The Response of the Southern Hemisphere Atmospheric Circulation to an Enhanced Greenhouse Gas Forcing

Jenny Brandefelt Department of Meteorology, Stockholm University, Stockholm, Sweden

Search for other papers by Jenny Brandefelt in
Current site
Google Scholar
PubMed
Close
and
Erland Källén Department of Meteorology, Stockholm University, Stockholm, Sweden

Search for other papers by Erland Källén in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The response of the atmospheric circulation to an enhanced radiative greenhouse gas forcing is investigated. It has been proposed that the response of the climate system to an enhanced forcing projects directly onto the preexisting natural modes of variability. An evaluation of this possibility and in particular of the implications of unchanged flow regimes is performed with a focus on the Southern Hemisphere extratropical atmospheric circulation. Low-pass-filtered mean sea level pressure and geopotential height at 500 and 200 hPa from a transient integration with a coupled global climate model is used.

The response to an enhanced forcing projects strongly onto the leading modes of present-day variability, in agreement with other studies. However, the spatial patterns of the leading modes are changed in response to enhanced forcing. The first and second modes of interweekly variability are the Pacific–South American modes, zonal wavenumber-3 wave trains from the central Pacific to the southern Atlantic. In response to the enhanced forcing, the spatial patterns of these modes change, and the wave train extends along a circumpolar path with amplitude also in the Eastern Hemisphere. This change in the spatial patterns is associated with a strengthening of the waveguide for barotropic Rossby waves.

Corresponding author address: Dr. Jenny Brandefelt, Department of Meteorology, Stockholm University, S-10691 Stockholm, Sweden. Email: jenny@misu.su.se

Abstract

The response of the atmospheric circulation to an enhanced radiative greenhouse gas forcing is investigated. It has been proposed that the response of the climate system to an enhanced forcing projects directly onto the preexisting natural modes of variability. An evaluation of this possibility and in particular of the implications of unchanged flow regimes is performed with a focus on the Southern Hemisphere extratropical atmospheric circulation. Low-pass-filtered mean sea level pressure and geopotential height at 500 and 200 hPa from a transient integration with a coupled global climate model is used.

The response to an enhanced forcing projects strongly onto the leading modes of present-day variability, in agreement with other studies. However, the spatial patterns of the leading modes are changed in response to enhanced forcing. The first and second modes of interweekly variability are the Pacific–South American modes, zonal wavenumber-3 wave trains from the central Pacific to the southern Atlantic. In response to the enhanced forcing, the spatial patterns of these modes change, and the wave train extends along a circumpolar path with amplitude also in the Eastern Hemisphere. This change in the spatial patterns is associated with a strengthening of the waveguide for barotropic Rossby waves.

Corresponding author address: Dr. Jenny Brandefelt, Department of Meteorology, Stockholm University, S-10691 Stockholm, Sweden. Email: jenny@misu.su.se

1. Introduction

The response of the climate system to an enhanced radiative forcing imposed by increasing atmospheric greenhouse gas (GHG) concentrations is of great importance to mankind. This enhanced forcing affects the mean state and the variability of the climate system. The variability is physically coupled to the mean state and may respond also to the changes in the mean state. The aim of this study is to try to establish physically based connections between the mean flow response to an enhanced GHG forcing and the response in extratropical variability, using a transient integration of a coupled global climate model (CGCM).

It has been argued that the spatial patterns of the response to anthropogenic forcing may project onto modes of natural climate variability (Palmer 1999). Support for this hypothesis comes from analysis of the observational record and CGCM output. Several papers have reported trends in the Antarctic Oscillation (AAO; Thompson et al. 2000; Marshall 2003) and the Arctic Oscillation (AO; Thompson et al. 2000; Feldstein 2002; Frauenfeld and Davis 2003; Ostermeier and Wallace 2003) over the past few decades. A trend in the AAO is also found in many transient greenhouse warming integrations (Fyfe et al. 1999; Kushner et al. 2001; Stone et al. 2001; Cai et al. 2003). The modeled response of the AO in transient greenhouse warming integrations, however, varies among different models. Fyfe et al. (1999) and Shindell et al. (1999) both find trends in the AO, corresponding in sign to the observed trend. However, Zorita and González-Rouco (2000) find that the trend in the AO is strongly dependent on the CGCM and on the initial conditions of the simulations. Shindell et al. (1999) find that the trend in the AO only exists in simulations with enhanced resolution in the stratosphere. Gillett and Allen (2002), however, find that the trend in the AO is not affected when they increase the stratospheric resolution and raise the upper boundary of their model.

The AAO and AO are often referred to as the annular modes (Limpasuvan and Hartmann 1999) due to their significant zonal mean components. The trends in these modes in response to anthropogenic forcing are thus mainly associated with the zonal mean response of the climate system. In particular, the observed trend in the AAO entails a strengthening of the circumpolar vortex and intensification in the westerlies that encircle Antarctica, and the observed trend in the AO entails a contraction of the Arctic polar vortex. The polar vortices result from strong meridional temperature gradients. The trends in the AAO and AO are thus to a large extent a result of changes in the position and the strength of these meridional temperature gradients.

The hypothesis that the response to anthropogenic forcing projects onto modes of natural variability implies that the modes of variability do not themselves change in response to the forcing. The alternative to this hypothesis of unchanged flow regimes is that the geographical structure of the flow patterns is altered in response to the forcing. Changes in the mean climate in response to anthropogenic forcing may entail changes in the geographical structure of the flow patterns. In this study, the response to an enhanced GHG forcing is evaluated separated into the zonal mean and the variability in the deviations from the zonal mean. This decomposition is likely to be better for the Southern Hemisphere (SH) extratropical atmospheric circulation than for the NH due to the weaker influence of the land–sea distribution and topography on the SH extratropical atmospheric circulation. Possible physically based connections between the zonal mean flow response to an enhanced GHG forcing and the response in extratropical variability are investigated. These ideas are found to be supported in model results from the SH extratropics. Therefore, this article focuses on the SH.

In this paper, the results of the statistical methods to assess the leading hemispheric modes of extratropical variability are complemented with local measures of extratropical variability. The main components of the extratropical variability are traveling cyclones and persistent anticyclones. A number of recent modeling studies address the possible influence of enhanced GHG concentrations on cyclone activity (König et al. 1993; Hall et al. 1994; Lambert 1995; Beersma et al. 1997; Zhang and Wang 1997; Carnell and Senior 1998; Schubert et al. 1998; Sinclair and Watterson 1999; Knippertz et al. 2000; Geng and Sugi 2003). All of these studies treat the NH, and four of them also treat the SH cyclone activity. The studies are performed with data from models with differing complexity and resolution, on transient integrations, equilibrium integrations, and time slice integrations (see Geng and Sugi 2003 for a summary). The collected results indicate a tendency for a decrease in the total number of cyclones and an increase in strong cyclones as an impact of global warming for both hemispheres. Recently, Fyfe (2003) finds a dramatic decrease in the number of cyclones in the region 40°–60°S in both observational data and a CGCM simulation of the SH. The decrease in this region is accompanied with an increase south of 60°S. Fyfe concludes that these changes in the number of cyclones are associated with a poleward shift in zonal mean baroclinicity.

This study aims to investigate possible physically based connections between the zonal mean flow response to an enhanced GHG forcing and the response in extratropical variability. The CGCM data and the methods used are described in section 2. As a basis for the study of the response in SH extratropical variability, the mean response in mean sea level pressure (MSLP) and geopotential height at 500 hPa (H500) and at 200 hPa (H200) is presented in section 3. The leading modes of intermonthly and interweekly variability in the CGCM are presented in section 4. The significance of these modes for midlatitude variability is also discussed. In section 5, the connection between the mean response to the enhanced GHG forcing and the response in midlatitude variability is examined. For comparison with other studies, the projection of the mean response onto the leading modes of present-day variability in the CGCM is determined in section 6. The spatial patterns of the leading modes of variability are shown to change in response to the enhanced forcing, and a possible dynamical connection to the zonal mean flow response is discussed. Section 7 summarizes the findings.

2. Model, data, and methods

a. Model and data

Data from the global coupled ocean–atmosphere–sea ice–land surface climate model ECHAM4/OPYC3 is used. The atmospheric part of this model has a spectral resolution of T42 in the horizontal dimension and 19 vertical layers. The coupling involves annual mean flux correction, restricted to heat and freshwater fluxes, in order to avoid climate drift. A 240-yr simulation of transient greenhouse warming is used. This simulation starts from equilibrium conditions with the ocean model spun up using present-day GHG concentrations (Roeckner et al. 1999). The atmospheric GHG forcing is prescribed using observations for the period from 1860 until 1990. Thereafter, the Intergovernmental Panel on Climate Change (IPCC) IS92a scenario for radiative GHG forcing is used (Houghton et al. 1992). The effects of sulphate aerosols and tropospheric ozone are not taken into account. Further details on the model and design of simulations used in this study are given in Roeckner et al. (1999). Also, Hu et al. (2001) describe the impact of global warming on the interannual and interdecadal climate modes in this model for the region 20°S–90°N.

Daily data are used in this study. The spatial domain is from the South Pole to 20°S. Periods of 30 yr from the transient integration (1860–89, 1890–1919, … , 2070–99) are considered. The analysis is performed on anomalies defined with respect to the seasonal cycle. The seasonal cycle at each grid point is obtained using harmonic analysis (retaining the first and second harmonic) with a Lanczos window. Prior to the harmonic analysis, the grand mean and linear trend are removed.

The CGCM is compared to reanalysis data from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR; Kalnay et al. 1996) for the years 1958–87. Reanalyses of the SH suffer from sparse observations. The reanalysis from the European Centre for Medium-Range Weather Forecasts (ECMWF; ECMWF 1997) for the years 1979–93 provides an improved representation compared to the NCEP–NCAR reanalysis due to inclusion of data from satellite measurements. The storm track activity, for instance, is less intense in the NCEP– NCAR reanalysis compared to the ECMWF reanalysis. However, to have the same period of data in the reanalysis as in the CGCM (30 yr), the NCEP–NCAR reanalysis is used for the comparison in this study.

b. Methods

In studies of the projection of the response to an enhanced GHG forcing onto leading modes of variability, annual mean (Stone et al. 2001; Cai et al. 2003), seasonal mean (Zorita and González Rouco 2000), or monthly mean (Fyfe et al. 1999) data are used. For comparison with the studies just mentioned, daily low-pass-filtered (periods greater than 30 days) data are used to determine the leading modes of intermonthly variability. The traveling cyclones and persistent anticyclones, characteristic of midlatitude variability, are active on time scales of a day to a week (midlatitude cyclones) and a week to a couple of weeks (persistent midlatitude anticyclones). The leading modes of interweekly variability are determined using daily low-pass-filtered (periods greater than 10 days) data to investigate the physical connections between the response in the mean flow and in the midlatitude variability. Midlatitude variability is here defined as the deviations from the zonal mean. The zonal mean is therefore removed from these filtered data before the leading modes are determined.

Daily anomalies (see section 2a) are filtered using a low-pass filter with a Lanczos window (Hamming 1989) with a cutoff period at 10 days and 30 days, respectively. These filtered data will henceforth be referred to as interweekly variability (IWV; periods greater than 10 days) and intermonthly variability (IMV; periods greater than 30 days).

The leading modes of variability are determined using empirical orthogonal function (EOF) analysis [i.e., principal component (PC) analysis] of daily IWV and IMV. Area weights, reflecting the area represented by each grid point, are used in computing the covariance matrix. For a field A(x, y), one may write
i1520-0442-17-22-4425-e1
where ei(x, y) are the EOFs. The projection (Pi) of a field A(x, y) onto a set of EOFs is defined as
i1520-0442-17-22-4425-e2
To determine if the EOFs are clearly separated, the criterion of North et al. (1982) is used. This rule of thumb says that there is evidence of mixing between two EOFs if the difference between their respective eigenvalues, (λjλi), is smaller than one standard error of the eigenvalue (Δλi); that is,
i1520-0442-17-22-4425-e3
where n is the number of independent samples. EOFs in the model integrations are retained only if (Δλi/|λjλi|) < 0.5.

Persistent events in the principal components (the time series of the amplitude of the EOF) associated with the leading EOFs are searched for using the procedure outlined by Dole and Gordon (1983). A persistent event is identified when the corresponding PC is greater than (less than) one standard deviation for 8 days or longer. The mean anomaly at each grid point for all days that belong to an event is determined. The connection between the large-scale statistical modes of variability and the midlatitude variability can thus be assessed.

Persistent anticyclones, or blocking, are determined from daily H500 anomalies (see section 2a) using the same technique. If a positive H500 anomaly of magnitude greater than 100 geopotential meters (gpm) persists for 5 days or longer, it is counted as a persistent positive anomaly event.

Following Blackmon (1976), baroclinic wave activity (i.e., midlatitude traveling cyclones) is often quantified by the variability of the bandpass-filtered H500, often called the storm track activity. Here, storm track activity is determined as the standard deviation of the (2.5–8 day) bandpass-filtered H500 following Christoph et al. (1995).

3. The mean response

The increase in atmospheric GHG concentrations is negligible during the first 30 yr of the transient integration (years 1860–89). This period is used to represent present-day climate, and the response to the enhanced GHG forcing is defined as the difference between the last and the first 30 yr of the transient integration (years 2070–99 minus 1860–89). The direct effect of the enhanced radiative GHG forcing is a heating of the climate system due to absorption of the increased thermal radiation. The increased thermal radiation is mainly absorbed at the earth's surface where the heating is larger over land than over ocean. The heating at the surface is redistributed in the vertical and horizontal by dynamical processes in the atmosphere (and oceans) such as convection, baroclinicity, winds, and ocean currents. The zonal mean surface temperature [temperature at 1000 hPa (T100)] response to the enhanced GHG forcing is shown in Fig. 1. The larger warming over land than over ocean results in a zonal mean meridional gradient in the heating response with a minimum over the Southern Ocean. The SH response is similar in winter and summer with an increase of the T1000 difference between the equator and approximately 60°S and a decrease of the T1000 difference between 60°S and the South Pole. The NH response in T1000 differs substantially from the SH response. In both winter and summer, the T1000 difference between the equator and approximately 45°N is decreased. The T1000 difference between the equator and the North Pole is decreased in NH winter (December–February) whereas the T1000 difference between 45°N and the Arctic is increased in NH summer (June–August). The substantial heating of the Arctic in NH winter is associated with melting of the sea ice. In the following, the response to the enhanced GHG forcing in the SH (south of 20°S) winter (June–August) and SH summer (December–February) MSLP, H500, and H200 is discussed.

The zonal mean winter and summer responses in H200, H500, and MSLP are shown in Fig. 2. The zonal mean response in MSLP demonstrates a strengthening of the meridional gradient between the subtropical anticyclones and the midlatitude minimum in MSLP associated with the traveling cyclones. The response in MSLP demonstrates a seasonal cycle that concurs with the seasonal cycle in the position of the Hadley circulation. A possible explanation for the amplification of the subtropical anticyclones is thus an increased Hadley circulation in response to the increased meridional temperature gradient. A possible explanation for the deepening of the midlatitude minimum in MSLP is increased cyclonic activity in response to the increased meridional temperature gradient. The response in midlatitude variability is discussed in section 5. These dynamical effects of the increased meridional temperature gradient are also seen in the H500 and H200 response in Fig. 2. The hemispheric mean response increases with height due to the increase in the hemispheric mean temperature. Furthermore, a heating maximum in the response to the enhanced GHG forcing is found in the tropical upper troposphere (Roeckner et al. 1999). The effect of this heating is seen in the increase in the response in H500 and H200 north of 40°S with height. The response in the meridional gradients in H500 and H200 also implies a response in the geostrophic wind.

The SH winter and summer responses to the enhanced GHG forcing are shown in Fig. 3. The responses in H200, H500, and MSLP are equivalent barotropic in winter and summer with an extratropical zonal wavenumber 3. The magnitude of this zonal wave is comparable to the magnitude of the zonal mean response. The spatial pattern of the response in H500 is compared to the response in midlatitude variability in section 5.

To assess the statistical significance of the response in T1000, H200, H500, and MSLP, a two-sided unpaired t test is applied at each grid point. The zonal mean response in T1000 is significant at the 99% confidence level in 100% (100%) of the grid points in winter (summer). The response in H200 is significant at the 99% confidence level in 98% (100%) of the grid points in winter (summer). The response in H500 is significant at the 99% confidence level in 65% (93%) of the grid points in winter (summer). The response in MSLP is significant at the 99% confidence level in 59% (67%) of the grid points in winter (summer). Thus the response in high altitudes is stronger and more significant. Therefore, possible connections between the zonal mean flow response and the response in the variability in H200 are investigated. Midlatitude variability is diagnosed from H500, which is often used to assess the traveling cyclones and persistent anticyclones.

4. The leading modes of variability

For comparison with other studies, the leading modes of IMV in H200, H500, and MSLP for the first 30 yr of the transient integration (years 1860–89) are determined. As mentioned in section 3, this period is used to represent present-day climate and climate variability. The projection of the response to the enhanced GHG forcing onto these modes of IMV is discussed in section 6. Traveling cyclones and persistent anticyclones, characteristic of midlatitude variability, are active on time scales of a day to a couple of weeks. The leading modes of IWV are therefore determined to investigate the physical connections between the response in the mean flow and the response in the midlatitude variability. Midlatitude variability is here defined as the deviations from the zonal mean. The zonal mean is therefore removed from the (IWV) data before the leading modes are determined. The leading modes of IWV for the first 30 yr of the transient integration (years 1860–89) are described in section 4b. These present-day modes of IWV are compared to the leading modes of IWV for the last 30 yr of the transient integration (years 2070–99) in section 6b. The strongest and most significant response to the enhanced GHG forcing is found in the upper troposphere. Therefore this article focuses on the EOFs of IMV and IWV in H200.

a. Intermonthly variability

EOF analysis is applied to daily IMV in H200, H500, and MSLP in the first 30 yr (years 1860–89) of the transient integration for the winter (June–August) and summer (December–February) season separately.

The first EOF of IMV in H200 in the first 30 yr of the transient integration (years 1860–89) is shown in Fig. 4. This mode resembles the AAO (Rogers and van Loon 1982; Thompson and Wallace 2000). In addition to the annular appearance of this mode, it exhibits a zonal wavenumber 3 of significant amplitude in summer. The spatial patterns of the first EOFs of MSLP and H500 are very similar to the spatial pattern of the first EOF of H200 (pattern correlation, 0.91–0.99). The variability explained by this AAO mode decreases with height. It accounts for 29 (23)% of IMV in MSLP, 22 (19)% of IMV in H500, and 18 (16)% of IMV in H200 in winter (summer).

Only the first EOF of winter MSLP clearly satisfies the criterion of North et al. (1982). For the other combinations of variable and season, a minimum of three EOFs satisfy the criterion. The second and third EOFs of winter H200 are shown in Fig. 5. These two modes exhibit zonal wavenumber-3 patterns in quadrature with each other and a well-defined wave train from the central Pacific to Argentina. They resemble the so-called Pacific–South American (PSA) modes (Mo and Ghil 1987; Lau et al. 1994). The PSA modes in H200 exhibit similar spatial patterns in summer and winter. The spatial patterns of the second and third EOFs of H500 and summer MSLP are similar to the corresponding modes of H200 (pattern correlation, 0.81–0.98). Each PSA mode accounts for 9%–16% of the IMV in H200, H500, and MSLP in each season.

b. Interweekly variability

EOF analysis is applied to daily IWV in H200, H500, and MSLP in the first 30 yr (years 1860–89) of the transient integration for the winter (June–August) and summer (December–February) seasons separately. The zonal mean is removed before the leading modes are determined.

The first and second EOFs of IWV in winter H200 are shown in Fig. 6. These two modes are similar to the PSA modes of IMV (see Fig. 5), respectively, and will henceforth be termed PSA1 (EOF1 of IWV) and PSA2 (EOF2 of IWV). Only the first two EOFs of IWV in H200 are clearly separated according to the North et al. (1982) criteria. The PSA1 explains 12% (14%) and the PSA2 explains 10% (12%) of IWV in H200 in winter (summer). The spatial patterns and variabilities explained by the PSA1 and PSA2 of IWV in MSLP and H500 are similar to the PSA1 and PSA2 in H200. The PSA modes of IWV in H200 exhibit similar spatial patterns in summer and winter.

The AAO (the first EOF of IMV; see Fig. 4) is not one of the leading modes of IWV. This result holds for IWV also when the zonal mean is not removed. In this case, the EOFs are linear combinations of a PSA mode and an AAO-like zonal mean pole-to-equator gradient. The AAO is associated with a weakening/strengthening of the midlatitude gradients in geopotential height and thus with the strength of the midlatitude westerlies. The absence of a distinct AAO from the top EOFs of IWV indicates that the characteristic time scale of this mode is longer than a week.

Persistent positive (negative) events in the EOFs of H200 are selected (see section 2b) in order to study the connection between the midlatitude variability and the statistical modes of variability (i.e., EOFs). The number of events found varies when the threshold is varied, but the results presented here are not sensitive to the threshold used to select events. For the period 1860–89 in the transient integration, 17 (17) positive events and 14 (22) negative events are found for the PSA1, and 15 (20) positive events and 16 (13) negative events are found for the PSA2 in winter (summer).

Figure 7 shows the difference between anomalies in MSLP, H500, and the storm track activity (STA) associated with persistent positive and negative events in PSA1 of winter H200. The difference is multiplied by 0.5 to obtain the anomalies associated with a persistent event. First, it is noted that these modes are equivalent barotropic. Second, persistent PSA1 events are associated with substantial anomalies in H500. The strongest anomalies are found in the Western Hemisphere with maximum amplitudes of more than 200 gpm. Thus, persistent events in PSA1 are associated with persistent cyclones and anticyclones in H500. The anomalies in the STA associated with persistent PSA1 events can be physically explained by the geostrophic wind anomalies associated with the cyclones and anticyclones in the wave train. Large positive (negative) anomalies in the STA are associated with westerly (easterly) geostrophic wind anomalies. Hence, in the SH, persistent cyclones and anticyclones are partly produced by persistent large-scale equivalent barotropic waves (see also Mo and Paegle 2001). For these occasions the anomalous transient eddy forcing appears to be a result of the large-scale wave pattern rather than a forcing mechanism for the persistent anomalies.

5. Midlatitude variability

The mean H500 response to the enhanced GHG forcing is discussed in section 3. In this section, the response in midlatitude variability is described and compared to the mean response in H500. The representation of midlatitude variability is also compared to reanalysis data.

a. Storm track activity

The mean STA is shown for winter and summer for the first 30 yr of the transient integration in Fig. 8. The maximum STA is found in the latitude band 40°–70°S in both winter and summer. The primary maximum extends from the southern Atlantic Ocean over the southern Indian Ocean to the Southern Ocean south of Australia in both winter and summer. The maximum STA in the extratropics is about 10%–20% higher in winter than in summer. For comparison, the STA for the period 1958–87 in the NCEP–NCAR reanalysis is also shown in Fig. 8. The spatial pattern of STA in the transient integration is similar to the pattern in the reanalysis and the maximum in STA is about 15%30% higher in the transient integration than in the NCEP–NCAR reanalysis (see section 2).

The changes in the STA in response to the enhanced GHG forcing are shown in Fig. 9. The winter STA is intensified in mid- and high latitudes, with the largest changes occurring in the midlatitudes. The summer STA response exhibits a poleward shift and a zonal wavenumber 3. To assess the statistical significance of the response in STA, a two-sided unpaired t test is applied at each grid point. The response in STA is significant at the 99% confidence level in 60% (61%) of the grid points in winter (summer). The response in STA can be physically explained by the geostrophic wind response associated with the H500 response. Comparison of the response in the STA with the response in H500 (see Figs. 2 and 3) shows that an increased (decreased) STA is associated with a strengthening (weakening) in the H500 gradient, that is, a strengthening (weakening) of the geostrophic wind.

b. Persistent anticyclones

For most of the time, the midlatitude variability is dominated by traveling cyclones. Occasionally an anticyclone becomes stationary and persistent at some location. In section 5a, it is shown that STA changes significantly in response to the enhanced GHG forcing. In this section, possible significant changes in the frequency of occurrence of persistent anticyclones in response to the enhanced forcing are investigated.

Figure 10 shows the number of persistent anticyclones (NPA) in winter and summer H500 for the period 1860–89 in the transient integration and for the period 1958–87 in the NCEP–NCAR reanalysis. The maximum NPA is found in the latitude band 50°–70°S in both winter and summer. The primary maximum extends from the date line over the southern Pacific Ocean to the southern tip of South America with a maximum of 55 (50) events in 30 yr for winter (summer). The geographical distribution of NPA in the transient integration is similar to the distribution in the reanalysis. The maximum NPA in the reanalysis is 60 (55) events in 30 yr for winter (summer).

Persistent anticyclones in the NH often form at the eastern ends of the storm tracks over the west coasts of Europe and North America. The physical mechanisms involved in the formation and maintenance of persistent anticyclones are less well understood than the mechanisms important for midlatitude storms, but it is widely believed that transient eddies play a role in their maintenance (e.g., Green 1977; Källén 1981, 1982; Illari 1984; Shutts 1983; Hoskins and Sardeshmukh 1987; Haines and Marshall 1987; Vautard and Legras 1988; Vautard et al. 1988; Maeda et al. 2000; Arai and Mukougawa 2002; etc). Comparison of Fig. 10 and Fig. 8 shows that the majority of the persistent anticyclones form at the eastern ends of the storm track also in the SH. Comparison of Fig. 10 and Fig. 7 shows that the location of the maximum amplitude of the PSA modes concurs with the maximum in the NPA (see also Mo and Paegle 2001). Persistent anticyclones are thus partly related to the large-scale statistical modes of variability discussed in section 4.

Figure 11 shows the response to the enhanced GHG forcing in the NPA. In winter the NPA decreases in the region 30°–70°S and increases south of 70°S. In summer the NPA decreases in the region 40°–60°S and increases in a region south of Africa. To assess the statistical significance of the response in NPA, the difference in the NPA between the periods 1860–89, 1890–1919, and 1920–49 is used to characterize the natural variability in NPA. In winter and summer this “natural variability” exceeds 10 gpm in several regions south of 30°S. The response in the NPA exceeds this natural variability in less than 13% (15%) of the SH extratropics in winter (summer). The response in the NPA is thus not statistically significant.

A possible explanation for the response in the NPA is the changes in the STA in response to the enhanced forcing, that is, that an increase (decrease) in the NPA would be associated with an upstream increase (decrease) in STA. This idea is not confirmed by the results presented here, which indicate a general decrease in the NPA associated with a general increase in STA. These results are in agreement with Carnell and Senior (1998), who find that a decrease in the NPA is associated with an increase in the storms in the NH response to increasing greenhouse gases. An alternative explanation is that the response in the NPA is associated with changes in the mean flow. The maximum (minimum) in the NPA (Fig. 10) concurs with the ridge (trough) in the zonal wavenumber-1 stationary wave in H500 (not shown). However, this connection between ridges in the stationary wave pattern and a high NPA is not found between the (zonal wavenumber 3) response in the stationary wave pattern (Fig. 3) and the response in the NPA. Carnell and Senior (1998) find that the NH response in the NPA is associated with changes in the stationary wave pattern. Persistent anticyclones are rare events and thus are associated with large natural variations. This makes it difficult to detect a possible response in the frequency of occurrence of these events.

6. The response in the leading modes of variability

The response to the enhanced GHG forcing in STA is shown to be coupled to the mean response in the atmospheric circulation in section 5. In this section, the hypothesis that the response to the enhanced GHG forcing projects onto the leading modes of present-day variability is tested. For comparison with other studies, the projection [see Eq. (2)] of the winter and summer response to the enhanced GHG forcing in MSLP onto the leading modes of IMV is determined. The AAO explains 49% (17%) of the winter (summer) response in MSLP. The zonal mean strengthening of the midlatitude meridional gradient in MSLP discussed in section 3 is well described by the zonal mean component of the AAO. The small projection of the winter response onto the AAO is due to the phase difference between the zonal wavenumber-3 component in the response (Fig. 3) and the zonal wavenumber-3 component in the winter AAO (Fig. 4). The zonal wavenumber 3 in the response to the enhanced forcing in MSLP is better explained by the PSA modes and particularly PSA2 (see Fig. 5). The PSA1 explains <0.1% and the PSA2 explains 21% of the response in summer MSLP [the PSA modes of winter MSLP do not satisfy the criterion of North et al. (1982)].

a. Interweekly variability

The leading modes of IWV in winter H200 for the last 30 yr of the transient integration are shown in Fig. 12. These modes are similar to the modes of IWV in H200 for the first 30 yr (Fig. 6). Figure 13 shows the difference between anomalies in MSLP, H500, and STA associated with persistent positive and negative events in PSA1 of winter H200. The difference is multiplied by 0.5 to obtain the anomalies associated with a persistent event. Again, PSA1 is an equivalent barotropic mode associated with substantial anomalies in MSLP, H500, and STA.

For comparison with the modes of IMV the projection of the response in H200 onto PSA1 and PSA2 in IWV in H200 is determined. The zonal mean is removed from the data of IWV before the leading modes are determined, and therefore the deviation from the zonal mean response in H200 is used here. The PSA1 and PSA2 for the years 1860–89 explain 13% (0.4%) and 27% (32%), respectively, of the response in winter (summer) H200. The PSA1 and PSA2 for the years 2070–99 explain 1% (2%) and 49% (56%), respectively, of the response in winter (summer) H200. Thus, the zonally asymmetric component of the response in H200 is best described by the PSA2 mode for the years 2070–99.

b. Barotropic Rossby wave propagation

The spatial patterns of the PSA modes of IWV in winter H200 for the years 2070–99 (Fig. 12) are similar to the PSA modes of IWV in winter H200 for the years 1860–89 (Fig. 6) over the southern Pacific and the southwestern Atlantic Oceans. However, the patterns differ substantially in the Eastern Hemisphere. The spatial patterns of the PSA modes resemble stationary wave patterns found in studies with far less complicated models than the one studied here. Yang et al. (1997) present results from a study of the existence of spherical resonance in a two-layer, high-resolution, quasi-geostrophic model forced by a single isolated mountain. They find that a modest change in the shape of the zonal mean wind gives a dramatic change in the stationary wave forced by the mountain and couple these results to the stationary Rossby wavenumber. According to Hoskins and Ambrizzi (1993), the degree to which the time mean zonal mean flow acts as a potential waveguide can be estimated. The stationary barotropic Rossby wavenumber (Ks) is determined according to
i1520-0442-17-22-4425-e4
where Ks = (k2s + l2s)1/2 and
i1520-0442-17-22-4425-e5
where ϕ is the latitude and υ = U/a cosϕ is the relative rotation rate of the atmosphere; βM is cos ϕ times the meridional gradient of absolute vorticity on the sphere (see Hoskins and Ambrizzi 1993). If the group velocity of the Rossby waves is considered, it may be shown that Rossby rays are always refracted toward latitudes with larger Ks (Hoskins and Ambrizzi 1993). These results are valid for stationary barotropic Rossby waves in a zonally symmetric basic-state flow. The PSA modes are equivalent barotropic (see Figs. 7 and 13) Rossby waves in a zonally asymmetric flow.

The time mean zonal mean Ks at 200 hPa for winter and summer and for eight 30-yr periods of the transient integration (1860–89, … , 2070–99) is shown in Fig. 14. A local maximum in Ks indicating a potential waveguide is found in all curves in the midlatitudes. In response to the enhanced GHG forcing, this waveguide gradually becomes more distinct from the first 30 yr to the last 30 yr of the transient integration. The response in the zonal mean wind that determines the response in Ks is to a large extent determined by the response in zonal mean H200 (see Fig. 2) and thus coupled directly and dynamically to the enhanced GHG forcing (see section 3). The resulting change in the dispersion properties for barotropic Rossby waves is a possible explanation for the changes in the spatial patterns associated with the leading modes.

The wave trains of the PSA modes for IWV in H200 for the first and last 30 yr of the transient integration differ over the southern Atlantic and Indian Oceans. Therefore, the sectorial mean Ks for longitudes 90°W– 90°E for the first and last 30 yr of the transient integration is also determined. These results are also shown in Fig. 14. Again a more distinct waveguide is found in the last 30 yr than in the first 30 yr in both winter and summer.

The changes in the spatial patterns associated with the leading modes and the development of a more distinct waveguide in the SH midlatitudes indicate an increase in quasi-stationary wave activity in this region. This is confirmed by the diagnosis of the variance in H200. The time mean zonal mean variances for IWV in H200 for winter and summer and for eight 30-yr periods of the transient integration (1860–89, … , 2070–99) are shown in Fig. 15. The variance is determined for deviations from the zonal mean. As for Ks, the sectorial mean variance for 90°W–90°E for the first and last 30 yr of the transient integration is also shown. The maximum zonal mean H200 variance in SH midlatitudes is increased by about 9% (1%) in winter (summer) between the first and the last 30 yr of the integration. The corresponding increase for the sectorial mean H200 variance is about 9% (7%) in winter (summer). The gradual increase in H200 variance shown in Fig. 15 is in accordance with the changes in the spatial patterns associated with the leading modes and the development of a more distinct waveguide.

7. Discussion and conclusions

The response of the extratropical SH atmospheric circulation to an enhanced GHG forcing is investigated using data from a transient integration with the ECHAM4/OPYC3 CGCM. The response of the SH atmospheric circulation to the prescribed GHG forcing in this integration is similar to the response in other CGCMs, which motivates the use of only one integration in the study. The aim of this study is to find physically based connections between the mean flow response to the enhanced forcing and the response in extratropical variability. The results of the study are interpreted in view of the hypothesis that the spatial patterns of the response to anthropogenic forcing project onto modes of natural climate variability (Palmer 1999).

The meridional temperature gradient changes in response to the enhanced GHG forcing. The zonal mean response in MSLP is similar in winter and summer, with an amplification of the subtropical anticyclones and a deepening of the midlatitude minimum associated with the traveling cyclones. A possible explanation for this response is found in an intensification of the dynamical processes associated with the present-day maximum in zonal mean MSLP in the subtropics and minimum in midlatitudes, that is, the sinking motion associated with the Hadley circulation and the midlatitude traveling cyclones, respectively. Both these dynamical processes are driven by meridional temperature differences. Along with these dynamical effects of the increased meridional temperature gradient, the zonal mean responses in H500 and H200 exhibit the direct effect of the increased meridional temperature gradient. The resulting zonal mean response demonstrates a strengthening of the meridional gradient in extratropical H500 and H200 and thus increased zonal mean geostrophic winds.

The SH response to the enhanced forcing is equivalent barotropic and exhibits an extratropical zonal wavenumber 3 in MSLP, H500, and H200. Maxima (minima) in the STA response are found in regions where the response in H500 implies increased (decreased) geostrophic winds. These results are in agreement with Fyfe (2003), who finds that the number of cyclones decreases in the region 40°–60°S and increases south of 60°S in association with a poleward shift in baroclinicity. Fyfe diagnoses baroclinicity in terms of the meridional temperature gradient, which also determines the strength of the geostrophic wind at high altitudes.

The leading mode of daily IMV in MSLP, H500, and H200 is the AAO. This essentially zonal mode explains 49% of the response in MSLP in winter but only 17% in summer. The difference between winter and summer is due to a phase difference between the zonal wavenumber-3 component in the response pattern and the AAO in winter. However, the zonal mean response in MSLP is well described by the AAO in both seasons. Other studies of (annual and monthly mean) data from transient GHG integrations also indicate an upward trend in the AAO (Fyfe et al. 1999; Kushner et al. 2001; Stone et al. 2001; Cai et al. 2003) in agreement with observations (Marshall 2003). The AAO mainly describes the variability in the strength and position of the zonal mean gradient in pressure/geopotential height. The zonal mean MSLP response to an enhanced GHG forcing is a change in the strength and position of this gradient, which explains the projection of the response onto the AAO. The zonal wavenumber 3 in the MSLP response to the enhanced forcing is better explained by the PSA modes (the second and third EOFs of IMV) and particularly PSA2. The PSA1 and PSA2 exhibit zonal wavenumber-3 patterns in quadrature with each other and a well defined wave train with large amplitude in the southern Pacific–South American sector. The PSA1 explains <0.1% and the PSA2 explains 21% of the response in summer MSLP.

Extratropical variability is dominated by traveling cyclones and persistent anticyclones that are active on time scales of days to weeks. The zonally asymmetric component of the response in daily IWV is therefore explored. Furthermore, the strongest and most significant response to the enhanced GHG forcing is found in the upper troposphere. The leading modes of IWV in daily zonally asymmetric H200 are the PSA modes. Episodes of persistence in these modes are associated with persistent anticyclones and cyclones in H500 in the southern Pacific–South American sector and anomalies in STA coupled to the anomalies in the westerlies associated with the large-scale anomalies in H500. The spatial patterns of these PSA modes change in response to the enhanced GHG forcing. The extension of the zonal wavenumber-3 wave changes, which results in increased amplitude in the Eastern Hemisphere (from the southern Atlantic Ocean to New Zealand). The variance in H200 is also increased in the SH midlatitudes in agreement with these changes in wave amplitude. These changes in IWV in H200 are found to be associated with a strengthened zonal mean waveguide for barotropic Rossby waves. The PSA modes describe anomalies from the zonal mean. The deviations from the zonal mean response in H200 project strongly onto PSA2 for the last 30 yr of the transient integration. This mode accounts for 49%–56% of the deviations from the zonal mean in the response. The zonal wavenumber 3 in the response may be interpreted as a stationary Rossby wave that results from the strengthened waveguide for barotropic Rossby waves.

In summary, the SH MSLP response projects onto the AAO. These results are in agreement with other studies (Fyfe et al. 1999; Kushner et al. 2001; Stone et al. 2001). Furthermore, the spatial patterns of extratropical IWV change in response to the enhanced forcing. These changes are interpreted as a result of changes in the propagation conditions for barotropic Rossby waves. A change in the propagation of large-scale anomalies may be of great importance to climate on regional scales. The results shown here indicate a substantial zonally asymmetric component in STA response to the enhanced forcing. This component is coupled to the zonally asymmetric response in the atmospheric circulation and implies large regional differences in the response to the enhanced GHG forcing in precipitation.

Regional aspects of climate change in the Northern Hemisphere have been found to be difficult to analyze with the methods used here. Earlier studies of changes in cyclone activity and blocking frequencies have given contrasting results, mainly dependent on the climate model used rather than any aspects of the forcing scenario (e.g., Hall et al. 1994; Zhang and Wang 1997; Carnell and Senior 1998; Schubert et al. 1998; Ulbrich and Christoph 1999; Knippertz et al. 2000). It has also been found that climate change response in the Arctic area is highly variable; in addition to being the region on earth most sensitive to an increased GHG forcing, it is also the region where the response is most variable (Räisänen 2001). This sensitivity of the Arctic region is probably connected to a response sensitivity in the large-scale waves as well as a sensitivity due to sea ice changes and atmosphere–ocean heat exchange in the Arctic region. The idea of a change in the zonal mean flow giving rise to an altered wave guide structure is not a viable explanation as the wave guide properties in the Northern Hemisphere remain unchanged in the model run analyzed here. However, due to the strong seasonal dependence of the Arctic warming (Fig. 1) and the strong zonal asymmetry in the NH response (Hu et al. 2001; Carnell and Senior 1998), this possibility may not be ruled out.

Another aspect of climate change and changes in large-scale wave patterns is stratospheric major warmings and ozone depletion. In September 2002, an unprecedented major warming occurred in the SH polar stratosphere (Sinnhüber et al. 2003; Allen et al. 2003). Although the sudden warming did not break the vortex isolation (Allen et al. 2003), the increased temperatures halted the ozone loss (Hoppel et al. 2003). Theoretical and observational studies (of NH major warmings) show that major warmings are caused by the amplification of planetary waves in the troposphere (e.g., Matsuno 1971; Labitzke 1982; Itoh and Harada 2004). The change in the propagation conditions for planetary waves reported on here could possibly be of importance for the probability of SH sudden warmings and thus for ozone depletion in the SH polar winter.

Acknowledgments

We wish to thank H. Thiemann at the Deutsche Klimarechenzentrum for kindly providing us with the climate model data. This work was performed as a part of the Swedish Regional Climate Modeling Programme (SWECLIM) which is supported by the Strategic Funds for Environmental Research (MISTRA). The Grid Analysis Display System (GrADS) was used for drawing Figs. 1–13.

REFERENCES

  • Allen, D. R., R. M. Bevilacqua, G. E. Neboluha, C. E. Randall, and G. L. Manney, 2003: Unusual stratospheric transport and mixing during the 2002 Antarctic winter. Geophys. Res. Lett.,30, 1599, doi:10.1029/2003GL017117.

    • Search Google Scholar
    • Export Citation
  • Arai, M., and H. Mukougawa, 2002: On the effectiveness of the eddy straining mechanism for the maintenance of blocking flows. J. Meteor. Soc. Japan, 80 , 10891102.

    • Search Google Scholar
    • Export Citation
  • Beersma, J. J., K. M. Rider, G. J. Komen, E. Kass, and V. V. Kharin, 1997: An analysis of extra-tropical storms in the North Atlantic region as simulated in a control and 2 × CO2 times slice experiment with a high resolution atmospheric model. Tellus, 49A , 347361.

    • Search Google Scholar
    • Export Citation
  • Blackmon, M. L., 1976: A climatological spectral study of the 500 mb geopotential height of the Northern Hemisphere. J. Atmos. Sci, 33 , 16071623.

    • Search Google Scholar
    • Export Citation
  • Cai, W., P. H. Whetton, and D. J. Karoly, 2003: The response of the Antarctic Oscillation to increasing and stabilized atmospheric CO2. J. Climate, 16 , 15251538.

    • Search Google Scholar
    • Export Citation
  • Carnell, R. E., and C. A. Senior, 1998: Changes in mid-latitude variability due to increasing greenhouse gases and sulfate aerosols. Climate Dyn, 14 , 369383.

    • Search Google Scholar
    • Export Citation
  • Christoph, M., U. Ulbrich, and U. Haak, 1995: Faster determination of the intraseasonal variability of storm tracks using Murakami's recursive filter. Mon. Wea. Rev, 123 , 578581.

    • Search Google Scholar
    • Export Citation
  • Dole, R. M., and N. D. Gordon, 1983: Persistent anomalies of the extra-tropical Northern Hemisphere wintertime circulation: Geographical distribution and regional persistence characteristics. Mon. Wea. Rev, 111 , 15671586.

    • Search Google Scholar
    • Export Citation
  • ECMWF, cited 1997: The description of the ECMWF Re-analysis Global Atmospheric Data Archive. [Available online at http:// www.ecmwf.int/products/data/archive/index.html.].

    • Search Google Scholar
    • Export Citation
  • Feldstein, S. B., 2002: The recent trend and variance increase of the annular mode. J. Climate, 15 , 8894.

  • Frauenfeld, O. W., and R. E. Davis, 2003: Northern Hemisphere circumpolar vortex trends and climate change implications. J. Geophys. Res.,108, 4423, doi:10.1029/2002JD002958.

    • Search Google Scholar
    • Export Citation
  • Fyfe, J. C., 2003: Extratropical Southern Hemisphere cyclones: Harbingers of climate change? J. Climate, 16 , 28022805.

  • Fyfe, J. C., G. J. Boer, and G. M. Flato, 1999: The Arctic and Antarctic Oscillations and their projected changes under global warming. Geophys. Res. Lett, 26 , 16011604.

    • Search Google Scholar
    • Export Citation
  • Geng, Q., and M. Sugi, 2003: Possible change of extratropical cyclone activity due to enhanced greenhouse gases and sulfate aerosols— Study with a high-resolution AGCM. J. Climate, 16 , 22622274.

    • Search Google Scholar
    • Export Citation
  • Gillett, N. P., and M. R. Allen, 2002: The role of stratospheric resolution in simulating the Arctic Oscillation response to greenhouse gases. Geophys. Res. Lett.,29, 1500, doi:10.1029/2001GL014444.

    • Search Google Scholar
    • Export Citation
  • Green, J. S. A., 1977: The weather during July 1976: Some dynamical considerations of the drought. Weather, 32 , 120128.

  • Haines, K., and J. Marshall, 1987: Eddy-forced coherent structures as a prototype of atmospheric blocking. Quart. J. Roy. Meteor. Soc, 113 , 681704.

    • Search Google Scholar
    • Export Citation
  • Hall, N. M. J., B. J. Hoskins, P. J. Valdes, and C. A. Senior, 1994: Storm tracks in a high resolution GCM with doubled carbon dioxide. Quart. J. Roy. Meteor. Soc, 120 , 12091230.

    • Search Google Scholar
    • Export Citation
  • Hamming, R. W., 1989: Digital filters. Prentice Hall International, 284 pp.

  • Hoppel, K., R. Bevilacqua, D. Allen, G. Neboluha, and C. Randall, 2003: POAM III observations of the anomalous 2002 Antarctic ozone hole. Geophys. Res. Lett.,30, 1394, doi:10.1029/2003GL016899.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and P. D. Sardeshmukh, 1987: A diagnostic study of the dynamics of the Northern Hemisphere winter of 1985–1986. Quart. J. Roy. Meteor. Soc, 113 , 759778.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci, 50 , 16611671.

    • Search Google Scholar
    • Export Citation
  • Houghton, J. T., B. A. Callendar, and S. K. Varney, Eds.,. 1992: Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press, 198 pp.

    • Search Google Scholar
    • Export Citation
  • Hu, Z. Z., L. Bengtsson, E. Roeckner, M. Christoph, A. Bacher, and J. M. Oberhuber, 2001: Impact of global warming on the interannual and interdecadal climate modes in a coupled GCM. Climate Dyn, 17 , 361374.

    • Search Google Scholar
    • Export Citation
  • Illari, L., 1984: A diagnostic study of the potential vorticity in a warm blocking anticyclone. J. Atmos. Sci, 41 , 35183526.

  • Itoh, H., and K. I. Harada, 2004: Coupling between tropospheric and stratospheric leading modes. J. Climate, 17 , 320336.

  • Källén, E., 1981: The nonlinear effects of orographic and momentum forcing in a low-order, barotropic model. J. Atmos. Sci, 38 , 21502163.

    • Search Google Scholar
    • Export Citation
  • Källén, E., 1982: Bifurcation properties of quasi-geostrophic, barotropic models and their relations to atmospheric blocking. Tellus, 34 , 255265.

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

  • Knippertz, P., U. Ulbrich, and P. Speth, 2000: Changing cyclones and surface wind speeds over the North Atlantic and Europe in a transient GHG experiment. Climate Res, 15 , 109122.

    • Search Google Scholar
    • Export Citation
  • König, W. R., R. Sausen, and F. Sielman, 1993: Objective identification of cyclones in GCM simulations. J. Climate, 6 , 22172231.

  • Kushner, P. J., I. M. Held, and T. L. Delworth, 2001: Southern Hemisphere atmospheric circulation response to global warming. J. Climate, 14 , 22382249.

    • Search Google Scholar
    • Export Citation
  • Labitzke, K., 1982: On the interannual variability of the middle stratosphere during northern winters. J. Meteor. Soc. Japan, 60 , 124139.

    • Search Google Scholar
    • Export Citation
  • Lambert, S. J., 1995: The effects of enhanced greenhouse warming on winter cyclone frequencies and strengths. J. Climate, 8 , 14471452.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., P. J. Sheu, and I. S. Kang, 1994: Multiscale low-frequency circulation modes in the global atmosphere. J. Atmos. Sci, 51 , 11691193.

    • Search Google Scholar
    • Export Citation
  • Limpasuvan, V., and D. L. Hartmann, 1999: Eddies and the annular modes of climate variability. Geophys. Res. Lett, 26 , 31333136.

  • Maeda, S., C. Kobayashi, K. Takano, and T. Tsuyuki, 2000: Relationship between singular modes of blocking and high-frequency eddies. J. Meteor. Soc. Japan, 78 , 631645.

    • Search Google Scholar
    • Export Citation
  • Marshall, G. J., 2003: Trends in the southern annular mode from observations and reanalyses. J. Climate, 16 , 41344143.

  • Matsuno, T., 1971: A dynamical model of the stratospheric sudden warming. J. Atmos. Sci, 28 , 14791494.

  • Mo, K. C., and M. Ghil, 1987: Statistics and dynamics of persistent anomalies. J. Atmos. Sci, 44 , 877901.

  • Mo, K. C., and J. N. Paegle, 2001: The Pacific–South American modes and their downstream effects. Int. J. Climatol, 21 , 12111229.

  • North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev, 110 , 699706.

    • Search Google Scholar
    • Export Citation
  • Ostermeier, G. M., and J. M. Wallace, 2003: Trends in the North Atlantic Oscillation–Northern Hemisphere annular mode during the twentieth century. J. Climate, 16 , 336341.

    • Search Google Scholar
    • Export Citation
  • Palmer, T. N., 1999: A nonlinear dynamical perspective on climate prediction. J. Climate, 12 , 575591.

  • Räisänen, J., 2001: CO2-induced climate change in CMIP2 experiments: Quantification of agreement and role of internal variability. J. Climate, 14 , 20882104.

    • Search Google Scholar
    • Export Citation
  • Roeckner, E., L. Bengtsson, J. Feichter, J. Lilieveld, and H. Rodhe, 1999: Transient climate change simulations with a coupled atmosphere–ocean GCM including the sulfur cycle. J. Climate, 12 , 30043032.

    • Search Google Scholar
    • Export Citation
  • Rogers, J. C., and H. van Loon, 1982: Spatial variability of sea level pressure and 500-mb height anomalies over the Southern Hemisphere. Mon. Wea. Rev, 110 , 13751392.

    • Search Google Scholar
    • Export Citation
  • Schubert, M. J., J. Perlwitz, R. Blender, K. Fraedrich, and F. Lunkeit, 1998: North Atlantic cyclones in CO2-induced warm climate simulations: Frequency, intensity, and tracks. Climate Dyn, 14 , 827837.

    • Search Google Scholar
    • Export Citation
  • Shindell, D. T., R. L. Miller, G. A. Schmidt, and L. Pandolfo, 1999: Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature, 399 , 452455.

    • Search Google Scholar
    • Export Citation
  • Shutts, G. J., 1983: The propagation of eddies in diffluent jetstreams: Eddy vorticity forcing of ‘blocking’ flow fields. Quart. J. Roy. Meteor. Soc, 109 , 737761.

    • Search Google Scholar
    • Export Citation
  • Sinclair, M. R., and I. G. Watterson, 1999: Objective assessment of extratropical weather systems in simulated climates. J. Climate, 12 , 34673485.

    • Search Google Scholar
    • Export Citation
  • Sinnhüber, M., M. Weber, A. Amankwah, and J. P. Burrows, 2003: Total ozone during the unusual Antarctic winter of 2002. Geophys. Res. Lett.,30, 1580, doi:10.1029/2002GL016798.

    • Search Google Scholar
    • Export Citation
  • Stone, D. A., A. J. Weaver, and R. J. Stouffer, 2001: Projection of climate change onto modes of atmospheric variability. J. Climate, 14 , 35513565.

    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., and J. M. Wallace, 2000: Annular modes in the extratropical circulation. Part I: Month-to-month variability. J. Climate, 13 , 10001016.

    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., J. M. Wallace, and G. C. Hegerl, 2000: Annular modes in the extratropical circulation. Part II: Trends. J. Climate, 13 , 10181036.

    • Search Google Scholar
    • Export Citation
  • Ulbrich, U., and M. Christoph, 1999: A shift in the NAO and increasing storm track activity over Europe due to anthropogenic greenhouse gas forcing. Climate Dyn, 15 , 551559.

    • Search Google Scholar
    • Export Citation
  • Vautard, R., and B. Legras, 1988: On the source of midlatitude low-frequency variability. Part II: Nonlinear equilibration of weather regimes. J. Atmos. Sci, 45 , 28452867.

    • Search Google Scholar
    • Export Citation
  • Vautard, R., B. Legras, and M. Déqué, 1988: On the source of midlatitude low-frequency variability. Part I: A statistical approach to persistence. J. Atmos. Sci, 45 , 28112844.

    • Search Google Scholar
    • Export Citation
  • Yang, S., B. Reinhold, and E. Källén, 1997: Multiple weather regimes and baroclinically forced spherical resonance. J. Atmos. Sci, 54 , 13971409.

    • Search Google Scholar
    • Export Citation
  • Zhang, Y., and W-C. Wang, 1997: Model-simulated northern winter cyclone and anticyclone activity under a greenhouse warming scenario. J. Climate, 10 , 16161634.

    • Search Google Scholar
    • Export Citation
  • Zorita, E., and F. González-Rouco, 2000: Disagreement between predictions of the future behavior of the Arctic Oscillation as simulated in two different climate models: Implications for global warming. Geophys. Res. Lett, 27 , 17551758.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

The zonal mean response to an enhanced GHG forcing (difference between the years 2070–99 and 1860–89) in T1000 in Dec–Feb (SH summer; solid line) and Jun–Aug (SH winter; dashed line). Units are in Kelvins

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 2.
Fig. 2.

The zonal mean response to an enhanced GHG forcing (difference between the years 2070–99 and 1860–89) in H200 (squares), H500 (filled circles), and MSLP (circles) in (top) SH winter (Jun–Aug) and (bottom) summer (Dec–Feb). The vertical axes designate (from left to right) H200, H500, and MSLP. Units are in gpm and hPa

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 3.
Fig. 3.

The mean response to an enhanced GHG forcing (difference between the years 2070–99 and 1860–89) in (a), (b) H200; (c), (d) H500; and (e), (f) MSLP in SH winter (Jun–Aug) for (a), (c), and (e) and summer (Dec–Feb) for (b), (d), and (f). Units are in gpm and hPa

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 4.
Fig. 4.

The first EOF (the AAO) of IMV in H200 for the years 1860–89 in (a) winter (Jun–Aug) and (b) summer (Dec–Feb)

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 5.
Fig. 5.

The (a) second and (b) third EOF (PSA1 and PSA2, respectively) of IMV in winter (Jun–Aug) H200 for the years 1860– 89

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 6.
Fig. 6.

The (a) first and (b) second (PSA1 and PSA2) of IWV in winter (Jun–Aug) H200 for the years 1860–89

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 7.
Fig. 7.

(a) MSLP, (b) H500, and (c) STA anomaly differences between positive and negative persistent events in PSA1 of IWV in winter (Jun–Aug) H200 for the years 1860–89. The differences are multiplied by 0.5. Units are in hPa and gpm

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 8.
Fig. 8.

(a), (c) Storm track activity for the years 1860–89 of the transient integration and (b), (d) for the years 1958– 87 of the NCEP–NCAR reanalysis in winter (Jun–Aug) for (a), (b) and summer (Dec–Feb) for (c), (d). Units are in gpm

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 9.
Fig. 9.

The response to an enhanced GHG forcing (difference between the years 2070–99 and 1860–89) in the storm track activity in (a) winter (Jun–Aug) and (b) summer (Dec–Feb). Units are in gpm

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 10.
Fig. 10.

Same as in Fig. 8, but for the number of persistent anticyclones in 30 yr. Units are in number of cases per 30 yr

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 11.
Fig. 11.

Same as in Fig. 9, but for the number of persistent anticyclones in 30 yr. Units are in number of cases per 30 yr

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 12.
Fig. 12.

Same as in Fig. 6, but for the years 2070–99

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 13.
Fig. 13.

Same as in Fig. 7, but for the years 2070–99

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 14.
Fig. 14.

The (a), (b) zonal mean and (c), (d) sectorial mean (90°W–90°E) stationary Rossby wavenumber (Ks) for (a), (c) winter (Jun–Aug) and (b), (d) summer (Dec–Feb). The zonal mean Ks is shown for eight consecutive 30-yr periods in the transient integration (1860–89, 1890–1919, … , 2070–99) with individual curves ranging from light gray (1860–89) to black (2070–99). The sectorial mean Ks is shown for the periods 1860–89 (light gray) and 2070– 99 (black)

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Fig. 15.
Fig. 15.

The (a), (b) zonal mean and (c), (d) sectorial mean (90°W–90°E) variance of IWV in H200 for (a), (c) winter (Jun–Aug) and (b), (d) summer (Dec–Feb). The variance is determined for deviations from the zonal mean. The zonal mean variance is shown for eight consecutive 30-yr periods in the transient integration (1860–89, 1890–1919, … , 2070–99) with individual curves ranging from light gray (1860–89) to black (2070–99). The sectorial mean variance is shown for the periods 1860–89 (light gray) and 2070–99 (black). Units are in gpm2

Citation: Journal of Climate 17, 22; 10.1175/3221.1

Save
  • Allen, D. R., R. M. Bevilacqua, G. E. Neboluha, C. E. Randall, and G. L. Manney, 2003: Unusual stratospheric transport and mixing during the 2002 Antarctic winter. Geophys. Res. Lett.,30, 1599, doi:10.1029/2003GL017117.

    • Search Google Scholar
    • Export Citation
  • Arai, M., and H. Mukougawa, 2002: On the effectiveness of the eddy straining mechanism for the maintenance of blocking flows. J. Meteor. Soc. Japan, 80 , 10891102.

    • Search Google Scholar
    • Export Citation
  • Beersma, J. J., K. M. Rider, G. J. Komen, E. Kass, and V. V. Kharin, 1997: An analysis of extra-tropical storms in the North Atlantic region as simulated in a control and 2 × CO2 times slice experiment with a high resolution atmospheric model. Tellus, 49A , 347361.

    • Search Google Scholar
    • Export Citation
  • Blackmon, M. L., 1976: A climatological spectral study of the 500 mb geopotential height of the Northern Hemisphere. J. Atmos. Sci, 33 , 16071623.

    • Search Google Scholar
    • Export Citation
  • Cai, W., P. H. Whetton, and D. J. Karoly, 2003: The response of the Antarctic Oscillation to increasing and stabilized atmospheric CO2. J. Climate, 16 , 15251538.

    • Search Google Scholar
    • Export Citation
  • Carnell, R. E., and C. A. Senior, 1998: Changes in mid-latitude variability due to increasing greenhouse gases and sulfate aerosols. Climate Dyn, 14 , 369383.

    • Search Google Scholar
    • Export Citation
  • Christoph, M., U. Ulbrich, and U. Haak, 1995: Faster determination of the intraseasonal variability of storm tracks using Murakami's recursive filter. Mon. Wea. Rev, 123 , 578581.

    • Search Google Scholar
    • Export Citation
  • Dole, R. M., and N. D. Gordon, 1983: Persistent anomalies of the extra-tropical Northern Hemisphere wintertime circulation: Geographical distribution and regional persistence characteristics. Mon. Wea. Rev, 111 , 15671586.

    • Search Google Scholar
    • Export Citation
  • ECMWF, cited 1997: The description of the ECMWF Re-analysis Global Atmospheric Data Archive. [Available online at http:// www.ecmwf.int/products/data/archive/index.html.].

    • Search Google Scholar
    • Export Citation
  • Feldstein, S. B., 2002: The recent trend and variance increase of the annular mode. J. Climate, 15 , 8894.

  • Frauenfeld, O. W., and R. E. Davis, 2003: Northern Hemisphere circumpolar vortex trends and climate change implications. J. Geophys. Res.,108, 4423, doi:10.1029/2002JD002958.

    • Search Google Scholar
    • Export Citation
  • Fyfe, J. C., 2003: Extratropical Southern Hemisphere cyclones: Harbingers of climate change? J. Climate, 16 , 28022805.

  • Fyfe, J. C., G. J. Boer, and G. M. Flato, 1999: The Arctic and Antarctic Oscillations and their projected changes under global warming. Geophys. Res. Lett, 26 , 16011604.

    • Search Google Scholar
    • Export Citation
  • Geng, Q., and M. Sugi, 2003: Possible change of extratropical cyclone activity due to enhanced greenhouse gases and sulfate aerosols— Study with a high-resolution AGCM. J. Climate, 16 , 22622274.

    • Search Google Scholar
    • Export Citation
  • Gillett, N. P., and M. R. Allen, 2002: The role of stratospheric resolution in simulating the Arctic Oscillation response to greenhouse gases. Geophys. Res. Lett.,29, 1500, doi:10.1029/2001GL014444.

    • Search Google Scholar
    • Export Citation
  • Green, J. S. A., 1977: The weather during July 1976: Some dynamical considerations of the drought. Weather, 32 , 120128.

  • Haines, K., and J. Marshall, 1987: Eddy-forced coherent structures as a prototype of atmospheric blocking. Quart. J. Roy. Meteor. Soc, 113 , 681704.

    • Search Google Scholar
    • Export Citation
  • Hall, N. M. J., B. J. Hoskins, P. J. Valdes, and C. A. Senior, 1994: Storm tracks in a high resolution GCM with doubled carbon dioxide. Quart. J. Roy. Meteor. Soc, 120 , 12091230.

    • Search Google Scholar
    • Export Citation
  • Hamming, R. W., 1989: Digital filters. Prentice Hall International, 284 pp.

  • Hoppel, K., R. Bevilacqua, D. Allen, G. Neboluha, and C. Randall, 2003: POAM III observations of the anomalous 2002 Antarctic ozone hole. Geophys. Res. Lett.,30, 1394, doi:10.1029/2003GL016899.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and P. D. Sardeshmukh, 1987: A diagnostic study of the dynamics of the Northern Hemisphere winter of 1985–1986. Quart. J. Roy. Meteor. Soc, 113 , 759778.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci, 50 , 16611671.

    • Search Google Scholar
    • Export Citation
  • Houghton, J. T., B. A. Callendar, and S. K. Varney, Eds.,. 1992: Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment. Cambridge University Press, 198 pp.

    • Search Google Scholar
    • Export Citation
  • Hu, Z. Z., L. Bengtsson, E. Roeckner, M. Christoph, A. Bacher, and J. M. Oberhuber, 2001: Impact of global warming on the interannual and interdecadal climate modes in a coupled GCM. Climate Dyn, 17 , 361374.

    • Search Google Scholar
    • Export Citation
  • Illari, L., 1984: A diagnostic study of the potential vorticity in a warm blocking anticyclone. J. Atmos. Sci, 41 , 35183526.

  • Itoh, H., and K. I. Harada, 2004: Coupling between tropospheric and stratospheric leading modes. J. Climate, 17 , 320336.

  • Källén, E., 1981: The nonlinear effects of orographic and momentum forcing in a low-order, barotropic model. J. Atmos. Sci, 38 , 21502163.

    • Search Google Scholar
    • Export Citation
  • Källén, E., 1982: Bifurcation properties of quasi-geostrophic, barotropic models and their relations to atmospheric blocking. Tellus, 34 , 255265.

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

  • Knippertz, P., U. Ulbrich, and P. Speth, 2000: Changing cyclones and surface wind speeds over the North Atlantic and Europe in a transient GHG experiment. Climate Res, 15 , 109122.

    • Search Google Scholar
    • Export Citation
  • König, W. R., R. Sausen, and F. Sielman, 1993: Objective identification of cyclones in GCM simulations. J. Climate, 6 , 22172231.

  • Kushner, P. J., I. M. Held, and T. L. Delworth, 2001: Southern Hemisphere atmospheric circulation response to global warming. J. Climate, 14 , 22382249.

    • Search Google Scholar
    • Export Citation
  • Labitzke, K., 1982: On the interannual variability of the middle stratosphere during northern winters. J. Meteor. Soc. Japan, 60 , 124139.

    • Search Google Scholar
    • Export Citation
  • Lambert, S. J., 1995: The effects of enhanced greenhouse warming on winter cyclone frequencies and strengths. J. Climate, 8 , 14471452.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., P. J. Sheu, and I. S. Kang, 1994: Multiscale low-frequency circulation modes in the global atmosphere. J. Atmos. Sci, 51 , 11691193.

    • Search Google Scholar
    • Export Citation
  • Limpasuvan, V., and D. L. Hartmann, 1999: Eddies and the annular modes of climate variability. Geophys. Res. Lett, 26 , 31333136.

  • Maeda, S., C. Kobayashi, K. Takano, and T. Tsuyuki, 2000: Relationship between singular modes of blocking and high-frequency eddies. J. Meteor. Soc. Japan, 78 , 631645.

    • Search Google Scholar
    • Export Citation
  • Marshall, G. J., 2003: Trends in the southern annular mode from observations and reanalyses. J. Climate, 16 , 41344143.

  • Matsuno, T., 1971: A dynamical model of the stratospheric sudden warming. J. Atmos. Sci, 28 , 14791494.

  • Mo, K. C., and M. Ghil, 1987: Statistics and dynamics of persistent anomalies. J. Atmos. Sci, 44 , 877901.

  • Mo, K. C., and J. N. Paegle, 2001: The Pacific–South American modes and their downstream effects. Int. J. Climatol, 21 , 12111229.

  • North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev, 110 , 699706.

    • Search Google Scholar
    • Export Citation
  • Ostermeier, G. M., and J. M. Wallace, 2003: Trends in the North Atlantic Oscillation–Northern Hemisphere annular mode during the twentieth century. J. Climate, 16 , 336341.

    • Search Google Scholar
    • Export Citation
  • Palmer, T. N., 1999: A nonlinear dynamical perspective on climate prediction. J. Climate, 12 , 575591.

  • Räisänen, J., 2001: CO2-induced climate change in CMIP2 experiments: Quantification of agreement and role of internal variability. J. Climate, 14 , 20882104.

    • Search Google Scholar
    • Export Citation
  • Roeckner, E., L. Bengtsson, J. Feichter, J. Lilieveld, and H. Rodhe, 1999: Transient climate change simulations with a coupled atmosphere–ocean GCM including the sulfur cycle. J. Climate, 12 , 30043032.

    • Search Google Scholar
    • Export Citation
  • Rogers, J. C., and H. van Loon, 1982: Spatial variability of sea level pressure and 500-mb height anomalies over the Southern Hemisphere. Mon. Wea. Rev, 110 , 13751392.

    • Search Google Scholar
    • Export Citation
  • Schubert, M. J., J. Perlwitz, R. Blender, K. Fraedrich, and F. Lunkeit, 1998: North Atlantic cyclones in CO2-induced warm climate simulations: Frequency, intensity, and tracks. Climate Dyn, 14 , 827837.

    • Search Google Scholar
    • Export Citation
  • Shindell, D. T., R. L. Miller, G. A. Schmidt, and L. Pandolfo, 1999: Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature, 399 , 452455.

    • Search Google Scholar
    • Export Citation
  • Shutts, G. J., 1983: The propagation of eddies in diffluent jetstreams: Eddy vorticity forcing of ‘blocking’ flow fields. Quart. J. Roy. Meteor. Soc, 109 , 737761.

    • Search Google Scholar
    • Export Citation
  • Sinclair, M. R., and I. G. Watterson, 1999: Objective assessment of extratropical weather systems in simulated climates. J. Climate, 12 , 34673485.

    • Search Google Scholar
    • Export Citation
  • Sinnhüber, M., M. Weber, A. Amankwah, and J. P. Burrows, 2003: Total ozone during the unusual Antarctic winter of 2002. Geophys. Res. Lett.,30, 1580, doi:10.1029/2002GL016798.

    • Search Google Scholar
    • Export Citation
  • Stone, D. A., A. J. Weaver, and R. J. Stouffer, 2001: Projection of climate change onto modes of atmospheric variability. J. Climate, 14 , 35513565.

    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., and J. M. Wallace, 2000: Annular modes in the extratropical circulation. Part I: Month-to-month variability. J. Climate, 13 , 10001016.

    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., J. M. Wallace, and G. C. Hegerl, 2000: Annular modes in the extratropical circulation. Part II: Trends. J. Climate, 13 , 10181036.

    • Search Google Scholar
    • Export Citation
  • Ulbrich, U., and M. Christoph, 1999: A shift in the NAO and increasing storm track activity over Europe due to anthropogenic greenhouse gas forcing. Climate Dyn, 15 , 551559.

    • Search Google Scholar
    • Export Citation
  • Vautard