Search Results

You are looking at 1 - 10 of 19 items for

  • Author or Editor: David B. Enfield x
  • All content x
Clear All Modify Search
David B. Enfield

Abstract

This study calculates a detailed climatological inventory of the oceanic heat balance in the equatorial Pacific. The gridded climatology of Weare et al. is used as an estimate of net surface heating. Zonal and meridional/vertical advection are estimated in a manner similar to that of Wyrtki, using the gridded climatologies for wind stress (Wyrtki and Meyers) and sea surface temperature (Reynolds), plus estimates of zonal transport. In addition, the meridional diffusion of heat into the cold tongue has been estimated from the work of Hansen and Paul and the terms of the heat flux has been examined for consistency with expectations about the remaining, vertical diffusion process. The effects of using the alternate climatologies of Esbensen and Kushnir and Reed for the net surface heating are also calculated.

The total advective heat flux divergence is calculated to be −27 ± 7, −91 ± 17 and −48 ± 17 W m2, respectively, in the western, central and eastern equatorial Pacific with meridional advection and upwelling removing about three times as much heat as zonal advection. The advective contribution are in approximate agreement with Wyrtki's “likely case” estimates for the 100°W–170°E longitude zone. The contribution from zonal advection, meridional advection and meridional diffusion are found to be greatest during the Boreal fall, winter and fall–winter seasons, respectively.

Depending on the climatology used for the net surface heat gain, the assumption of a uniform meridional diffusivity of 2 × 104 m2 s−1 leads to physically unrealistic residual flux divergences that imply a heat gain from vertical turbulence in the central Pacific or that vertical turbulence removes much more heat from the western and eastern Pacific than from the central Pacific. Total neglect of the meridional diffusion exacerbates the problems. Increasing the meridional diffusivity to 6 × 104 m2 s−1 in the central Pacific, consistent with direct estimates by Hansen and Paul, gives zonally uniform, negative residuals that are physically consistent with existing measurements of equatorial turbulence. With the model so tuned, the “best guess” heat balance in the central Pacific involves significant contributions from all terms, in the western Pacific between surface heat gain from the atmosphere and losses due to vertical diffusion, and in the eastern Pacific between surface gain and losses due to meridional advection (upwelling) and vertical diffusion.

Full access
David B. Enfield

Abstract

The time and space variability of low-level winds over the Southeast Tropical Pacific (SETP) region is described for the 6-year period 1974–80. The data set consists of monthly averaged low-level cloud-motion vector winds supplemented by coastal surface winds and pressure-related indices at fixed stations, and by long-term averages of ship-reported surface winds. The data are statistically analyzed in terms of correlations and empirical orthogonal functions (EOF's).

The annual cycle of the SETP low-level wind speeds is most prominent north of 15°S, with minimum and maximum intensifies during the January–March and July–September periods, respectively, in phase with the larger scale southeast trades. South of 15°S the annual variability is small and characterized by minimum wind speeds from May through July, when the anticyclonic circulation center is at its northernmost position (∼26°S). There is an SETP core region of maximum wind speeds that annually migrates (along 85°W) from 20°5 in January–March to 15°S in June–August. Peru coastal winds are seasonally in phase with the southeast trade circulation at Talara (4°5) and San Juan (15°5) but considerably out of phase at Chimbote (9°5), Lima (12°S) and Tacna (17°S).

The nonseasonal component of the SETP circulation is characterized by areal coherence over the 15–30°S subregion. When the SETP circulation is unseasonably weak (strong), so is the intensity of the high-speed core, and both the core and the anticyclonic circulation center tend to lie north (south) of their climatological positions. A weak SETP circulation with northward lying positional characteristics prevailed during the 1976–77 El Nin̄o. Nonseasonal variability is poorly correlated or uncorrelated between the coastal winds and various measures of either the basinwide or SETP circulations. Monthly anomalies of the low-level cloud-motion winds at 20°S, 85°W are an effective index of the nonseasonal atmospheric, circulation of the SETP region.

Full access
David B. Enfield

Abstract

Daily sea level and surface winds at eastern Pacific shore locations and equatorial islands, together with gridded five-day averages of 850 mb winds, have been analyzed for the 1979–84 period to determine how the 40–60 day intraseasonal oscillation of eastern Pacific sea levels is forced, as described by Spillane et al. for 1971–75. The oscillation was also present in 1980–84 from Callao, Peru, to San Francisco, with maximum energy near 52–57 days and band limits of 43 and 65 days. During 1980–84 there was no evidence for forcing of the large-scale oscillation in the eastern Pacific, although a local contribution of forcing was superimposed on the remote signal at the California stations. Interannual fluctuations in amplitude were evident in the sea level time series, consistent with those of the corresponding wind oscillation in the western equatorial Pacific. The oscillation was best developed in both variables in 1980–82 and became weak or nonexistent during the recovery phase of the 1982–83 El Niñ, similar to a weakening that occurred following the 1972–73 episode, noted by Spillane et al. The sea level oscillations have the characteristics of lowest baroclinic mode Kelvin waves that are primarily forced by a similar, energetic oscillation in the winds in the western equatorial Pacific. During the 1980–82 period a significant component of the wind signal extended into the central Pacific and was associated with sea level propagation speeds of about 5 m s−1, suggesting a more extensive forcing along the equatorial waveguide at that time. In 1982–84, when the oscillation was weak, the sea level propagation was about 3 m s−1, consistent with the free propagation of lowest baroclinic mode Kelvin waves in the central Pacific.

Full access
Chunzai Wang and David B. Enfield

Abstract

Variability of the tropical Western Hemisphere warm pool (WHWP) of water warmer than 28.5°C, which extends seasonally over parts of the eastern North Pacific, the Gulf of Mexico, the Caribbean, and the western tropical North Atlantic (TNA), was previously studied by Wang and Enfield using the da Silva data from 1945–93. Using additional datasets of the NCEP–NCAR reanalysis field and the NCEP SST from 1950–99, and the Levitus climatological subsurface temperature, the present paper confirms and extends the previous study of Wang and Enfield. The WHWP alternates with northern South America as the seasonal heating source for the Walker and Hadley circulations in the Western Hemisphere. During the boreal winter a strong Hadley cell emanates northward from the Amazon heat source with subsidence over the subtropical North Atlantic north of 20°N, sustaining a strong North Atlantic anticyclone and associated northeast (NE) trade winds over its southern limb in the TNA. This circulation, including the NE trades, is weakened during Pacific El Niño winters and results in a spring warming of the TNA, which in turn induces the development of an unusually large summer warm pool and a wetter Caribbean rainy season. As the WHWP develops in the late boreal spring, the center of tropospheric heating and convection shifts to the WHWP region, whence the summer Hadley circulation emanates from the WHWP and forks into the subsidence regions of the subtropical South Atlantic and South Pacific. During the summers following El Niño, when the warm pool is larger than normal, the increased Hadley flow into the subtropical South Pacific reinforces the South Pacific anticyclone and trade winds, probably playing a role in the transition back to the cool phase of ENSO.

Seasonally, surface heat fluxes seem to be primarily responsible for warming of the WHWP. Interannually, all of the datasets suggest that a positive ocean–atmosphere feedback through longwave radiation and associated cloudiness seems to operate in the WHWP. During the winter preceding a large warm pool, there is a strong weakening of the Hadley cell that serves as a “tropospheric bridge” for transferring El Niño effects to the Atlantic sector and inducing warming of the warm pool. Associated with the warm SST anomalies is a decrease in sea level pressure 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. This data-inferred hypothesis of the longwave radiation feedback process needs to be further investigated for its validation in the WHWP.

Full access
David B. Enfield and Eric J. Alfaro

Abstract

Seasonally stratified analyses of rainfall anomalies over the intra-Americas sea and surrounding land areas and of onset and end dates of the Central American rainy season show that the variability of the tropical Atlantic sea surface temperature anomaly (SSTA) is more strongly associated with rainfall over the Caribbean and Central America than is tropical eastern Pacific SSTA. Seasonal differences include the importance of antisymmetric configurations of tropical Atlantic SSTA in the dry season but not in the rainy season. Both oceans are related to rainfall, but the strength of the rainfall response appears to depend on how SSTA in the tropical Atlantic and eastern Pacific combine. The strongest response occurs when the tropical Atlantic is in the configuration of a meridional dipole (antisymmetric across the ITCZ) and the eastern tropical Pacific is of opposite sign to the tropical North Atlantic. When the tropical North Atlantic and tropical Pacific are of the same sign, the rainfall response is weaker. The rainy season in lower Central America tends to start early and end late in years that begin with warm SSTs in the tropical North Atlantic, and the end dates are also delayed when the eastern equatorial Pacific is cool. This enhancement of date departures for zonally antisymmetric configurations of SSTA between the North Atlantic and Pacific is qualitatively consistent with the results for rainfall anomalies.

Full access
David B. Enfield and Alberto M. Mestas-Nuñez

Abstract

El Niño–Southern Oscillation (ENSO) is a global phenomenon with significant phase propagation within and between basins. This is captured and described in the first mode of a complex empirical orthogonal function (CEOF) analysis of sea surface temperature anomaly (SSTA) from the midnineteenth century through 1991. The global ENSO from the SSTA data, plus a linear trend everywhere, are subsequently removed in order to consider other global modes of variability uncontaminated by the intra- and interbasin effects of ENSO. An ordinary EOF analysis of the SSTA residuals reveals three non-ENSO modes of low-frequency variability that are related to slow oceanic and climate signals described in the literature. The first two modes have decadal to multidecadal timescales with high loadings in the Pacific. They bear some spatial similarities to the ENSO pattern but are broader, more intense at high latitudes, and differ in the time domain. A CEOF analysis confirms that they are not merely the phase-related components of a single mode and that all three modes are without significant phase propagation. The third mode is a multidecadal signal with maximal realization in the extratropical North Atlantic southeast of Greenland. It is consistent with studies that have documented connections between North Atlantic SSTA and the tropospheric North Atlantic Oscillation (NAO).

All three SSTA modes have midtropospheric associations related to previously classified Northern Hemisphere teleconnection patterns. The relationships between SSTA modes and tropospheric patterns are consistent with the ocean–atmosphere interactions discussed in previous studies to explain low-frequency climate oscillations in the North Pacific and North Atlantic sectors. The first three leading modes of non-ENSO SSTA are most related, to the tropospheric patterns of the Pacific North American, the North Pacific, and the Arctic oscillations (AO), respectively. The 500-hPa pattern associated with the third SSTA mode also bears similarities to the NAO in its Atlantic sector. This North Atlantic mode has a region of high, positive SSTA loadings in the Gulf of Alaska, which appear to be connected to the North Atlantic SSTA by a tropospheric bridge effect in the AO.

Full access
David B. Enfield and Sang-ki Lee

Abstract

The thermodynamic development of the Western Hemisphere warm pool and its four geographic subregions are analyzed. The subregional warm pools of the eastern North Pacific and equatorial Atlantic are best developed in the boreal spring, while in the Gulf of Mexico and Caribbean, the highest temperatures prevail during the early and late summer, respectively. For the defining isotherms chosen (≥27.5°, ≥28.0°, ≥28.5°C) the warm pool depths are similar to the mixed-layer depth (20–40 m) but are considerably less than the Indo–Pacific warm pool depth (50–60 m). The heat balance of the WHWP subregions is examined through two successive types of analysis: first by considering a changing volume (“bubble”) bounded by constant temperature wherein advective fluxes disappear and diffusive fluxes can be estimated as a residual, and second by considering a slab layer of constant dimensions with the bubble diffusion estimates as an additional input and the advective heat flux divergence as a residual output. From this sequential procedure it is possible to disqualify as being physically inconsistent four of seven surface heat flux climatologies: the NCEP–NCAR reanalysis (NCEP1) and the ECMWF 15-yr global reanalysis (ERA-15) because they yield a nonphysical diffusion of heat into the warm pools from their cooler surroundings, and the unconstrained da Silva and Southampton datasets because their estimated diffusion rates are inconsistent with the smaller rates of the better understood Indo–Pacific warm pool when the bubble analysis is applied to both regions. The remaining surface flux datasets of da Silva and Southampton (constrained) and Oberhuber have a much narrower range of slab surface warming (+25 ± 5 W m−2) associated with bubble residual estimates of total diffusion of –5 to –20 W m−2 (±5 W m−2) and total advective heat flux divergence of –2 to –14 W m−2 (±5 W m−2). The latter are independently confirmed by direct estimates using wind stress data and drifters for the Gulf of Mexico and eastern North Pacific subregions.

Full access
David B. Enfield and Luis Cid S.

Abstract

Although there are indications from numerical models that El Niño-Southern Oscillation (ENSO) may be an internal mode of the coupled Pacific ocean-atmosphere system, sensitive to climatic background parameters, it has not yet been possible to find significant changes in ENSO variability between the Little Ice Age and the present. Yet a number of authors have found qualitative indications in anecdotal and proxy records of shorter, century-scale variations in the return-interval statistics for El Niño episodes. To objectively determine what nonstationarities exist, we statistically examine the El Niño occurrences since 1525, compiled by Quinn et al. We have stratified the return intervals both for strong events and for all events according to two null hypotheses. 1) return intervals are stationary over periods of 200–500 years, and 2) the intervals are stationary on a centenary time scale, between epochs of contrasting solar variability. Two-parameter Weibull distributions are fit to subsamples of the data using an optimized bootstrap procedure, and the scale parameters are compared between groups. At the 95% significance level, only the null hypothesis for high/low solar levels and strong El Niño events can be rejected. The corresponding hypothesis for all events rejects at the 90% level, while overall stationarity cannot be rejected at any reasonable level, for either class of events. The significant results are that 1) the El Niño recurrence rate is stationary with respect to long-term climate changes, while 2) return intervals of strong El Niño events are significantly nonstationary at centenary time scales and 3) events of all intensifies exhibit the same nonstationarity but less clearly. There is too little data to reject the possibility that the association with solar epochs is coincidental, however, we have advanced a hypothesis to explain such a connection.

Full access
Alberto M. Mestas-Nuñez and David B. Enfield

Abstract

A varimax rotation was applied to the EOF modes of global SST derived by Enfield and Mestas-Nuñez. The SST anomaly record is more than a century long, with a global complex EOF representation of ENSO and a linear trend removed at every grid point. The rotated EOF modes capture localized centers of variability that contribute to the larger-scale spatial patterns of the unrotated modes. The first rotated EOF represents a multidecadal signal with larger response in the North Atlantic. The second rotated EOF represents an interdecadal fluctuation with larger response in the eastern North Pacific and out of phase fluctuations of smaller amplitude in the central North Pacific. The third rotated EOF captures interdecadal fluctuations in the eastern tropical Pacific with a dominant peak that coincides with the 1982/83 ENSO. The fourth rotated EOF has interdecadal to multidecadal nature with larger response in the central equatorial Pacific and quasi-symmetric out-of-phase response in the western North and South Pacific. The fifth mode represents multidecadal fluctuations with large response at about 40°N in the North Pacific. The sixth mode has interannual to interdecadal timescales with largest response confined to the South Atlantic. The authors’ rotated modes are dominated by intra- rather than interocean fluctuations supporting the hypothesis that the non-ENSO variability is more regional than global in nature. Analyses of sea level pressure and surface wind stress show that in general the non-ENSO rotated EOFs are consistent with an ocean response to local atmospheric forcing. An exception is the eastern tropical Pacific mode, which is more consistent with an atmospheric response to changes in the ocean SST.

Full access
David B. Enfield and J. S. Allen

Abstract

Case history analysis, cross spectra and multiple regression analysis have been used in a study of low-pass filtered sea level records from the Pacific mainland coast of Mexico in 1971 and 1973–75. During the summer-fall season (May–October), sea level variability is characterized by strong alongshore coherence and nondispersive, poleward phase propagation over a wide frequency range (0.02–0.37 cpd). The strength and clarity of the propagating signals seem to be related primarily to large-amplitude events of elevation (10–30 cm) that are generated off the southern coast of Mexico by tropical storms. These events are typically forced by the alongshore, poleward movements of the storms to as far north as 20°N, and thereafter continue to propagate freely at least as far as Guyamas (28°N). Large, variable phase speeds (250–500 km day−1 are observed in the southern region, consistent with the alongshore speeds of the forcing. A multiple-input statistical forcing model, in which adjusted sea level is regressed on local wind, large-scale atmospheric pressure and remotes sea level and wind, confirms that the disturbances are forced in the south and propagate freely in the north. North of 20°N, propagation speeds are similar than in the south and relatively invariant at each station, and show a steady decrease from 250–300 km day−1 near Mazatlan (23°N) to 180–230 km day−1 near Guaymas (28°N). These characteristics of the northern propagation are consistent with those from the theory of free, linear, hybrid coastal trapped waves, as computed from an inviscid numerical model by Brink (1982). The observed speeds in the south, however, are much faster than predicted by theory, consistent with their forced nature. The model results show a strongly barotropic velocity structure over the continental shelf, and a baroclinic structure farther offshore. In winter (November–April), the alongshore coherence of sea level is less than in summer at all stations, and only appreciable between Acapulco (17°N) and Mazatlan (23°N). The winter phase propagation (140–230 km day−1) is generally slower than in summer, most notably in the south, where it is consistent with a lack of forcing.

Full access