Thermodynamic Mechanisms Responsible for the Tropospheric Response to SST Anomalies in the Antarctic Circumpolar Wave

Warren B. White Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California

Search for other papers by Warren B. White in
Current site
Google Scholar
PubMed
Close
and
Shyh-Chin Chen Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California

Search for other papers by Shyh-Chin Chen in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Tropospheric temperature and vorticity budgets for the Antarctic Circumpolar Wave (ACW) are diagnosed utilizing the National Centers for Environment Prediction–National Center for Atmospheric Research reanalysis datasets from 1983 to 1992, focusing on the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean where remote forcing from the Tropics has been observed to be weak. There, warm sea surface temperature (SST) anomalies are found in the ACW propagating eastward together with anomalous upward latent heat flux, positive precipitation, low-level convergence, upper-level divergence, midlevel ascent, and poleward surface wind. Diagnosing the anomalous temperature budget finds SST-induced latent heat flux instigating anomalous mid- to upper-level diabatic heating and low-level diabatic cooling in the absence of significant eddy heat flux divergence. This diabatic heating profile is balanced by a combination of vertical and horizontal heat advection, giving rise to anomalous ascent and poleward wind throughout the column. The thermodynamics of this deep diabatic heating scenario are different from those of Palmer and Sun. An intrinsic feedback from atmosphere to ocean is indicated by reduced sensible-plus-latent heat flux displaced 45° to 90° of phase to the east of warm SST anomalies, yielding an anomalous SST warming tendency that contributes both to eastward phase propagation and amplitude maintenance of the ACW. Diagnosing the anomalous potential vorticity budget finds the vertical gradient of anomalous diabatic heating, negative over most of the column, balanced by the anomalous advection of planetary vorticity, the mean advection of anomalous relative vorticity, and net vortex tube advection, together yielding a poleward equivalently barotropic wind response to warm SST anomalies. This deep diabatic heating scenario is contrasted against the remote forcing scenario in the eastern Pacific and western Atlantic sectors of the Southern Ocean where remote forcing associated with the El Niño–Southern Oscillation (ENSO) in the Tropics can now be seen to drive the SST tendency in the ACW.

Corresponding author address: Dr. Warren B. White, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093-0230. Email: wbwhite@ucsd.edu

Abstract

Tropospheric temperature and vorticity budgets for the Antarctic Circumpolar Wave (ACW) are diagnosed utilizing the National Centers for Environment Prediction–National Center for Atmospheric Research reanalysis datasets from 1983 to 1992, focusing on the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean where remote forcing from the Tropics has been observed to be weak. There, warm sea surface temperature (SST) anomalies are found in the ACW propagating eastward together with anomalous upward latent heat flux, positive precipitation, low-level convergence, upper-level divergence, midlevel ascent, and poleward surface wind. Diagnosing the anomalous temperature budget finds SST-induced latent heat flux instigating anomalous mid- to upper-level diabatic heating and low-level diabatic cooling in the absence of significant eddy heat flux divergence. This diabatic heating profile is balanced by a combination of vertical and horizontal heat advection, giving rise to anomalous ascent and poleward wind throughout the column. The thermodynamics of this deep diabatic heating scenario are different from those of Palmer and Sun. An intrinsic feedback from atmosphere to ocean is indicated by reduced sensible-plus-latent heat flux displaced 45° to 90° of phase to the east of warm SST anomalies, yielding an anomalous SST warming tendency that contributes both to eastward phase propagation and amplitude maintenance of the ACW. Diagnosing the anomalous potential vorticity budget finds the vertical gradient of anomalous diabatic heating, negative over most of the column, balanced by the anomalous advection of planetary vorticity, the mean advection of anomalous relative vorticity, and net vortex tube advection, together yielding a poleward equivalently barotropic wind response to warm SST anomalies. This deep diabatic heating scenario is contrasted against the remote forcing scenario in the eastern Pacific and western Atlantic sectors of the Southern Ocean where remote forcing associated with the El Niño–Southern Oscillation (ENSO) in the Tropics can now be seen to drive the SST tendency in the ACW.

Corresponding author address: Dr. Warren B. White, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093-0230. Email: wbwhite@ucsd.edu

1. Introduction

Namias (1972) hypothesized that the large heat capacity in the upper ocean associated with sea surface temperature (SST) anomalies in the midlatitude North Pacific Ocean could produce significant and persistent influences on the tropospheric circulation directly overhead and downstream over North America. This proposed tropospheric response was characterized by high sea level pressure (SLP) anomalies forming ∼90° of phase to the east of warm SST anomalies on monthly, seasonal, and interannual period scales. Davis (1976) challenged this hypothesis, finding monthly SST anomalies temporally lagging monthly SLP anomalies, indicating the opposite response. Wallace and Jiang (1987) embraced both hypotheses within a coupled ocean–atmosphere scenario. They proposed that monthly and seasonal SLP anomalies occur in quasi-stationary equilibrium with SST-induced sensible-plus-latent heat flux anomalies, while the anomalous SST tendency is driven by wind- and air temperature–induced sensible-plus-latent heat fluxes associated with the SLP anomalies.

The advent of satellite SST and tropospheric profiling over the Southern Ocean in the early 1980s allowed year-to-year climate variability in the upper ocean and lower atmosphere to be detected and resolved. White and Peterson (1996) and Jacobs and Mitchell (1996) found monthly SST and SLP anomalies in the Southern Ocean dominated by interannual signals of ∼4-yr-period scale, propagating slowly eastward around the Southern Ocean in fixed phase with one another. They found both variables of global zonal wavenumber-2 scale taking ∼8 yr to circle the globe. They called this interannual signal in covarying SST and SLP anomalies the Antarctic Circumpolar Wave (ACW). Curiously, high SLP anomalies in the ACW were found displaced ∼90° of phase to the east of warm SST anomalies as observed by Namias (1972) and Wallace and Jiang (1987) in the midlatitude North Pacific Ocean.

Subsequently, following the lead of Wallace and Jiang (1987), analytical ocean–atmosphere coupled models of the ACW were constructed by Qiu and Jin (1997), White et al. (1998), and Baines and Cai (2000). These models yielded both eastward phase propagation and amplitude maintenance of the ACW, but utilized different thermodynamic mechanisms to achieve similar ACW characteristics (e.g., Talley 1999). This emphasized the need for diagnosing the thermodynamics involved in the coupling. Thus, in the present study we diagnose the anomalous tropospheric temperature and vorticity budgets of the ACW in the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis dataset (Kalnay et al. 1996).

Detection of the tropospheric response to SST anomalies in the ACW is complicated by remote forcing stemming from interannual SST anomalies in the western tropical Pacific Ocean associated with the El Niño–Southern Oscillation (ENSO). The ENSO has been observed producing meridional atmospheric teleconnections through a variety of transfer mechanisms (e.g., Horel and Wallace 1981; Wallace and Gutzler 1981; Karoly 1989; Lau and Nath 1990; Sardeshmukh and Hoskins 1991; White et al. 2002). This remote forcing yields mid- and high-latitude SLP anomalies, which in turn drive an SST tendency through air–sea fluxes of heat, momentum, and kinetic energy (e.g., Kushnir and Held 1996). Karoly (1989) found the standing mode of ENSO in the Tropics generating a standing mode of SLP anomaly in the eastern Pacific and western Atlantic sectors of the Southern Ocean, known as the Pacific–South America (PSA) pattern. Cai and Baines (2001) proposed that during the transition between El Niño and La Niña, when this remote forcing becomes weak or absent, the covarying SST and SLP anomalies that form the PSA subsequently propagates eastward into the eastern Atlantic, Indian, and western Pacific sectors of the Southern Ocean as the ACW. As a variation on this theme, White et al. (2002) found the ACW in the eastern Pacific and western Atlantic sectors to be in damped resonance with remote forcing stemming from the slow eastward phase propagation of covarying SST and SLP anomalies in the global ENSO wave (GEW; White and Cayan 2000) across the western tropical Pacific Ocean. Both Cai and Baines (2001) and White et al. (2002) found these meridional atmosphere teleconnections weakly influencing troposphere circulation in the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean. In the present study, we find the thermodynamics of the ACW over this latter domain to be very different from those in the eastern Pacific and Atlantic sectors.

White et al. (2002) also computed the phase velocities of interannual SLP and SST anomalies over the Southern Ocean, finding the ACW propagating eastward along a path different from that of the Antarctic Circumpolar Current (ACC). In the initial discovery of the ACW (White and Peterson 1996; Jacobs and Mitchell 1996), the eastward advection of SST anomalies in the ACW by the ACC was assumed. This assumption appears to have been incorrect. These phase velocities yield a path for the ACW that follows that of maximum autumn–winter cyclogenesis density between 35° and 50°S (Simmonds and Keay 2000), traveling north of the ACC at a much slower speed. The exception to this occurs in the vicinity of Drake Passage where the paths of the ACC and the ACW become aligned. In the present study, we find the SST anomalies following the path of the ACW across the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean (where meridional teleconnections are weak) not only in fixed phase with SLP anomalies, but in fixed phase with anomalous latent heat flux, precipitation, low-level divergence, upper-level divergence, and midlevel ascent. These phase relationships are very different from those computed over the path segment in the eastern Pacific and western Atlantic sectors, the latter region dominated by remote forcing from the Tropics.

Earlier Palmer and Sun (1985) examined the temperature and vorticity budgets in the tropospheric response to SST anomalies in the midlatitude North Atlantic Ocean, differentiating between the shallow diabatic heating scenario of Hoskins and Karoly (1981) and a deep diabatic heating scenario operating in atmosphere general circulation model (AGCM) simulations. To balance the temperature and vorticity budgets in the shallow diabatic heating scenario, Hoskins and Karoly (1981) found a cold-core high SLP anomaly displaced 90° of phase to the west of a warm SST anomaly. This is very different from the situation observed by Namias (1972) and Wallace and Jiang (1987) in the midlatitude North Pacific Ocean, by Ratcliffe and Murray (1970) and Palmer and Sun (1985) in the midlatitude North Atlantic Ocean, and by White and Peterson (1996) in the Southern Ocean. On the other hand, the deep diabatic heating scenario proposed by Palmer and Sun (1985) does explain these observations, with a warm-core high SLP anomaly displaced 90° of phase to the east of warm SST anomaly. They proposed that warm SST anomalies induce positive diabatic heating anomalies throughout the troposphere, with net anomalous heating raised to midlevel through low-level cooling by anomalous eddy heat flux divergences. In the present study, we find a deep diabatic heating scenario operating in the ACW over the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean. Yet, the thermodynamics are different from those proposed by Palmer and Sun (1985). We find warm SST anomalies associated with mid- to upper-level diabatic heating anomalies in the absence of a significant anomalous eddy heat flux divergence.

The extratropical troposphere response to monthly SST anomalies has been examined more recently by Peng et al. (1995, 1997), and Peng and Whitaker (1999) in AGCM studies of the western midlatitude North Pacific and North Atlantic oceans during autumn–winter. They found warm SST anomalies producing different tropospheric responses in different months, depending on the influences that different basic states have on the transient eddy interaction with the weak SST-induced circulation. In the present study, we find warm SST anomalies coinciding with anomalous mid- to upper-level diabatic heating and low-level cooling throughout the available record. We find this profile of diabatic heating balanced by both vertical and horizontal heat advection, yielding anomalous ascent and poleward wind over the column. We also find an intrinsic feedback from atmosphere to ocean operating through the sensible-plus-latent heat flux anomaly, which yields an anomalous SST tendency that contributes both to eastward phase propagation and amplitude maintenance of the ACW. We contrast this deep diabatic heating scenario observed in the ACW over the eastern Atlantic, Indian, and western Pacific sectors of the Southern Ocean with a remote forcing scenario for the ACW in the eastern Pacific and western Atlantic sectors straddling the Drake Passage (Cai and Baines 2001; White et al. 2002).

2. Data and methods

In this study we analyze 10 variables, and their various horizontal derivatives, from two sources over the 18 yr from 1982 to 1999, focusing attention on the 10 yr from 1983 to 1992 when the ACW was particularly robust (White et al. 2002). We analyze monthly SST, SLP, sensible heat flux (QS), latent heat flux (QE), tropospheric temperature (T), tropospheric wind (V), tropospheric pressure velocity (ω), and the horizontal eddy heat flux divergence (Div〈V″T″〉, where V″ and T″ represent 6-hourly deviations about the monthly mean, and where 〈 〉 represents the monthly mean operator) from NCEP–NCAR reanalysis (Kalnay et al. 1996). Over the global ocean the NCEP–NCAR reanalysis incorporates the Comprehensive Ocean–Atmosphere Data Set (COADS) surface marine weather observations (Slutz et al. 1985), the Reynolds' SST dataset (Reynolds and Marsico 1993), and atmospheric soundings from weather ships and satellites (Kalnay et al. 1996). These in situ and satellite data are assimilated into the NCEP–NCAR reanalysis AGCM, made available monthly on a 2° longitude by 2° latitude grid extending over the globe from 90°S to 90°N. Over this same period, we analyze the NCEP Climate Prediction Center's (CPC's) Merged Analysis of Precipitation (CMAP) reanalysis (Xie and Arkin 1997). This CMAP dataset is made available monthly on a 2.5° latitude–longitude grid over the globe. Monthly precipitation (PCP) data are based on gauge observations over the land, and on satellite estimates, atoll gauge estimates, and numerical model output over the ocean, the latter derived from the NCEP–NCAR reanalysis.

Maximum standard errors for monthly NCEP–NCAR SST, SLP, QS, and QE, and CMAP PCP estimates are approximately ±0.2 K, ±0.2 hPa, ±20 W m−2, ±20 W m−2, and ±2 mm month−1, respectively. Monthly estimates were subsequently interpolated onto a common 2.5° latitude by 5° longitude grid determined by White (1995) to be optimal for resolving gyre-scale climate variability over the global upper ocean. This reduces random error by ∼1.4. Monthly anomalies for each variable were computed about long-term monthly means defining the mean annual cycle. As shown below, zonal wavenumber–frequency spectra of monthly SST and SLP anomalies in the ACW yield peak spectral energy density on interannual period scales of ∼4 yr, allowing us to isolate interannual signals from higher- and lower-frequency variability by bandpass filtering time sequences of monthly anomalies using a period admittance window with half-power points at 36 and 72 months (Kaylor 1977). This particular admittance window isolates interannual signals of 4–5-yr periodicity in the ACW from the weaker background variability. To reduce loss of data at the ends of each 18-yr time sequence due to high-pass filtering, maximum entropy spectral analysis was applied (Andersen 1974) using spectral coefficients to extend these time sequences by an amount equal to half the filter width; this procedure allows half the variance of the signal at the end points to be faithfully represented (White 2000). This filtering procedure reduced standard errors of monthly anomalies by another factor of ∼6 (i.e., the square root of ∼36 independent monthly estimates in the high-pass portion of the filter). Thus, spatial smoothing and bandpass filtering together reduce the standard errors for interannual anomalies to approximately ±0.02 K, ±0.02 hPa, ±2 W m−2, and ±0.2 mm month−1.

We diagnose anomalous tropospheric temperature and vorticity budgets in this study using T, V, Div〈V″T″〉, and ω data located at six levels in the troposphere (i.e., 1000, 850, 700, 500, 300, and 200 hPa). In these budgets, we compute the various terms using spherical harmonic smoothing, truncated to rhomboidal 5. The latter allows us to focus on the global space scales such as those associated with the ACW. The ∼4-yr period scale of the ACW allows two or three cycles of the signal to be sampled over the 10-yr record. Thus, the foregoing analysis must be viewed as a case study. We enter into this study with the objective of testing two different hypotheses and answering some fundamental questions. The two hypotheses are the shallow diabatic heating scenario of Hoskins and Karoly (1981) and the deep diabatic heating scenario of Palmer and Sun (1985). We gain additional confidence in our results by achieving closure of the anomalous vorticity budget and by finding internal consistency between the temperature, vorticity, and potential vorticity budgets.

3. Covarying SST and SLP anomalies in the ACW

We begin by displaying animation sequences of interannual SST and SLP anomalies over the 10-yr record (Fig. 1). Each map in the sequence extends meridionally from 30° to 70°S and around the globe zonally from 30° to 30°E. Here we can see individual SST and SLP anomalies of ∼4-yr periodicity taking ∼8 yr to circle the globe, traveling at a zonal average speed of 45° (lon) yr−1 (i.e., ∼0.08 m s−1 at 50°S). Following the cold SST anomaly in the Indian Ocean sector of the Southern Ocean from June 1984 to June 1992, it propagates slowly eastward along the south coast of Australia from 1985 to 1988, propagates eastward through the Tasman Sea and across New Zealand from 1987 to 1989, and across the western and central Pacific sectors from 1989 to 1991. Thereafter, it turns south and propagates eastward through the Drake Passage and on into the Atlantic sector by 1992, subsequently turning north to continue propagating eastward just south of southern Africa. A similar tracking follows the high SLP anomaly in the Indian sector of the Southern Ocean in June 1986, achieving the western Atlantic sector in 1993.

To estimate the relative intensity of propagating and standing wave activity in the ACW (Fig. 1), we compute zonal wavenumber–frequency spectra of monthly SST and SLP anomalies around the Southern Ocean at 50°S over 18 yr from 1982 to 1999 (Fig. 2a; Bendat and Piersol 1986, 361–424). These spectra achieve significant peaks for periods of ∼4 yr and zonal wavelengths of ∼180° longitude in the right-hand quadrant, with eastward phase propagation dominating. This is confirmed by forming the standing wave spectra (Fig. 2b) and the propagating wave spectra (Fig. 2c), the former calculated by taking the smallest spectral energy density in either right- or left-hand quadrant, and the latter calculated by taking their difference. Standing wave spectra for both SST and SLP anomalies display significant peaks at ∼4-yr period and ∼180° zonal wavelength, but with magnitudes ∼1/3 of those in the propagating wave spectra. This is consistent with the ratio of propagating-to-standing wave amplitude observed recently by White et al. (2002).

4. Associations between SST, QS+QE, QE, and PCP anomalies in the ACW

For coupling to occur between ocean and atmosphere in the ACW, warm SST anomalies must drive both upward latent heat flux anomalies and high precipitation anomalies, both required to drive positive diabatic heating (QD) anomalies in the overlying troposphere. To examine this issue, we begin by temporally cross correlating interannual SST anomalies with interannual QS+QE, QE, and PCP anomalies over the Southern Ocean during the 10-yr record (Fig. 3). The distribution of cross correlations between interannual SST and QS+QE anomalies (Fig. 3a), between interannual SST and QE anomalies (Fig. 3b), and between SST and PCP anomalies (Fig. 3c) finds both QS+QE, QE, and PCP anomalies positively correlated with SST anomalies over most of the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean from 35° to 50°S, and negatively correlated over the western Pacific and eastern Atlantic sectors of the Southern Ocean from 50° to 60°S.

Recently, White et al. (2002) computed the phase velocities of covarying SST and SLP anomalies in Fig. 1, finding the path of the ACW to be displaced north of the ACC in the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean, with phase velocities much less than ACC velocities. This indicated that SST anomalies in the ACW over this portion of the Southern Ocean are not advected by the ACC, as proposed initially by White and Peterson (1996) and Jacobs and Mitchell (1996). Instead, White et al. (2002) found the ACW path following that of maximum autumn–winter cyclogenesis density between 35° and 50°S (Simmonds and Keay 2000). Here we find the ACW path (solid thick gray band in Fig. 3) coinciding with positive cross correlations between SST and QS+QE, QE, and PCP anomalies. Moreover, it coincides with the path of the autumn–winter synoptic storm cyclogenesis density (thin gray band in Fig. 3) from Simmonds and Keay (2000). These associations are consistent with the hypothesis that SST anomalies in the ACW heat the overlying troposphere in this longitude domain through their influence on the synoptic storm aggregate. On the other hand, in the eastern Pacific and western Atlantic sectors of the Southern Ocean, the path of the ACW (dashed thick gray band in Fig. 3) is associated with negative correlations between SST and QS+QE, QE, and PCP anomalies. These associations are consistent with the hypothesis that the anomalous SST tendency is remotely forced from the Tropics (Cai and Baines 2001; White et al. 2002).

5. Zonal phase relationships between SST, QE, PCP, D200, and SLP anomalies in the ACW

To establish zonal phase relationships between SST, QE, PCP, 200-hPa divergence (D200), and SLP anomalies along the path of the ACW (thick gray band in Fig. 3), we construct time–distance diagrams along this path over the 10-yr record for two domains: 1) along the path across the eastern Atlantic, Indian, western and central Pacific sectors of the Southern Ocean from 0° eastward to 120°W (Fig. 4a); and 2) along the path across the eastern Pacific and western Atlantic sectors of the Southern Ocean straddling the Drake Passage (Fig. 4b). We also construct time–distance diagrams along these paths for 850-hPa divergence (D850) anomalies and for 850-hPa horizontal eddy heat flux divergence anomalies, but we do not display them for reasons of economy. The former propagate eastward directly out of phase with the D200 anomalies in both domains. The latter do not propagate eastward with the other variables in either domain, with magnitude smaller by half an order than that of latent heat flux anomalies in comparable units.

The time–distance diagrams over the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean (Fig. 4a, middle) show all five variables propagating slowly eastward together over the domain, with warm SST anomalies, upward QE anomalies, high PCP anomalies, and positive D200 anomalies approximately in phase and with high SLP anomalies displaced ∼90° of phase to the east of warm SST anomalies. The latter associates warm SST anomalies with poleward near-surface geostrophic wind anomalies. This is confirmed by computing zonal-lag cross correlations between SST and QE anomalies, SST and PCP anomalies, SST and D200 anomalies, and SST and SLP anomalies over the 10-yr record (Fig. 4a, bottom). Zonal-lag cross correlations between warm SST and upward QE anomalies are positively correlated at 0° phase lag, but largest upward QE anomalies are displaced to the west of warm SST anomalies by ∼45° of phase or so. This arises because QE anomalies are influenced by both ocean and atmosphere; that is, by SST anomalies to be sure but also by surface air temperature and surface wind anomalies (Cayan 1992). Zonal-lag cross correlation between D850 and SST anomalies (Fig. 4a, bottom) displays the opposite phase relationship between D200 and SST anomalies, indicating that upper-level divergence anomalies are accompanied by low-level convergence anomalies. From considerations of mass continuity, this indicates the presence of anomalous vertical ascent near midlevel in phase with warm SST and high PCP anomalies. This is confirmed below. These phase relationships are consistent with the deep diabatic heating scenario proposed by Palmer and Sun (1985).

Now we contrast these results (Fig. 4a) with those following the path of the ACW across the eastern Pacific and western Atlantic sectors of the Southern Ocean straddling the Drake Passage (Fig. 4b), finding the same variables propagating eastward but with weaker intensities and different zonal phase relationships. This occurs because this path is well south of the mean path of autumn–winter cyclogenesis density (Simmonds and Keay 2000) along which air–sea heat exchange and precipitation achieve maximum mean value (Peixoto and Oort 1992). We find high SLP anomalies displaced to the east of warm SST anomalies but by ∼45° of phase, indicating an encroachment of the warm-core high SLP anomaly onto the warm SST anomaly. We find significant negative cross correlation between SST, QE, and PCP anomalies, the opposite of that observed in Fig. 4a, suggesting that the influence of warm SST anomaly on QE anomaly is overwhelmed by that of the warm surface air temperature anomaly near the center of the warm-core high SLP anomaly. We also find cross correlations between SST and D200 and D850 anomalies (Fig. 4b, bottom) displaced 90° of phase from their alignment observed in Fig. 4a, consistent with the remote forcing scenario (e.g., Cai and Baines 2001; White et al. 2002).

6. Discussion of the deep diabatic heating scenario of Palmer and Sun

Palmer and Sun (1985) examined the anomalous linear vorticity budget of the extratropical troposphere on seasonal timescales; that is,
VζVζβυf0V
where (-) and ( )′ represent the annual time-mean basic state and the interannual anomaly about it, respectively; with ζ, υ, and V representing relative vorticity, meridional wind, and wind velocity, respectively; and with β representing the meridional derivative of the Coriolis parameter f. Other symbols are in standard notation. Hereafter the anomalous eddy vorticity flux divergence has been considered negligible and ζf0 throughout the troposphere, justifying this simplified form for the vorticity conservation equation compared to its more general form (e.g., Peixoto and Oort 1992, p. 46).
Palmer and Sun (1985) argued that at low level, with mean wind and relative vorticity weak, Eq. (6.1) should reduce to the Sverdrup balance (Sverdrup 1947), where the anomalous meridional advection of planetary vorticity balances anomalous wind divergence; that is,
βυf0V
At the upper level, in the vicinity of maximum zonal mean westerly wind, Palmer and Sun (1985) expected a balance between anomalous wind divergence and the zonal mean advection of anomalous relative vorticity [i.e., V·∇ζ′ in Eq. (6.1)] with the anomalous advection of planetary vorticity playing a significant but secondary role; that is,
Vζβυf0V
In the present study we find these two balances operating in the ACW by demonstrating statistically significant positive cross covariance between warm SST anomalies and anomalous low-level convergence, upper-level divergence, low-level poleward advection of planetary vorticity, and upper-level mean horizontal advection of anomalous relative vorticity, computed below.
Palmer and Sun (1985) also examined the anomalous linear temperature budget of the extratropical troposphere on seasonal timescales; that is,
i1520-0442-15-18-2577-e64
where T represents temperature and θ represents potential temperature, ω represents the pressure velocity in the p direction, Div〈VT″〉 represents eddy heat flux divergence in units of K s−1, QD/(ρCP) represents the diabatic heating in units of K s−1, CP represents the specific heat of air, and ρ represents the density of air decreasing upward. The anomalous diabatic heating includes the release of latent heat through anomalous precipitation and the net radiational heating through anomalous cloud fraction. In this expression the mean vertical advection of anomalous temperature have been neglected, yielding this simplified form for the temperature conservation equation (e.g., Peixoto and Oort 1992, p. 377). Palmer and Sun (1985) proposed that Eq. (6.4) should reduce to
i1520-0442-15-18-2577-e65
with the mean horizontal advection of anomalous temperature [i.e., V·∇T′ in Eq. (6.4)] tending to cancel the anomalous horizontal advection of mean temperature [i.e., V′·∇T in Eq. (6.4)], so that the anomalous net horizontal heat advection can be considered negligible compared to the anomalous vertical heat advection [i.e., ω′(T/θ)θ/∂p in Eq. (6.4)]. Palmer and Sun (1985) proposed that anomalous midlevel ascent (i.e., ω′ < 0) arises from the anomalous temperature balance in Eq. (6.5), where the uniform diabatic heating of the troposphere by underlying SST anomalies is accompanied by low-level cooling by anomalous eddy heat flux divergence. In the present study, we find a different anomalous temperature balance (see below). First, we cannot neglect the anomalous net horizontal heat advection in Eq. (6.4), and, second, we can neglect the anomalous eddy heat flux divergence.
The importance of the net horizontal heat advection in the temperature budget examined in the present study indicates that the potential vorticity budget comes into play, the latter computed from vorticity and temperature budgets [Eqs. (6.1) and (6.4)] by eliminating the anomalous pressure velocity (ω′) between them; that is,
i1520-0442-15-18-2577-e66
where the hydrostatic approximation (∂p/∂z = −ρg) has been assumed, the anomalous horizontal eddy heat flux divergence in Eq. (6.4) has been neglected as small, N2 represents the buoyancy frequency [(g/θ)θ/∂z] (Peixoto and Oort 1992, p. 49), and V′·∇f = βυ′. Normally the derivation of the potential vorticity equation assumes the thermal wind balance to hold throughout the column [i.e., ∂V′/∂z = [g/(Tf0)]k × ∇T′ and ∂V/∂z = [g/(Tf0)]k × ∇T; Peixoto and Oort (1992), p. 156] on the global space scales and the interannual timescales (Pedlosky 1987), in which case the third term in brackets on the lhs of Eq. (6.6) goes to zero, and ∇T′ in the first term and ∇T in the second term can be expressed in terms of ∂V′/∂z and ∂V/∂z, respectively. This would have yielded a simpler expression for Eq. (6.6), where V′ can be evaluated directly in terms of the vertical gradient of anomalous diabatic heating [i.e., ∂/∂z{QD′/(ρCP)[f0g/(N2T)]} in Eq. (6.6)]. However, we find the thermal wind approximation near the sea surface to be less accurate compared to its mid- to upper-level counterpart, requiring us to evaluate all the terms in Eq. (6.6) so that we can yield a residual estimate for ∂/∂z{QD′/(ρCP)[f0g/(N2T)]} that is consistent with QD′/(ρCP) in Eq. (6.4). So, Eq. (6.6) allows us to examine the relative importance of anomalous absolute vorticity advection [i.e., V·∇ζ′ + V′·(∇ζ + ∇f) in Eq. (6.6)] and anomalous net vortex tube advection [i.e., V·∂/∂z{[f0g/(N2T)]∇T′} + V′·∂/∂z{[f0g/(N2T)]∇T} in Eq. (6.6) in response to the vertical gradient of diabatic heating [i.e., ∂/∂z{QD′/(ρCP)[f0g/(N2T)]} in Eq. (6.6)].

In the shallow diabatic heating scenario of Hoskins and Karoly (1981), anomalous SST-induced low-level diabatic heating is balanced by the anomalous equatorward advection of mean temperature, yielding equatorward wind anomalies directly from Eq. (6.4) and anomalous midlevel descent indirectly from the Sverdrup balance in Eq. (6.2). The decrease in equatorward wind anomaly with height yields, through the thermal wind balance, a cold-core high SLP anomaly displaced ∼90° of phase to the west of a warm SST anomaly (Fig. 5a). In the deep diabatic heating scenario of Palmer and Sun (1985), anomalous SST-induced midlevel net heating [i.e., QD′/(ρCP) − Div〈VT″〉′ in Eq. (6.5)] is balanced by vertical heat advection [ω′(T/θ)θ/∂p in Eq. (6.5)], yielding anomalous midlevel ascent directly from the temperature balance in Eq. (6.5) and poleward wind anomalies indirectly from the Sverdrup balance in Eq. (6.2). The increase in poleward wind anomaly with height yields, through the thermal wind balance, a warm-core high SLP anomaly displaced ∼90° of phase to the east of a warm SST anomaly (Fig. 5b).

To this point in the analysis, we find no evidence supporting the shallow diabatic heating scenario in the ACW. On the contrary, all evidence points to the deep diabatic heating scenario, which can explain the displacement of warm-core high SLP anomalies ∼90° of phase to the east of the warm SST anomalies (Fig. 4a). Thus, throughout the remainder of this study we focus on developing the thermodynamics of this deep diabatic heating scenario, as the ACW propagates eastward across the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean.

7. Diagnosis of the anomalous temperature budget in the troposphere

To establish whether SST-induced latent heat flux and precipitation anomalies along the path of the ACW (Fig. 4a) lead to midlevel diabatic heating anomalies, we compute vertical sections of the zonal-lag regression coefficients between individual terms in Eq. (6.4) to SST variability directly underneath over the 10-yr record (Fig. 6). Usual regression coefficients are subsequently multiplied by the standard deviation of the SST anomalies to yield term magnitude and phase information associated with unit-K SST variability. We conduct this regression analysis for the apparent deep diabatic heating scenario operating in the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean (Fig. 6a), contrasting it with the remote forcing scenario operating in the western Pacific and eastern Atlantic sectors in the vicinity of the Drake Passage (Fig. 6b).

In the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean (Fig. 6a), zonal-lag regression between SST and the diabatic heating anomalies yield mid- to upper-level heating and low-level cooling ranging from −0.4 × 10−6 to 0.6 × 10−6 K s−1. As in Palmer and Sun (1985), the anomalous advection of the mean temperature and the mean advection of anomalous temperature tend to cancel each other but the former is larger than the latter throughout the column, yielding a warming tendency ranging from 0.2 × 10−6 to 0.6 × 10−6 K s−1, similar to that of the mid- to upper-level diabatic heating. This is not true of the zonal-lag regression between SST anomalies and the anomalous eddy heat flux divergence, the magnitude of which is negligible. Zonal-lag regression with anomalous vertical heat advection increases upward to a maximum value near 300 hPa, yielding a cooling tendency throughout the column. Thus, positive SST-induced latent heat flux instigates mid- to upper-level diabatic heating in the absence of significant eddy heat flux divergence; this diabatic heating is balanced by a combination of vertical heat advection and net horizontal heat advection, giving rise to anomalous midlevel ascent and poleward wind.

The positive mid- to upper-level diabatic heating anomaly is displaced slightly to the east of the warm SST anomaly and the upward latent heat flux anomaly, but occurs in phase with the positive precipitation anomaly. It is associated with the negative low-level diabatic heating anomaly of weaker value, the latter balanced by the warming tendency from the anomalous net horizontal heat advection. The source of this anomalous low-level cooling can only stem from consideration of anomalous radiational heating. Farther to the east, it is associated with the downward sensible-plus-latent heat flux anomaly, the latter representing an intrinsic feedback from the atmosphere to the ocean. Since the peak downward sensible-plus-latent heat flux anomaly is displaced 45°–90° of phase to the east of the warm SST anomalies, it yields an anomalous SST tendency that contributes both to eastward phase propagation and amplitude maintenance of the ACW (White et al. 1998).

The coincidence of latent heat flux, precipitation, and mid- to upper-level diabatic heating anomalies is a major finding of this study. The profile of anomalous diabatic heating, with low-level cooling and mid- to upper-level heating, must arise from some combination of latent heat release and radiational heating. Indeed, the vertical mass integral of the anomalous diabatic heating profile requires a net flux into the column of ∼0.5 W m−2, while the magnitude of anomalous latent heat flux at the sea surface is ∼1.2 W m−2 and the anomalous precipitation indicates a net flux of ∼1.6 W m−2 into the column. This may occur if low-level radiational cooling attends upper-level latent heat release when an abnormal number of high towers are generated during synoptic storm development in the presence of upward SST-induced latent heat flux anomaly. A detailed examination of the sources of anomalous diabatic heating remains for future study.

If anomalous vertical heat advection were the only term in Eq. (6.4) balancing the anomalous diabatic heating, then the profile of the anomalous pressure velocity (Fig. 7a) would resemble that of anomalous diabatic heating, with anomalous ascent at mid- to upper-level associated with anomalous descent at the low level (Fig. 6a). Yet here we observe anomalous ascent throughout the column (Fig. 7a). This occurs because the SST-induced diabatic heating anomaly is balanced by a combination of anomalous vertical heat advection (dominating at the upper level) and net horizontal heat advection (dominating at the low to midlevel), giving rise to anomalous ascent and poleward wind throughout the column.

This deep diabatic heating scenario is contrasted with the remote forcing scenario in the vicinity of the Drake Passage (Fig. 6b) where the zonal-lag regression between anomalous SST and anomalous vertical heat advection is near zero directly above the warm SST anomaly. Here we find anomalous advection of mean temperature and mean advection of anomalous temperature tending to cancel each other but with the former larger than the latter over most of the column. The residual of these three terms balances the anomalous diabatic heating, which is negative over most of the column, maximum at the low level. Here, the anomalous horizontal eddy heat flux divergence cannot be ignored. Thus, warm SST anomalies do not initiate midlevel ascent over this domain (Fig. 7b). Since the negative low-level diabatic heating anomaly is associated with the downward sensible-plus-latent heat flux anomaly, an anomalous SST warming tendency is indicated, responding to the anomalous poleward advection of warm moist air (Cayan 1992).

8. Diagnosis of the anomalous vorticity budget in the troposphere

Now we diagnose the vorticity budget to establish the balance between anomalous low-level divergence and upper-level convergence and the horizontal advection of the absolute vorticity in Eq. (6.1). We examine vertical sections of zonal-lag regression coefficients between individual terms in Eq. (6.1) to unit-K SST variability directly underneath (Fig. 8) in a manner consistent with Fig. 7. We conduct regression for the deep diabatic heating scenario in the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean (Fig. 8a) and for the remote forcing scenario in the vicinity of the Drake Passage (Fig. 8b).

In the former region (Fig. 8a), zonal-lag regression between SST anomalies and the anomalous divergence term can be seen displaying the vertical reversal at midlevel near 400 hPa, with warm SST anomalies associated with anomalous low-level convergence and upper-level divergence, consistent with zonal-lag cross correlations in Fig. 4a. At the low level, we find anomalous convergence balanced principally by the anomalous meridional advection of planetary vorticity as in Eq. (6.2), the latter dominating the mean advection of anomalous relative vorticity, with both mean wind and mean relative vorticity weak at the low level. However, at the upper level where the mean wind, mean relative vorticity, and anomalous winds are all much stronger, all the anomalous vorticity terms potentially play a role in balancing anomalous divergence. Here, we find the mean advection of the anomalous relative vorticity dominating the planetary vorticity advection, indicating that anomalous upper-level divergence is balanced principally by the former as in Eq. (6.3). The agreement between the residual divergence on the lhs of Eq. (6.1) and the measured divergence on the rhs of Eq. (6.1) (Fig. 8a) indicates that the vorticity budget is nominally closed without having to consider anomalous horizontal eddy vorticity flux divergence.

The quality of anomalous divergence can be problematic in evaluating the diabatic heating as a residual term in the anomalous temperature budget. However, at the midlatitude the abundant satellite- and ground-based observations, as well as the skillful AGCM simulation over this region (Kistler et al. 2001), assures the accuracy of the circulation, including divergence. Sardeshmukh (1993) suggested using rotational wind in the full vorticity equation to solve for divergence (chi problem), claiming that more consistent divergent winds can be derived. Yet, the striking similarity between residual and analyzed divergence anomalies (Fig. 8a) obtained in this analysis indicates that had we used the residual divergence anomaly to compute the residual diabatic heating anomaly in the temperature balance (Fig. 6a), it would differ little from our current result.

This deep diabatic heating scenario is contrasted with the remote forcing scenario in the vicinity of the Drake Passage (Fig. 8b), wherein warm SST anomalies can be seen located near the middle of anomalous warm-core high SLP anomalies, with equatorward and poleward winds displaced east and west, respectively. Thus, SST anomalies are nearly in quadrature with the midlevel ascent associated with anomalous low-level convergence and upper-level divergence (Fig. 7b), consistent with the hypothesis that SLP anomalies associated with the ACW in this domain are due to remote forcing from the Tropics (e.g., White et al. 2002).

9. Diagnosis of the anomalous potential vorticity budget in the troposphere

To examine the response of the troposphere circulation to the vertical gradient of the anomalous diabatic heating, we examine the anomalous potential vorticity budget [Eq. (6.6)]. This also allows us to compare the relative magnitudes of anomalous absolute vorticity and vortex tube advection, the latter deriving from the horizontal heat advection in the anomalous temperature budget. We establish the relative importance of the potential vorticity terms by computing their phase and magnitude (Fig. 9) in association with unit K SST variability as in Fig. 6 for temperature terms and Fig. 8 for vorticity terms.

At the low level (Fig. 9a), we find most of the potential vorticity terms on the lhs of Eq. (6.6) to be of similar magnitude, with maximum values ranging from 2 × 10−12 to 4 × 10−12 s−2, and with anomalous advection of mean relative vorticity negligible. The mean advection of anomalous vortex tubes and anomalous advection of mean vortex tubes are both negative, with magnitude larger than that of the positive anomalous ageostrophic vortex tube advection term by 1 × 10−12 to 2 × 10−12 s−2. Since the anomalous advection of planetary vorticity (βυ′) is negative and dominates the positive mean advection of anomalous relative vorticity by a similar amount, then the combination of planetary vorticity advection and vortex tube advection provides the principal balance with the vertical gradient of anomalous diabatic heating, the latter negative at the low level. At the upper level (Fig. 9a), with the anomalous vortex tube advection terms on the lhs of Eq. (6.6) nearly canceling each other, the mean advection of anomalous relative vorticity is positive and dominates the negative anomalous advection of planetary vorticity by 1 × 10−12 to 2 × 10−12 s−2. Thus, mean advection of anomalous relative vorticity provides the principal balance with the vertical gradient of the anomalous diabatic heating, the latter positive at the upper level.

At the midlevel (Fig. 9a), the potential vorticity terms on the lhs of Eq. (6.6) are comparable. Here, the anomalous net vortex tube advection ranges from 2 × 10−12 to 3 × 10−12 s−2, and the mean advection of anomalous relative vorticity and the anomalous advection of planetary vorticity are comparable to this. Thus, the potential vorticity balance has the mean advection of the anomalous relative vorticity, the anomalous advection of planetary vorticity, and the anomalous net vortex tube advection providing the balance with the vertical gradient of anomalous diabatic heating, the latter negative at the midlevel.

We know that the various terms in the potential vorticity budget are computed correctly, since their residual on the rhs of Eq. (6.6) (Fig. 9a) yields an estimate for the vertical gradient of anomalous diabatic heating that is consistent with the residual estimate for anomalous diabatic heating on the rhs of Eq. (6.4) (Fig. 6a). Moreover, this can occur only if internal consistency is achieved among the three conservation equations [i.e., Eqs. (6.1), (6.4), and (6.6)]. We find the net horizontal heat advection anomaly in the diagnosis of Eq. (6.4) contributing as much to the anomalous ascent (Fig. 7) as the anomalous diabatic heating anomaly. This carries over into the anomalous potential vorticity budget in Eq. (6.6), producing a net vortex tube advection that alters the absolute vorticity balance over most of the column.

10. Discussion and conclusions

We display animation sequences of SLP and SST anomalies in the ACW from 30° to 70°S for the 10 yr from 1983 to 1992. Both variables can be seen propagating slowly eastward around the Southern Ocean in fixed phase with one another. The global distribution of phase velocities of the ACW (White et al. 2002) finds its mean path taking it as far north as 35°S in the Indian sector of the Southern Ocean and as far south as 60°S in the eastern Pacific and western Atlantic sectors straddling the Drake Passage. During this 10-yr portion of the available record, the ACW was particularly robust and the phase relationships among SST and SLP anomalies were relatively stable. Zonal wavenumber–frequency spectra of monthly SST and SLP anomalies along 50°S yields two significant findings; first, the standing wave component of interannual variability has an amplitude that is ∼1/3 that of the ACW, consistent with the complex empirical orthogonal function (CEOF) analysis of both variables conducted by White et al. (2002); and second, both variables display clear significant spectral peaks for interannual periods of 3–6 years and for zonal wavelengths of ∼180° of longitude, demonstrating that both the ocean and the atmosphere participate in the ACW phenomenon.

Since the ACW propagates slowly eastward around the Southern Ocean in covarying SST and SLP anomalies, three different explanations have been proposed to explain it: 1) the extratropical ocean and atmosphere couple to one another to produce the ACW; 2) the SLP anomalies propagate eastward for different reasons and the SST anomalies are forced to propagate along with them; and 3) the SST anomalies are generated by stochastic variability in the atmosphere in damped resonance with advection by the ACC. In the latter explanation, simulated in coupled ocean–atmosphere general circulation models (e.g., Christoph et al. 1998; Motoi et al. 1998; Cai et al. 1999), the SLP anomalies do not propagate with the SST anomalies. Thus, they simulate some model phenomenon different from that observed in the ACW. On the other hand, we obtain results that are consistent with the first two explanations, each dominated in different regions of the Southern Ocean, with the first explanation appearing to hold in the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean and the second explanation appearing to hold in the western Pacific and eastern Atlantic sectors.

In the eastern Pacific and western Atlantic sectors of the Southern Ocean, Cai and Baines (2001) proposed that the ACW is forced by the PSA pattern of SLP anomaly straddling the Drake Passage, established there by meridional atmospheric teleconnections associated with the standing mode of ENSO (Karoly 1989). On the other hand, White et al. (2002) proposed that the ACW is forced remotely by the meridional atmospheric teleconnections associated with the propagation of the GEW across the warm pool in the tropical western Pacific Ocean. As important, both Cai and Baines (2001) and White et al. (2002) demonstrated that remote forcing from the Tropics is restricted to the eastern Pacific and western Atlantic sectors of the Southern Ocean. This occurs for two reasons: first, tropical deep convection associated with ENSO is restricted to that portion of the warm pool extending across the western tropical Pacific Ocean from 120°E to 180°; and, second, the generation of quasi-stationary Rossby waves in the upper-level westerly winds is restricted to the vicinity of the east coast of Australia where mean winds are strongest, as predicted in model simulations by Sardeshmukh and Hoskins (1991). Thus, Cai and Baines (2001) and White et al. (2002) could offer explanations for the eastward phase propagation of the ACW in the vicinity of the Drake Passage in terms of remote forcing from the Tropics, but they could not explain it over the remainder of the Southern Ocean.

In this latter region, White et al. (2002) demonstrated that the path of the ACW across the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean does not coincide with that of the ACC, as proposed by White and Peterson (1996) and Jacobs and Mitchell (1996). They found the ACW displaced north of the ACC and traveling much slower, following the path of maximum autumn–winter cyclogenesis density between 35° and 50°S (Simmonds and Keay 2000). This suggested that ocean–atmosphere coupling in the ACW involves the influence of SST anomalies on extratropical cyclone activity in the synoptic storm aggregate, as suggested by Peng and Whitaker (1999). Subsequent analysis by White and Simmonds (2002, manuscript submitted to J. Climate, hereafter WS) supports this contention, with SLP anomalies in the ACW found to be associated with extratropical cyclone density and intensity anomalies.

So we began the present study by showing that the full tropospheric column propagates slowly eastward with covarying SST and SLP anomalies in the ACW across the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean. Along this path, extending across 240° of longitude, we found warm SST anomalies associated with anomalous upward latent heat flux, high precipitation, low-level convergence, and upper-level divergence, with midlevel ascent inferred by continuity. These associations are significant at the 90% confidence level and are consistent with the tropospheric response to SST anomalies in the deep diabatic heating scenario of Palmer and Sun (1985). This suggested that anomalous midlevel ascent occurs in response to SST-induced midlevel diabatic heating anomalies.

To examine these tentative conclusions, we subsequently diagnosed the anomalous temperature, vorticity, and potential vorticity budgets along the path of the ACW, focusing on the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean. In the temperature budget, we found warm SST anomalies generating anomalous mid- to upper-level diabatic heating and low-level diabatic cooling, different from that expected from the diabatic heating scenario of Palmer and Sun (1985). This anomalous diabatic heating profile is balanced by a combination of anomalous positive vertical heat advection and net horizontal heat advection, yielding anomalous ascent and poleward wind over the column. In the vorticity budget, we found the anomalous ascent, maximum at mid- to upper-level, instigating anomalous weak low-level convergence and strong upper-level divergence. The low-level convergence is balanced by the anomalous advection of planetary vorticity and vortex tubes, yielding a poleward wind anomaly of ∼0.1 m s−1 that is in geostrophic balance with the SLP anomaly in the ACW (White and Peterson 1996). On the other hand, the more intense upper-level divergence is balanced principally by the mean advection of anomalous relative vorticity, also yielding a poleward wind anomaly. Thus, anomalous low-level convergence and upper-level divergence in response to peak anomalous mid- to upper-level ascent produce an equivalently barotropic poleward wind response to warm SST anomalies. Upon diagnosing the potential vorticity budget, we found the magnitude of the anomalous net vortex tube advection over most of the troposphere comparing with that of the anomalous advection of planetary vorticity and the mean advection of anomalous relative vorticity, all three terms required to balance the vertical gradient of anomalous diabatic heating.

This deep diabatic heating scenario for the ACW can be summarized in a cartoon (Fig. 10). In the anomalous temperature budget, the anomalous mid- to upper-level diabatic heating near 300 hPa and the anomalous low-level diabatic cooling near 900 hPa occur in response to SST anomalies. This profile of anomalous diabatic heating and cooling is balanced by the anomalous vertical and horizontal heat advection, yielding anomalous ascent and poleward wind throughout the column. This poleward wind anomaly has sufficient vertical shear so that the corresponding net horizontal vortex tube advection in the anomalous potential vorticity balance becomes significant, complicating the relatively simple absolute vorticity balance given by Palmer and Sun (1985). At the low-level where QD < 0, anomalous ascent arises principally from the anomalous poleward advection of mean temperature. At the mid- to upper level where QD > 0, anomalous ascent arises principally from the anomalous vertical advection of mean potential temperature.

A major finding of this study is the coincidence between SST-induced latent heat flux, precipitation, and mid- to upper-level diabatic heating anomalies in the ACW. Another coincidence has been found between anomalous SLP and extratropical cyclone density and intensity in the ACW (WS). Taken together, these coincidences suggest that warm SST anomalies enhance extratropical cyclone development by supplying more latent heat to the synoptic storm aggregate. The resulting profile of anomalous diabatic heating, with low-level cooling and mid- to upper-level heating, obviously arises from some combination of latent heat release and radiational heating, since sensible heating by the transient eddies is insignificant. Indeed, the vertical mass integral of the anomalous diabatic heating profile requires a net flux into the column of ∼0.5 W m−2, while the magnitude of anomalous latent heat flux at the sea surface is ∼1.2 W m−2 and the anomalous precipitation indicates a net flux of ∼1.6 W m−2 flux into the column. This suggests that radiational cooling must contribute significantly to the anomalous diabatic heating profile. For example, low-level radiational cooling may attend the upper-level latent heat release if an abnormal number of high towers are generated in the synoptic storm response to warm SST anomalies. Another major finding is that the anomalous eddy heat flux divergence is negligible in the temperature budget, indicating that SST anomalies do not significantly alter the baroclinic instability process responsible for cyclone development in the synoptic storm aggregate. A detailed examination of the sources of anomalous diabatic heating remains to be done.

Thus, the deep diabatic heating scenario operating in the ACW differs from the deep diabatic heating scenario by Palmer and Sun (1985) in three principle ways. First, Palmer and Sun (1985) found SST-induced midlevel diabatic heating anomalies arising from anomalous cooling by the low-level eddy heat flux divergence. We find the latter to be negligible in the ACW. Second, we find SST-induced mid- to upper-level diabatic heating and low-level diabatic cooling balanced by vertical and horizontal heat advection of similar magnitude, yielding anomalous ascent and poleward wind over the entire column. Palmer and Sun (1985) did not find poleward wind generated in the temperature budget; rather, it was driven by anomalous divergence in the vorticity budget. Third, the anomalous net horizontal heat advection in the temperature balance produces a significant anomalous net horizontal vortex tube advection in the potential vorticity balance, contributing to the absolute vorticity balance proposed by Palmer and Sun (1985).

The magnitude of the anomalous SST-induced latent heat flux of ∼1 W m−2 in the deep diabatic heating scenario is small when compared with ∼100 W m−2 expected in a synoptic storm, but in the integral sense it produces a larger transfer of energy. To see this, consider that the ACW takes ∼2 years to transition from its warm phase to its cool phase at any one location, with the change in SST over the upper 10 m of ocean at least (corresponding to the change in air temperature over the entire column of the troposphere) experiencing a change of ∼1.5°C. Moreover, this anomalous change occurs not only at a point; it occurs over ∼45° of longitude and ∼20° of latitude. Thus, the total amount of energy being exchanged between the ocean and the atmosphere is O(1027 J), as compared with O(1026 J) exchanged during a synoptic storm in the Southern Hemisphere, with an anomalous air–sea heat flux of ∼100 W m−2 extending over 1/10 this domain for ∼2 days.

The present analysis reveals an intrinsic feedback from the atmosphere to the ocean as proposed initially by Wallace and Jiang (1987), and subsequently exploited by a number of analytical coupled models of the ACW (Qiu and Jin 1997; White et al. 1998; Baines and Cai 2000). In the diagnosis of the anomalous temperature budget, we find the warm SST anomalies associated with low-level diabatic heating anomalies, the latter associated with downward sensible-plus-latent anomalies displaced 45° to 90° of phase to the east. The latter can be expected to drive an anomalous SST tendency in the upper ocean, which can account for both the eastward phase propagation and amplitude maintenance of the ACW (White et al. 1998).

Throughout this study we have contrasted the deep diabatic heating scenario of the ACW in the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean with the remote forcing scenario of the ACW in the eastern Pacific and western Atlantic sectors. In this latter region we find significant negative associations between SST anomalies and sensible-plus-latent heat flux and precipitation anomalies, nearly the opposite of those found over the former region. Diagnosing the temperature budget finds warm SST anomalies associated with anomalous diabatic cooling throughout the troposphere, maximum at the low level where it is associated with downward sensible-plus-latent heat flux anomalies, the latter required to contribute to an anomalous SST warming tendency. Diagnosing the vorticity budget finds the warm SST anomaly nearly underlying the warm-core high SLP anomaly, with the meridional wind anomalies on either side equivalently barotropic. These diagnostics are consistent with the remote forcing scenarios proposed by Cai and Baines (2001) and White et al. (2002).

Acknowledgments

This research was supported by the Office of Global Programs of NOAA (NOAA Grant NA 77RJ0453) in concert with the Experimental Climate Prediction Center at SIO. Warren White is also supported by the National Aeronautics and Space Administration (NASA) under Contract JPL 1205106 and by the National Science Foundation (Grant OCE-9920730). Discussions with D. Cayan and M. Flatau were most helpful and deeply appreciated. Our thanks extend to Ted Walker who provided the computational support and to Andrea Fincham who developed the final figures.

REFERENCES

  • Andersen, N., 1974: On the calculation of filter coefficients for maximum entropy spectral analysis. Geophysics, 39 , 6972.

  • Baines, P. G., and W. Cai, 2000: Analysis of an interactive instability mechanism for the Antarctic Circumpolar Wave. J. Climate, 13 , 18311844.

    • Search Google Scholar
    • Export Citation
  • Bendat, J. S., and A. G. Piersol, 1986: Random Data: Analysis and Measurement Procedures. John Wiley and Sons, 566 pp.

  • Cai, W., and P. G. Baines, 2001: Forcing the Antarctic Circumpolar Wave by ENSO teleconnections. J. Geophys. Res., 106 , 90199038.

  • Cai, W., and H. B. Gordon, 1999: Southern mid- to high-latitude variability, a zonal wavenumber-3 pattern, and the Antarctic Circumpolar Wave in the CSIRO coupled model. J. Climate, 12 , 30873104.

    • Search Google Scholar
    • Export Citation
  • Cayan, D. R., 1992: Latent and sensible heat flux anomalies over the northern oceans: Driving the sea surface temperature anomalies. J. Phys. Oceanogr., 22 , 859881.

    • Search Google Scholar
    • Export Citation
  • Christoph, M., T. P. Barnett, and E. Roechner, 1998: The Antarctic Circumpolar Wave in a coupled ocean–atmosphere GCM. J. Climate, 11 , 16591672.

    • Search Google Scholar
    • Export Citation
  • Davis, R. E., 1976: Predictability of sea surface temperature and sea level pressure anomalies over the North Pacific Ocean. J. Phys. Oceanogr., 6 , 249266.

    • Search Google Scholar
    • Export Citation
  • Horel, J. D., and J. M. Wallace, 1981: Planetary-scale atmospheric phenomena associated with the Southern Oscillation. Mon. Wea. Rev., 109 , 813829.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and D. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38 , 11791196.

    • Search Google Scholar
    • Export Citation
  • Jacobs, G. A., and J. L. Mitchell, 1996: Ocean circulation variations associated with the Antarctic Circumpolar Wave. Geophys. Res. Lett., 23 , 29472950.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors. 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Karoly, D. J., 1989: Southern Hemisphere circulation features associated with El Niño–Southern Oscillation events. J. Climate, 2 , 12391252.

    • Search Google Scholar
    • Export Citation
  • Kaylor, R. E., 1977: Filtering and decimation of digital time series. Tech. Rep. Note BN 850, Institute of Physics, Science, and Technology, University of Maryland, College Park, 14 pp.

    • Search Google Scholar
    • Export Citation
  • Kistler, R., and Coauthors. 2001: The NCEP–NCAR 50-Year Reanalysis: Monthly mean CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247267.

    • Search Google Scholar
    • Export Citation
  • Kushnir, Y., and I. M. Held, 1996: Equilibrium atmospheric response to North Atlantic SST anomalies. J. Climate, 9 , 12081220.

  • Lau, N. C., and M. J. Nath, 1990: A general circulation model of the atmospheric response to extratropical SST anomalies observed in 1950–1970. J. Climate, 3 , 965989.

    • Search Google Scholar
    • Export Citation
  • Motoi, T., A. Kito, and H. Koide, 1998: The Antarctic circumpolar wave in a coupled ocean–atmosphere model. Ann. Glaciol., 27 , 483487.

    • Search Google Scholar
    • Export Citation
  • Namias, J., 1972: Space scales of sea surface temperature patterns and their causes. Fish. Bull., 70 , 611617.

  • Palmer, T. N., and Z. Sun, 1985: A modeling and observational study of the relationship between sea surface temperature in the northwest Atlantic and the atmospheric general circulation. Quart. J. Roy. Meteor. Soc., 111 , 947975.

    • Search Google Scholar
    • Export Citation
  • Pedlosky, J., 1987: Geophysical Fluid Dynamics. Springer-Verlag, 710 pp.

  • Peixoto, J. P., and A. H. Oort, 1992: Physics of Climate. American Institute of Physics, 520 pp.

  • Peng, S., and J. S. Whitaker, 1999: Mechanisms determining the atmospheric response to midlatitude SST anomalies. J. Climate, 12 , 13931408.

    • Search Google Scholar
    • Export Citation
  • Peng, S., L. A. Mysak, H. Ritchie, J. Derome, and B. Dugas, 1995: The differences between early and midwinter atmospheric responses to sea surface temperature anomalies in the northwest Atlantic. J. Climate, 8 , 137157.

    • Search Google Scholar
    • Export Citation
  • Peng, S., W. A. Robinson, and M. P. Hoerling, 1997: The modeled atmospheric response to midlatitude SST anomalies and its dependence on background circulation states. J. Climate, 10 , 971987.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., and F-F. Jin, 1997: Antarctic circumpolar wave: An indication of ocean–atmosphere coupling in the extratropics. Geophys. Res. Lett., 24 , 25852588.

    • Search Google Scholar
    • Export Citation
  • Ratcliffe, R. A. S., and R. Murray, 1970: New lag associations between North Atlantic sea temperatures and European pressure, applied to long-range weather forecasting. Quart. J. Roy. Meteor. Soc., 96 , 226246.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., and D. C. Marsico, 1993: An improved real-time global sea surface temperature analysis. J. Climate, 6 , 114119.

  • Sardeshmukh, P. D., 1993: The baroclinic chi problem and its application to the diagnosis of atmospheric heating rates. J. Atmos. Sci., 50 , 10991112.

    • Search Google Scholar
    • Export Citation
  • Sardeshmukh, P. D., and B. J. Hoskins, 1991: The generation of global rotational flow by idealized tropical divergence. J. Atmos. Sci., 45 , 12281251.

    • Search Google Scholar
    • Export Citation
  • Simmonds, I., and K. Keay, 2000: Mean Southern Hemisphere extratropical cyclone behavior on the 40-year NCEP–NCAR reanalysis. J. Climate, 13 , 873885.

    • Search Google Scholar
    • Export Citation
  • Slutz, R. J., S. J. Lubker, J. D. Hiscox, S. D. Woodruff, R. L. Jenne, D. H. Joseph, P. M. Steurer, and J. D. Elms, 1985: Comprehensive ocean–atmosphere data set; release 1. NOAA Environmental Research Laboratory, Boulder, CO, 268 pp.

    • Search Google Scholar
    • Export Citation
  • Snedecor, G. W., and W. G. Cochran, 1980: Statistical Methods. Iowa State University Press, 507 pp.

  • Sverdrup, H. U., 1947: Wind-driven currents in a baroclinic ocean; with application to the equatorial currents of the eastern Pacific. Proc. Natl. Acad. Sci. USA, 33 , 318326.

    • Search Google Scholar
    • Export Citation
  • Talley, L. D., 1999: Simple coupled midlatitude climate models. J. Phys. Oceanogr., 29 , 20162037.

  • Wallace, J. M., and D. Gutzler, 1981: Teleconnections in the geopotential height field during Northern Hemisphere winter. Mon. Wea. Rev., 109 , 784812.

    • Search Google Scholar
    • Export Citation
  • Wallace, J. M., and Q-R. Jiang, 1987: On the observed structure of the interannual variability of the observed ocean/atmosphere climate system. Atmospheric and Oceanic Variability, H. Cattle, Ed., Roy. Meteor. Soc., 17–43.

    • Search Google Scholar
    • Export Citation
  • White, W. B., 1995: Design of a global observing system for basin-scale upper ocean temperature anomalies. Progress in Oceanography, Vol. 36, Pergamon, 169–217.

    • Search Google Scholar
    • Export Citation
  • White, W. B., . 2000: Tropical coupled Rossby waves in the Pacific ocean–atmosphere system. J. Phys. Oceanogr., 30 , 12451264.

  • White, W. B., and R. Peterson, 1996: An Antarctic circumpolar wave in surface pressure, wind, temperature, and sea ice extent. Nature, 380 , 699702.

    • Search Google Scholar
    • Export Citation
  • White, W. B., and D. R. Cayan, 2000: A global ENSO wave in surface temperature and pressure and its interdecadal modulation from 1900 to 1996. J. Geophys. Res., 105 , 1122311242.

    • Search Google Scholar
    • Export Citation
  • White, W. B., S-C. Chen, and R. Peterson, 1998: The Antarctic Circumpolar Wave: A beta-effect in ocean–atmosphere coupling over the Southern Ocean. J. Phys. Oceanogr., 28 , 23452361.

    • Search Google Scholar
    • Export Citation
  • White, W. B., and R. J. Allan, 2002: Positive feedbacks between the Antarctic circumpolar wave and the El Niño–Southern Oscillation wave. J. Geophys. Res., in press.

    • Search Google Scholar
    • Export Citation
  • Xie, P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78 , 25392558.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Animation sequences of maps of interannual SST and SLP anomalies over the Southern Ocean for 10 yr from Jan 1983 to Jan 1993, with maps displayed every 6 months. Each map in the sequence extends meridionally from 30° to 70°S and around the globe from 30° to 30°E. Positive (negative) anomalies range from yellow to red (blue to purple) with contours of 0.1 K and 0.25 hPa

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 2.
Fig. 2.

(a) Zonal wavenumber–frequency spectra of SST and SLP anomalies along 50°S, extending around the globe, computed for 18 yr from 1982 to 1999. (b) Standing wave spectra computed by taking the smallest spectral energy estimate in the left- and right-hand quadrants in each zonal wavenumber–frequency band. (c) Propagation wave spectra computed by taking the difference between spectral energy estimates in the left- and right-hand quadrants in each zonal wavenumber–frequency band. Contours give the 90% confidence levels (Snedecor and Cochran 1980). Shading is for effect

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 3.
Fig. 3.

(a) Distribution of the temporal cross correlation between interannual SST and QS+QE anomalies at each grid point for the 10 yr from Jan 1983 to Jan 1992 over the Southern Ocean from 20° to 60°S. (b) Distribution of the temporal correlation between interannual SST and QE anomalies at each grid point for the 10 yr from Jan 1983 to Jan 1992 over the Southern Ocean from 20° to 60°S. (c) Distribution of the temporal correlation between interannual SST and PCP anomalies at each grid point for the same period. Positive (negative) correlations are unhatched (hatched). Correlations greater than 0.4 are significant at the 90% confidence level for ∼15 effective temporal degrees of freedom (Snedecor and Cochran 1980) computed using a temporal decorrelation scale of 8 months. The thick gray band follows the path of the ACW given in White et al. (2002). In the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean, this path overlies positive correlations between SST anomalies and QS+QE, QE, and PCP anomalies. The thin gray band follows the path of maximum cyclogenesis density for winter (Jun–Jul–Aug; Simmonds and Keay 2000).>

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 4
Fig. 4

a. (a) The path of the ACW across the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean from 0° eastward to 120°W, taken from White et al. (2002). (b) Time–distance diagrams of interannual SST, QE, PCP, D200, and SLP anomalies for the 10 yr from 1984 to 1993 along the solid gray path in (a). Positive (negative) anomalies are unhatched (hatched). (c) Zonal-lag cross correlations between SST anomalies and QE, PCP, D200, D850, and SLP anomalies in (b). The 90% confidence level is 0.4 for ∼15 effective temporal degrees of freedom (Snedecor and Cochran 1980). Units and contour intervals are given in the heading of each time–distance diagram

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 4
Fig. 4

b. Same as in Fig. 4a but for the solid gray path of the ACW across the eastern Pacific and western Atlantic sectors of the Southern Ocean in the vicinity of the Drake Passage from 120°W eastward to 0°.

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 5.
Fig. 5.

Schematic diagram over an extratropical east–west cross section in the Southern Ocean for (a) the shallow diabatic heating scenario proposed by Hoskins and Karoly (1981) and (b) the deep diabatic heating scenario proposed by Palmer and Sun (1985). In the shallow diabatic heating scenario, the anomalous circulation shows a warm-core low (WL) displaced to the east and a cold-core high (CH) displaced to the west of low-level diabatic heating (QD) anomaly, with equatorward wind anomalies decreasing with height directly overhead and vertical descent occurring at the midlevel. In the deep diabatic heating scenario, the circulation shows a warm-core high (WH) displaced to the east and a cold-core low (CL) displaced to the west of the midlevel net heating from the sum of diabatic heating anomaly (QD) and horizontal eddy heat flux divergence anomaly (ED), with poleward wind anomalies increasing with height and vertical ascent occurring at the midlevel. The time–mean wind (u) in both scenarios are westerly, increasing with height

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 6.
Fig. 6.

(a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and components of the anomalous temperature budget [Eq. (6.4)]; i.e., V·∇T′, V′·∇T, ω′(T/θ)θ/∂p, Div〈VT″〉′, and their sum, the latter giving the residual computation of QD′/(ρCP) in Eq. (6.4). These regression coefficients extend over the height of the troposphere from 900 to 200 hPa, with zonal-lag regression computed along the path following the time–distance diagrams in Fig. 4a from 1984 to 1993. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Also displayed are zonal-lag regression coefficients between interannual SST anomalies and interannual QS+QE, QE, and PCP anomalies. Usual regression coefficients are multiplied by the rms of the SST anomalies so that these regression coefficients yield both magnitude and phase information in units of 10−6 K s−1 associated with unit-K SST variability. Similar regression coefficients for PCP, QS+QE, and QE anomalies are in units of mm month−1, W m−2, and W m−2, respectively, in association with unit-K SST variability

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 7.
Fig. 7.

(a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and the vertical pressure velocity anomaly (−ω′, positive upward) in Eq. (6.4), extending over the height of the troposphere from 900 to 200 hPa along a path following the time–distance diagrams in Fig. 4a from 1984 to 1993. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Also displayed are zonal-lag regression coefficients between interannual SST anomalies and interannual QS+QE, QE, and PCP anomalies. Usual regression coefficients are multiplied by the rms of the SST anomalies so that ω′ has units of 10−4 hPa s−1 in association with unit-K SST variability. Similar regression coefficients for PCP, QS+QE, and QE anomalies are in units of mm month−1, W m−2, and W m−2, respectively, in association with unit-K SST variability

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 8.
Fig. 8.

(a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and components of the anomalous vorticity budget [Eq. (6.1)]; i.e., V·ζ′, V′·ζ, βυ′, and their sum, the latter expected to yield a residual −f0∇·V′, which is similar to that computed directly. These regression coefficients extend over the height of the troposphere from 900 to 200 hPa, with zonal-lag regression computed along the path following the time–distance diagrams in Fig. 4a from 1983 to 1992. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Also displayed are zonal-lag regression coefficients between interannual SST anomalies and interannual QS+QE, QE, and PCP anomalies. Usual regression coefficients are multiplied by the rms of the SST anomalies so that these regression coefficients yield both magnitude and phase in units of 10−12 s−2 associated with unit-K SST variability. Similar regression coefficients for PCP, QS+QE, and QE anomalies are in units of mm month−1, W m−2, and W m−2, respectively, in association with unit-K SST variability

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 9.
Fig. 9.

(a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and components of the anomalous potential vorticity budget [Eq. (6.6); i.e., V·ζ′, V·∂/∂z{[f0g/(N2T)]T′}, βυ′, V′·ζ, V′·∂/∂z{[f0g/(N2T)]∇T}, [f0g/(N2T)](∂V/∂z·T′ + ∂V′/∂z·T), and their residual (i.e., ∂/∂z{QD′/(ρCP)[f0g/(N2T)]}). These regression coefficients extend over the height of the troposphere from 900 to 300 hPa along the path following the time–distance diagrams in Fig. 4a from 1983 to 1992. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Usual regression coefficients are multiplied by the rms of the SST anomalies so that displayed regression coefficients yield both magnitude and phase in units of 10−12 s−2 associated with unit-K SST variability

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Fig. 10.
Fig. 10.

Schematic diagram along a zonal cross section of the troposphere in the Southern Ocean summarizing the deep diabatic heating scenario observed in the ACW. Positive SST-induced latent heat flux anomalies are associated with anomalous mid- to upper-level diabatic heating (QD > 0) and anomalous low-level diabatic cooling (QD < 0). This profile of anomalous diabatic heating is balanced by a combination of anomalous vertical and horizontal heat advection, yielding anomalous ascent and poleward wind over the column. The anomalous ascent (ω′) achieves maximum value between 500 and 300 hPa, producing weak low-level convergence and strong upper-level divergence anomalies. The low-level convergence is balanced principally by the planetary vorticity advection and the upper-level divergence is balanced principally by relative vorticity advection, yielding a poleward equivalently barotropic wind response to SST anomalies

Citation: Journal of Climate 15, 18; 10.1175/1520-0442(2002)015<2577:TMRFTT>2.0.CO;2

Save
  • Andersen, N., 1974: On the calculation of filter coefficients for maximum entropy spectral analysis. Geophysics, 39 , 6972.

  • Baines, P. G., and W. Cai, 2000: Analysis of an interactive instability mechanism for the Antarctic Circumpolar Wave. J. Climate, 13 , 18311844.

    • Search Google Scholar
    • Export Citation
  • Bendat, J. S., and A. G. Piersol, 1986: Random Data: Analysis and Measurement Procedures. John Wiley and Sons, 566 pp.

  • Cai, W., and P. G. Baines, 2001: Forcing the Antarctic Circumpolar Wave by ENSO teleconnections. J. Geophys. Res., 106 , 90199038.

  • Cai, W., and H. B. Gordon, 1999: Southern mid- to high-latitude variability, a zonal wavenumber-3 pattern, and the Antarctic Circumpolar Wave in the CSIRO coupled model. J. Climate, 12 , 30873104.

    • Search Google Scholar
    • Export Citation
  • Cayan, D. R., 1992: Latent and sensible heat flux anomalies over the northern oceans: Driving the sea surface temperature anomalies. J. Phys. Oceanogr., 22 , 859881.

    • Search Google Scholar
    • Export Citation
  • Christoph, M., T. P. Barnett, and E. Roechner, 1998: The Antarctic Circumpolar Wave in a coupled ocean–atmosphere GCM. J. Climate, 11 , 16591672.

    • Search Google Scholar
    • Export Citation
  • Davis, R. E., 1976: Predictability of sea surface temperature and sea level pressure anomalies over the North Pacific Ocean. J. Phys. Oceanogr., 6 , 249266.

    • Search Google Scholar
    • Export Citation
  • Horel, J. D., and J. M. Wallace, 1981: Planetary-scale atmospheric phenomena associated with the Southern Oscillation. Mon. Wea. Rev., 109 , 813829.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and D. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38 , 11791196.

    • Search Google Scholar
    • Export Citation
  • Jacobs, G. A., and J. L. Mitchell, 1996: Ocean circulation variations associated with the Antarctic Circumpolar Wave. Geophys. Res. Lett., 23 , 29472950.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors. 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Karoly, D. J., 1989: Southern Hemisphere circulation features associated with El Niño–Southern Oscillation events. J. Climate, 2 , 12391252.

    • Search Google Scholar
    • Export Citation
  • Kaylor, R. E., 1977: Filtering and decimation of digital time series. Tech. Rep. Note BN 850, Institute of Physics, Science, and Technology, University of Maryland, College Park, 14 pp.

    • Search Google Scholar
    • Export Citation
  • Kistler, R., and Coauthors. 2001: The NCEP–NCAR 50-Year Reanalysis: Monthly mean CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247267.

    • Search Google Scholar
    • Export Citation
  • Kushnir, Y., and I. M. Held, 1996: Equilibrium atmospheric response to North Atlantic SST anomalies. J. Climate, 9 , 12081220.

  • Lau, N. C., and M. J. Nath, 1990: A general circulation model of the atmospheric response to extratropical SST anomalies observed in 1950–1970. J. Climate, 3 , 965989.

    • Search Google Scholar
    • Export Citation
  • Motoi, T., A. Kito, and H. Koide, 1998: The Antarctic circumpolar wave in a coupled ocean–atmosphere model. Ann. Glaciol., 27 , 483487.

    • Search Google Scholar
    • Export Citation
  • Namias, J., 1972: Space scales of sea surface temperature patterns and their causes. Fish. Bull., 70 , 611617.

  • Palmer, T. N., and Z. Sun, 1985: A modeling and observational study of the relationship between sea surface temperature in the northwest Atlantic and the atmospheric general circulation. Quart. J. Roy. Meteor. Soc., 111 , 947975.

    • Search Google Scholar
    • Export Citation
  • Pedlosky, J., 1987: Geophysical Fluid Dynamics. Springer-Verlag, 710 pp.

  • Peixoto, J. P., and A. H. Oort, 1992: Physics of Climate. American Institute of Physics, 520 pp.

  • Peng, S., and J. S. Whitaker, 1999: Mechanisms determining the atmospheric response to midlatitude SST anomalies. J. Climate, 12 , 13931408.

    • Search Google Scholar
    • Export Citation
  • Peng, S., L. A. Mysak, H. Ritchie, J. Derome, and B. Dugas, 1995: The differences between early and midwinter atmospheric responses to sea surface temperature anomalies in the northwest Atlantic. J. Climate, 8 , 137157.

    • Search Google Scholar
    • Export Citation
  • Peng, S., W. A. Robinson, and M. P. Hoerling, 1997: The modeled atmospheric response to midlatitude SST anomalies and its dependence on background circulation states. J. Climate, 10 , 971987.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., and F-F. Jin, 1997: Antarctic circumpolar wave: An indication of ocean–atmosphere coupling in the extratropics. Geophys. Res. Lett., 24 , 25852588.

    • Search Google Scholar
    • Export Citation
  • Ratcliffe, R. A. S., and R. Murray, 1970: New lag associations between North Atlantic sea temperatures and European pressure, applied to long-range weather forecasting. Quart. J. Roy. Meteor. Soc., 96 , 226246.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., and D. C. Marsico, 1993: An improved real-time global sea surface temperature analysis. J. Climate, 6 , 114119.

  • Sardeshmukh, P. D., 1993: The baroclinic chi problem and its application to the diagnosis of atmospheric heating rates. J. Atmos. Sci., 50 , 10991112.

    • Search Google Scholar
    • Export Citation
  • Sardeshmukh, P. D., and B. J. Hoskins, 1991: The generation of global rotational flow by idealized tropical divergence. J. Atmos. Sci., 45 , 12281251.

    • Search Google Scholar
    • Export Citation
  • Simmonds, I., and K. Keay, 2000: Mean Southern Hemisphere extratropical cyclone behavior on the 40-year NCEP–NCAR reanalysis. J. Climate, 13 , 873885.

    • Search Google Scholar
    • Export Citation
  • Slutz, R. J., S. J. Lubker, J. D. Hiscox, S. D. Woodruff, R. L. Jenne, D. H. Joseph, P. M. Steurer, and J. D. Elms, 1985: Comprehensive ocean–atmosphere data set; release 1. NOAA Environmental Research Laboratory, Boulder, CO, 268 pp.

    • Search Google Scholar
    • Export Citation
  • Snedecor, G. W., and W. G. Cochran, 1980: Statistical Methods. Iowa State University Press, 507 pp.

  • Sverdrup, H. U., 1947: Wind-driven currents in a baroclinic ocean; with application to the equatorial currents of the eastern Pacific. Proc. Natl. Acad. Sci. USA, 33 , 318326.

    • Search Google Scholar
    • Export Citation
  • Talley, L. D., 1999: Simple coupled midlatitude climate models. J. Phys. Oceanogr., 29 , 20162037.

  • Wallace, J. M., and D. Gutzler, 1981: Teleconnections in the geopotential height field during Northern Hemisphere winter. Mon. Wea. Rev., 109 , 784812.

    • Search Google Scholar
    • Export Citation
  • Wallace, J. M., and Q-R. Jiang, 1987: On the observed structure of the interannual variability of the observed ocean/atmosphere climate system. Atmospheric and Oceanic Variability, H. Cattle, Ed., Roy. Meteor. Soc., 17–43.

    • Search Google Scholar
    • Export Citation
  • White, W. B., 1995: Design of a global observing system for basin-scale upper ocean temperature anomalies. Progress in Oceanography, Vol. 36, Pergamon, 169–217.

    • Search Google Scholar
    • Export Citation
  • White, W. B., . 2000: Tropical coupled Rossby waves in the Pacific ocean–atmosphere system. J. Phys. Oceanogr., 30 , 12451264.

  • White, W. B., and R. Peterson, 1996: An Antarctic circumpolar wave in surface pressure, wind, temperature, and sea ice extent. Nature, 380 , 699702.

    • Search Google Scholar
    • Export Citation
  • White, W. B., and D. R. Cayan, 2000: A global ENSO wave in surface temperature and pressure and its interdecadal modulation from 1900 to 1996. J. Geophys. Res., 105 , 1122311242.

    • Search Google Scholar
    • Export Citation
  • White, W. B., S-C. Chen, and R. Peterson, 1998: The Antarctic Circumpolar Wave: A beta-effect in ocean–atmosphere coupling over the Southern Ocean. J. Phys. Oceanogr., 28 , 23452361.

    • Search Google Scholar
    • Export Citation
  • White, W. B., and R. J. Allan, 2002: Positive feedbacks between the Antarctic circumpolar wave and the El Niño–Southern Oscillation wave. J. Geophys. Res., in press.

    • Search Google Scholar
    • Export Citation
  • Xie, P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78 , 25392558.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Animation sequences of maps of interannual SST and SLP anomalies over the Southern Ocean for 10 yr from Jan 1983 to Jan 1993, with maps displayed every 6 months. Each map in the sequence extends meridionally from 30° to 70°S and around the globe from 30° to 30°E. Positive (negative) anomalies range from yellow to red (blue to purple) with contours of 0.1 K and 0.25 hPa

  • Fig. 2.

    (a) Zonal wavenumber–frequency spectra of SST and SLP anomalies along 50°S, extending around the globe, computed for 18 yr from 1982 to 1999. (b) Standing wave spectra computed by taking the smallest spectral energy estimate in the left- and right-hand quadrants in each zonal wavenumber–frequency band. (c) Propagation wave spectra computed by taking the difference between spectral energy estimates in the left- and right-hand quadrants in each zonal wavenumber–frequency band. Contours give the 90% confidence levels (Snedecor and Cochran 1980). Shading is for effect

  • Fig. 3.

    (a) Distribution of the temporal cross correlation between interannual SST and QS+QE anomalies at each grid point for the 10 yr from Jan 1983 to Jan 1992 over the Southern Ocean from 20° to 60°S. (b) Distribution of the temporal correlation between interannual SST and QE anomalies at each grid point for the 10 yr from Jan 1983 to Jan 1992 over the Southern Ocean from 20° to 60°S. (c) Distribution of the temporal correlation between interannual SST and PCP anomalies at each grid point for the same period. Positive (negative) correlations are unhatched (hatched). Correlations greater than 0.4 are significant at the 90% confidence level for ∼15 effective temporal degrees of freedom (Snedecor and Cochran 1980) computed using a temporal decorrelation scale of 8 months. The thick gray band follows the path of the ACW given in White et al. (2002). In the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean, this path overlies positive correlations between SST anomalies and QS+QE, QE, and PCP anomalies. The thin gray band follows the path of maximum cyclogenesis density for winter (Jun–Jul–Aug; Simmonds and Keay 2000).>

  • Fig. 4

    a. (a) The path of the ACW across the eastern Atlantic, Indian, and western and central Pacific sectors of the Southern Ocean from 0° eastward to 120°W, taken from White et al. (2002). (b) Time–distance diagrams of interannual SST, QE, PCP, D200, and SLP anomalies for the 10 yr from 1984 to 1993 along the solid gray path in (a). Positive (negative) anomalies are unhatched (hatched). (c) Zonal-lag cross correlations between SST anomalies and QE, PCP, D200, D850, and SLP anomalies in (b). The 90% confidence level is 0.4 for ∼15 effective temporal degrees of freedom (Snedecor and Cochran 1980). Units and contour intervals are given in the heading of each time–distance diagram

  • Fig. 4

    b. Same as in Fig. 4a but for the solid gray path of the ACW across the eastern Pacific and western Atlantic sectors of the Southern Ocean in the vicinity of the Drake Passage from 120°W eastward to 0°.

  • Fig. 5.

    Schematic diagram over an extratropical east–west cross section in the Southern Ocean for (a) the shallow diabatic heating scenario proposed by Hoskins and Karoly (1981) and (b) the deep diabatic heating scenario proposed by Palmer and Sun (1985). In the shallow diabatic heating scenario, the anomalous circulation shows a warm-core low (WL) displaced to the east and a cold-core high (CH) displaced to the west of low-level diabatic heating (QD) anomaly, with equatorward wind anomalies decreasing with height directly overhead and vertical descent occurring at the midlevel. In the deep diabatic heating scenario, the circulation shows a warm-core high (WH) displaced to the east and a cold-core low (CL) displaced to the west of the midlevel net heating from the sum of diabatic heating anomaly (QD) and horizontal eddy heat flux divergence anomaly (ED), with poleward wind anomalies increasing with height and vertical ascent occurring at the midlevel. The time–mean wind (u) in both scenarios are westerly, increasing with height

  • Fig. 6.

    (a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and components of the anomalous temperature budget [Eq. (6.4)]; i.e., V·∇T′, V′·∇T, ω′(T/θ)θ/∂p, Div〈VT″〉′, and their sum, the latter giving the residual computation of QD′/(ρCP) in Eq. (6.4). These regression coefficients extend over the height of the troposphere from 900 to 200 hPa, with zonal-lag regression computed along the path following the time–distance diagrams in Fig. 4a from 1984 to 1993. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Also displayed are zonal-lag regression coefficients between interannual SST anomalies and interannual QS+QE, QE, and PCP anomalies. Usual regression coefficients are multiplied by the rms of the SST anomalies so that these regression coefficients yield both magnitude and phase information in units of 10−6 K s−1 associated with unit-K SST variability. Similar regression coefficients for PCP, QS+QE, and QE anomalies are in units of mm month−1, W m−2, and W m−2, respectively, in association with unit-K SST variability

  • Fig. 7.

    (a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and the vertical pressure velocity anomaly (−ω′, positive upward) in Eq. (6.4), extending over the height of the troposphere from 900 to 200 hPa along a path following the time–distance diagrams in Fig. 4a from 1984 to 1993. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Also displayed are zonal-lag regression coefficients between interannual SST anomalies and interannual QS+QE, QE, and PCP anomalies. Usual regression coefficients are multiplied by the rms of the SST anomalies so that ω′ has units of 10−4 hPa s−1 in association with unit-K SST variability. Similar regression coefficients for PCP, QS+QE, and QE anomalies are in units of mm month−1, W m−2, and W m−2, respectively, in association with unit-K SST variability

  • Fig. 8.

    (a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and components of the anomalous vorticity budget [Eq. (6.1)]; i.e., V·ζ′, V′·ζ, βυ′, and their sum, the latter expected to yield a residual −f0∇·V′, which is similar to that computed directly. These regression coefficients extend over the height of the troposphere from 900 to 200 hPa, with zonal-lag regression computed along the path following the time–distance diagrams in Fig. 4a from 1983 to 1992. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Also displayed are zonal-lag regression coefficients between interannual SST anomalies and interannual QS+QE, QE, and PCP anomalies. Usual regression coefficients are multiplied by the rms of the SST anomalies so that these regression coefficients yield both magnitude and phase in units of 10−12 s−2 associated with unit-K SST variability. Similar regression coefficients for PCP, QS+QE, and QE anomalies are in units of mm month−1, W m−2, and W m−2, respectively, in association with unit-K SST variability

  • Fig. 9.

    (a) Vertical sections of zonal-lag regression coefficients between interannual SST anomalies and components of the anomalous potential vorticity budget [Eq. (6.6); i.e., V·ζ′, V·∂/∂z{[f0g/(N2T)]T′}, βυ′, V′·ζ, V′·∂/∂z{[f0g/(N2T)]∇T}, [f0g/(N2T)](∂V/∂z·T′ + ∂V′/∂z·T), and their residual (i.e., ∂/∂z{QD′/(ρCP)[f0g/(N2T)]}). These regression coefficients extend over the height of the troposphere from 900 to 300 hPa along the path following the time–distance diagrams in Fig. 4a from 1983 to 1992. (b) Same as in (a) but for the path following the time–distance diagrams in Fig. 4b. Usual regression coefficients are multiplied by the rms of the SST anomalies so that displayed regression coefficients yield both magnitude and phase in units of 10−12 s−2 associated with unit-K SST variability

  • Fig. 10.

    Schematic diagram along a zonal cross section of the troposphere in the Southern Ocean summarizing the deep diabatic heating scenario observed in the ACW. Positive SST-induced latent heat flux anomalies are associated with anomalous mid- to upper-level diabatic heating (QD > 0) and anomalous low-level diabatic cooling (QD < 0). This profile of anomalous diabatic heating is balanced by a combination of anomalous vertical and horizontal heat advection, yielding anomalous ascent and poleward wind over the column. The anomalous ascent (ω′) achieves maximum value between 500 and 300 hPa, producing weak low-level convergence and strong upper-level divergence anomalies. The low-level convergence is balanced principally by the planetary vorticity advection and the upper-level divergence is balanced principally by relative vorticity advection, yielding a poleward equivalently barotropic wind response to SST anomalies

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 230 31 4
PDF Downloads 91 43 5