical form to the delayed oscillator ENSO model. As in delayed oscillator models, although the wave-propagation time delay is crucial to the existence and period of the oscillation, the oscillation period is also critically dependent on the relative importance of the growth rate of the instability and the negative feedback. Consequently, the period of the oscillation can be several times the wave propagation time. Unlike delayed oscillator models, the wave propagation time delay is dependent on x(t) since the wave’s propagation time depends on the distance of the edge of the warm pool from the ocean boundaries. If the wave propagation delay time is approximated as a co...