Abstract

The realistic simulation of El Niño–Southern Oscillation (ENSO) by the University of California, Los Angeles (UCLA), coupled atmosphere–ocean general circulation model (CGCM) is used to test two simple theoretical models of the phenomenon: the recharge oscillator model of Jin and the delayed oscillator model of Schopf, Suarez, Battisti, and Hirst (SSBH). The target for the simple models is provided by the CGCM results prefiltered with singular spectrum analysis to extract the leading oscillatory mode. In its simplest form, the Jin model can be reduced to two first ordinary differential equations. If the parameters of the model are fit in this reduced form, it appears to capture the period of the CGCM oscillatory mode. If the Jin model is instead fit using the individual physical balances that are used to derive it, substantial misfits to the CGCM are encountered. The SSBH model can likewise be expressed either in a condensed form or a larger set of individual physical balances with highly analogous results.

It is shown that the misfits in both simple models can be greatly reduced by introducing a spinup timescale for wind stress relative to eastern equatorial Pacific SST. In the CGCM, this spinup time appears to be associated with a combination of atmospheric and ocean mixed layer processes in a way consistent with the “mixed mode” regime discussed by Syu and Neelin, which is not included in the Jin and SSBH models. These appear indistinguishable in this analysis, although the latter is more sensitive to fitting.

This paper provides a bridge between work on ENSO by theoreticians and numerical modelers. The CGCM results validate the conceptual framework of the simple models by demonstrating that they can provide a plausible representation of ENSO with realistic sets of parameters. The results also suggest that, in terms of realistic ENSO variability, the framework of the simple models can be made substantially more complete by including the adjustment time between wind stress and eastern Pacific SST required by the coupled spinup of the atmosphere and the ocean mixed layer processes outside this region.

Abstract

This study focuses on precipitation in Uruguay and the Brazilian state of Rio Grande do Sul, which extend along the Atlantic coast of southern South America. The present paper has two principal goals: 1) to describe the annual cycle of precipitation and 2) to investigate the relationships between its anomalies and those in sea surface temperature (SST) in the Pacific and Atlantic oceans. The dataset is provided by 40 rainfall stations almost evenly distributed in space and covers the period 1917–80. The tools used in support of this research include principal component and canonical correlation analyses.

It is found that total precipitation tends to be evenly distributed during the year. The largest spatial variability in the monthly deviations from the annual mean appears as a west–east (inland–coastal) dipole with the largest positive values in the west during early fall and midspring, and in the east along the Atlantic coast during winter. The second mode of rainfall variability appears as a north–south dipole with the largest positive values in the south during late summer and late fall, and in the north during early spring and early summer. The third mode appears primarily as a north–south dipole along the western boundary with the largest positive values in the southwest during fall and in the northwest during early spring. These modes explain 60%, 19%, and 8% of the total variance. Five subregions are identified according to similarities between the characteristics of the annual cycles in their rainfall stations.

It is shown that there are significant relationships between anomalies in rainfall and in SST in the Pacific and Atlantic oceans. Some of these relationships confirm the results of previous studies, such as the links between the El Niño–Southern Oscillation phenomenon in the equatorial Pacific Ocean and rainfall anomalies in Uruguay during late spring–early summer and late fall–early winter. Other relationships have not been reported before, such as the links between SST anomalies in the southwestern Atlantic Ocean and rainfall anomalies in the entire region during October–December and April–July. It is also found that when SST anomalies are considered in both oceans simultaneously, their links with rainfall anomalies are in some cases enhanced and in others weakened.

Abstract

The influence of Atlantic sea surface temperature (SST) anomalies on the atmospheric circulation over the North Atlantic sector during winter is investigated by performing experiments with an atmospheric general circulation model. These consist of a 30-yr run with observed SST anomalies for the period 1961–90 confined geographically to the Atlantic Ocean, and of a control run with climatological SSTs prescribed globally. A third 30-yr integration with observed SSTs confined to the South Atlantic is made to confirm present findings.

The simulated interannual variance of 500-hPa wintertime geopotential heights over the North Atlantic attains much more realistic values when observed Atlantic SSTs are prescribed. Circulation patterns that resemble the positive phase of the North Atlantic oscillation (NAO) become more pronounced in terms of the leading EOF of winter means, and a cluster analysis of daily fields. The variance of an interannual NAO index increases by fivefold over its control value. Atlantic SST variability is also found to produce an appreciable rectified response in the December–February time mean.

Interannual fluctuations in the simulated NAO are found to be significantly correlated with SST anomalies over the tropical and subtropical South Atlantic. These SST anomalies are accompanied by displacements in the simulated summer monsoonal circulation over South America and the cross-equatorial regional Hadley circulation.

Abstract

The El Niño–Southern Oscillation (ENSO) has an established impact on precipitation in Uruguay during austral spring (October–December). This impact is absent in peak summer (January–February), and returns weakly in fall–winter (March–July). Interannual and intraseasonal variability of the atmospheric circulation over South America and the South Pacific is investigated using the NCEP–NCAR reanalysis data for these seasons. The leading empirical orthogonal function (EOF) of seasonally averaged 200-hPa winds over South America is found to be associated with ENSO through a pronounced Walker cell component in all seasons. However, during spring, this pattern acquires an extratropical teleconnection that links the circulation over southeastern South America (SESA) with ENSO. This extratropical teleconnection disappears in summer, when the circulation over SESA is dominated by variability in the South Atlantic convergence zone. In fall, extratropical South America again becomes affected by a wavelike pattern that extends over the South Pacific, but it is uncorrelated with ENSO.

On intraseasonal timescales, a cluster analysis of daily geopotential height fields over the South Pacific sector reveals three wave train–like circulation regimes with similar structures in all seasons. During the transition seasons (but not summer), the frequencies of occurrence of two of these regimes are found to be significantly different from normal in years when the interannual wavelike EOFs are pronounced. Interannual anomaly patterns are constructed from the intraseasonal regimes according to these changes in their frequency of occurrence, and shown to resemble quite closely the interannual EOFs over the South Pacific sector. These results provide evidence that the interannual teleconnection patterns seen over the South Pacific in austral spring and fall–winter are due to changes in the frequency of occurrence and amplitude of intraseasonal circulation regimes. The Rossby wave source composited over ENSO years suggests that ENSO heating anomalies are able to trigger these changes in regime occurrence and amplitude during October–December through Rossby wave propagation, leading to the known ENSO teleconnection in austral spring. By contrast, the interannual teleconnection over the South Pacific during fall–winter appears to be due to essentially random changes in the frequency of occurrence of the intraseasonal circulation regimes, which are found to be much larger than during austral summer when no extratropical teleconnection pattern exists.

Abstract

The sensitivity of the atmospheric circulation to sea surface temperature (SST) anomalies in the tropical and subtropical South Atlantic Ocean is studied by means of simulations with an atmospheric general circulation model (GCM). Two types of prescribed SST anomalies are used, motivated by previous analyses of data. The first occurs during austral summers in association with a strengthening of the South Atlantic convergence zone (SACZ) and consists of cold SST anomalies over the subtropical South Atlantic. The second is the leading seasonally varying empirical orthogonal function of SST, consisting of warm basin-scale anomalies with maximum amplitude in the subtropics during January–March and at the equator in June. An ensemble of about 10 seasonal simulations is made using each type of anomaly, focusing on the January–March period in the first case and the January–June seasonal evolution in the second.

During January–March both experiments yield a statistically significant baroclinic response over the subtropical Atlantic with dipolar SACZ-like anomalies. Some evidence of positive feedback is found. The response is shown to be fairly similar in pattern as well as amplitude to the linear regression of observed interannual low-level wind anomalies with subtropical SST anomalies. However, in the first experiment with cold SST anomalies, the simulated response contrasts with the leading interannual mode of observed SACZ variability.

Warm basin-scale anomalies are found to have their largest impact during boreal summer, with a strong statistically significant equatorial baroclinic response and positive rainfall anomalies over the equatorial ocean. The latter do not extend appreciably into the adjacent continents, although there are significant positive rainfall anomalies over the Sahel in April–June and negative anomalies over the western Indian Ocean. In the upper troposphere, a statistically significant wave train extends southwestward to southern South America and northeastward to Europe in April–June, while there is some linkage between the tropically and subtropically forced responses during January–March.

Abstract

Variations of zonal mean and eddy properties in the Southern Hemisphere during the winter of 1979 are studied. Several periods of enhanced wave activity and rapid zonal-mean changes are observed. During these periods of rapid change both the actual zonal wind tendency and the Eliassen-Palm flux divergence have a dipole pattern in the upper stratosphere, with positive values in high latitudes and negative values in middle and low latitudes. The positive values of E-P flux divergence near the pole are particularly interesting, since they suggest a possible source of wave activity in this region. The stationary component of wavenumber 1 and the eastward traveling component of wavenumber 2 contribute most to the dipole pattern of wave driving in the stratosphere.

A nearly uniform deceleration of the mean flow in the troposphere is contributed to by all zonal wavenumbers from 1 to 10. The lower wavenumbers contribute most to the driving in high latitudes, while the driving in lower latitudes is contributed mostly by high wavenumbers. Wave forcing of the mean flow as measured by the E-P flux divergence and actual changes in the mean flow are correlated with each other, but the actual mean flow tendencies are often significantly smaller than the eddy driving. The largest correlations between E-P flux divergence and the observed zonal mean wind changes exceed 0.5 and occur both in the upper stratosphere near the jet core and near the tropopause across a broad range of latitudes.

Abstract

The climatology and interannual variability of wave–mean flow interaction in the Southern Hemisphere (20–80°S, 0–55 km) is described for the winter months of June–September based on a sample of four years, 1979–82. The stratospheric jet stream shifts downward and poleward over the course of the winter in response to seasonal variations in thermal forcing. The shift occurs at different times in different years, however, so that the months of July and August show substantial interannual variability of monthly mean zonal winds. The poleward and downward shift of the jet axis in an individual year is usually abrupt and occurs in association with a burst of upwardly propagating planetary waves. The driving of the mean flow in the stratosphere generally has a dipolar structure with easterly accelerations near 40°S and westerly accelerations in polar latitudes. The structure of the wave driving is consistent with the structure of the observed mean flow accelerations.

Abstract

The intraseasonal variability of the Southern Hemisphere stratosphere and troposphere is studied using multilevel geopotential height data for nine winters (1979–87). The study uses empirical orthogonal function (EOF) analysis of unfiltered data at five tropospheric and five stratospheric levels.

The four leading EOFs at all tropospheric levels exhibit the patterns previously detected at 500 mb. Study of the corresponding principal components (PCs) at each level shows that the quasi-stationary anomalies associated with the leading EOFs are equivalent barotropic and exhibit no preference for early, middle or late winter.

The five leading EOFs in the stratosphere fall into two classes. The first three EOFs at all levels form the first class. This class represents anomalies that are dominated by zonal wavenumber one (wave 1), exhibit strong westward tilt with height and travel slowly eastward or remain stationary. Most cases of large, persistent PC values for this class occur in early winter. The fourth and fifth EOFs form the other class. This class represents anomalies that are dominated by wavenumber two, and tilt noticeably, but less strongly than the first class, westward with height. These anomalies tend to develop mostly in late winter and to travel eastward more rapidly. The intraseasonal variability in the stratosphere resides therewith, as expected, in structures dominated by the longest planetary waves.

No systematic connections between tropospheric and stratospheric persistent anomalies are apparent in the dataset.

Abstract

The early winter (mid-April to mid-July) circulation in the Southern Hemisphere stratosphere is studied. Emphasis is placed on the evolution of strong disturbances with structures dominated by the zonal wavenumber-1 component of the flow (wave 1). The approach to this investigation is based on analysis of 12 years (1979–90) of observational data and comparative analyses of control and hypothesis-testing simulations with a three-dimensional primitive equation model of the stratosphere and mesosphere.Considerable interannual variability is found in both the intensity and timing of wave-1 amplification during early winter. Though usually quasi-stationary, there are six extended periods in the dataset when wave 1 travels steadily eastward and is of large amplitude. Two of these periods (June 1980 and June 1985) are examined in detail. The evolution of the circulation in these two cases resembles that during Canadian warmings in the Northern Hemisphere in several ways. First, there is a large, eastward-moving disturbance with a nearly equivalent barotropic structure, with the largest amplitude in the lower and middle stratosphere. Second, temperature increases are smaller than those observed during final warmings in the Southern Hemisphere. Third, irreversible buckling of contours of Ertel's potential vorticity takes place in a region well away from the zero-wind line. Owing to their geographical preference for development over the South Pacific, wave-1 events in the southern stratosphere during early winter are referred to as South Pacific warmings.The hypothesis-testing simulations suggest that the development of South Pacific warmings is connected with the amplification of wave 1 at 100 mb and that the eastward propagation of the disturbances requires eastward propagation of wave 1 at 100 mb. In addition, the results suggest that development of stratospheric disturbances in the southern stratosphere during early winter depends more on the intensity of wave 1 at 100 mb than on the structure of the zonal-mean flow in the stratosphere.

Abstract

The South Atlantic anticyclone is a major feature of the austral winter climatology. An atmospheric general circulation model (AGCM) is used to study the dynamics of the South Atlantic anticyclone by means of control simulations and experiments to investigate sensitivity to prescribed orography, sea surface temperatures, and soil wetness. The South Atlantic anticyclone in the first control simulation is unrealistically zonally elongated and centered too far west—errors typical of coupled ocean–atmosphere GCMs. Results of the sensitivity experiments suggest that these deficiencies are associated with another family of systematic model errors: the overprediction of convection over the tropical land surfaces, particularly over eastern tropical Africa and India, and the concurrent large-scale westward shift in the divergence center at upper levels and the convergence center at lower levels. The results also confirm the important role of South American and African orography in localizing the South Atlantic anticyclone over the ocean. Other factors, however, like the regional zonal gradients of sea surface temperatures, are found to have only a minor impact on the anticyclone. To further substantiate these findings, the wintertime anticyclone is examined using a revised version of the atmospheric GCM. Improvements are found in both the anticyclone as well as the Asia–African summer monsoon circulations. The results demonstrate the existence of links between intensity and structure of the wintertime South Atlantic anticyclone and the major summer monsoons in the Northern Hemisphere.