Eliassen-Palm Flux Divergence Research Articles

Climatology and trends in the forcing of the stratospheric zonal-mean flow

Abstract. The momentum budget of the Transformed Eulerian-Mean (TEM) equation is calculated using the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA-40) and the National Centers for Environmental Prediction (NCEP) Reanalysis 2 (R-2). This study outlines the considerable contribution of unresolved waves, deduced to be gravity waves, to the forcing of the zonal-mean flow. A trend analysis, from 1980 to 2001, shows that the onset and break down of the Northern Hemisphere (NH) stratospheric polar night jet has a tendency to occur later in the season in the more recent years. This temporal shift follows long-term changes in planetary wave activity that are mainly due to synoptic waves, with a lag of one month. In the Southern Hemisphere (SH), the polar vortex shows a tendency to persist further into the SH summertime. This also follows a statistically significant decrease in the intensity of the stationary EP flux divergence over the 1980–2001 period. Ozone depletion is well known for strengthening the polar vortex through the thermal wind balance. However, the results of this work show that the SH polar vortex does not experience any significant long-term changes until the month of December, even though the intensification of the ozone hole occurs mainly between September and November. This study suggests that the decrease in planetary wave activity in November provides an important feedback to the zonal wind as it delays the breakdown of the polar vortex. In addition, the absence of strong eddy feedback before November explains the lack of significant trends in the polar vortex in the SH early spring. A long-term weakening in the Brewer-Dobson (B-D) circulation in the polar region is identified in the NH winter and early spring and during the SH late spring and is likely driven by the decrease in planetary wave activity previously mentioned. During the rest of the year, there are large discrepancies in the representation of the B-D circulation and the unresolved waves between the two reanalyses, making trend analyses unreliable.

Influence of Gravity Waves in the Tropical Upwelling: WACCM Simulations

Abstract The annual cycle of tropical upwelling and contributions by planetary and gravity waves are investigated from climatological simulations using the Whole Atmosphere Community Climate Model (WACCM) including three gravity wave drag (GWD) parameterizations (orographic, nonstationary background, and convective GWD parameterizations). The tropical upwelling is estimated by the residual mean vertical velocity at 100 hPa averaged over 15°S–15°N. This is well matched with an upwelling estimate from the balance of the zonal momentum and the mass continuity. A clear annual cycle of the tropical upwelling is found, with a Northern Hemispheric (NH) wintertime maximum and NH summertime minimum determined primarily by the Eliassen–Palm flux divergence (EPD), along with a secondary contribution from the zonal wind tendency. Gravity waves increase tropical upwelling throughout the year, and of the three sources the contribution by convective gravity wave drag (CGWD) is largest in most months. The relative contribution by all three GWDs to tropical upwelling is not larger than 5%. However, when tropical upwelling is estimated by net upward mass flux between turnaround latitudes where upwelling changes downwelling, annual mean contribution by all three GWDs is up to 19% at 70 hPa by orographic and convective gravity waves with comparable magnitudes. Effects of CGWD on upwelling are investigated by conducting an additional WACCM simulation without CGWD parameterization. It was found that including CGWD parameterization increases tropical upwelling not only directly by adding CGWD forcing, but also indirectly by modulating EPD and zonal wind tendency terms in the tropics.

Interrelation of the intraseasonal precipitation variability over Eurasia to the Eliassen-Palm flux characteristics over the Northern Hemisphere

Conditional averages of principal components of the Eliassen-Palm flux divergence variability are projected onto the daily precipitation amounts chart. The conditions of calculating the average values are determined by the dates classified in three equiprobable precipitation categories. The classification and calculation of characteristics are performed for the summer and winter seasons. Using the rotated principal component analysis, several regions are revealed of statistically significant interrelation between extreme precipitation and the first EP flux divergence variability modes, a simplified exploratory interpretation of interrelations is given and several recommendations are formulated for correcting seasonal forecasts of meteorological conditions with the use of results obtained.

Dynamical Control of the Middle Atmosphere

This review recapitulates the concept of the wave-driven residual circulation in the stratosphere and mesosphere. The residual circulation is defined as the conventional mean meridional circulation corrected by the quasi-linear Stokes drift due to atmospheric waves. Only when the zonal-mean primitive equations are transformed using the residual circulation, they reflect the causality arising from the Eliassen-Palm (EP) theorem. The EP theorem states that the proper wave-mean flow interaction, defined as the EP flux divergence, vanishes for waves that are linear, conservative, and steady. In the real atmosphere, this theorem is violated mainly due to wave breaking and turbulence. The resulting EP flux divergence then drives a residual circulation which causes the observed substantial deviations from some hypothetical radiatively determined state. With regard to this dynamical control we discuss the different contributions of Rossby waves and gravity waves. Recapitulation of Lindzen’s theory of gravity-wave saturation allows us to interpret various phenomena in the upper mesosphere such as interhemispheric coupling or modulations of the gravity-wave driven branch of the residual circulation by solar proton effects and thermal tides. In addition we discuss the relative importance of changes in radiative transfer and tropospheric gravity-wave sources on the long-term temperature trends in the summer mesosphere.

Simulation of changes in global ozone and atmospheric dynamics in the 21st century with the chemistry-climate model SOCOL

The solar climate ozone links (SOCOL) three-dimensional chemistry-climate model is used to estimate changes in the ozone and atmospheric dynamics over the 21st century. With this model, four numerical time-slice experiments were conducted for 1980, 2000, 2050, and 2100 conditions. Boundary conditions for sea-surface temperatures, sea-ice parameters, and concentrations of greenhouse and ozone-depleting gases were set following the IPCC A1B scenario and the WMO A1 scenario. From the model results, a statistically significant cooling of the model stratosphere was obtained to be 4–5 K for 2000–2050 and 3–5 K for 2050–2100. The temperature of the lower atmosphere increases by 2–3 K over the 21st century. Tropospheric heating significantly enhances the activity of planetary-scale waves at the tropopause. As a result, the Eliassen-Palm flux divergence considerable increases in the middle and upper stratosphere. The intensity of zonal circulation decreases and the meridional residual circulation increases, especially in the winter-spring period of each hemisphere. These dynamic changes, along with a decrease in the concentrations of ozone-depleting gases, result in a faster growth of O3 outside the tropics. For example, by 2050, the total ozone in the middle and high latitudes approaches its model level of 1980 and the ozone hole in Antarctica fills up. The superrecovery of the model ozone layer in the middle and high latitudes of both hemispheres occurs in 2100. The tropical ozone layer recovers far less slowly, reaching a 1980 level only by 2100.

Finite-Amplitude Wave Activity and Mean Flow Adjustments in the Atmospheric General Circulation. Part I: Quasigeostrophic Theory and Analysis

Abstract A diagnostic relationship between finite-amplitude wave activity and the associated adiabatic adjustments to the zonal-mean zonal wind and temperature is developed in the quasigeostrophic (QG) framework and is applied to a 23-yr segment (1979–2001) of the 40-yr ECMWF Re-Analysis (ERA-40) data. Wave activity is defined in terms of an instantaneous areal displacement of QG potential vorticity (PV) from zonal symmetry. Unlike previous forms, the tendency of wave activity equals exactly the negative of the eddy PV flux (Eliassen–Palm flux divergence) in the conservative limit, even at finite amplitude. This allows one to integrate the transformed Eulerian mean (TEM) theory in time and quantify the departure (adiabatic adjustment) of the zonal-mean state from an eddy-free reference state in terms of the observed wave activity. The structure of wave activity identifies synoptic eddies in the extratropics and planetary waves in the high latitudes of winter-to-spring stratosphere. In addition, a thin layer of high wave activity is found at the top of the lowermost stratosphere (∼17 km) in the summer extratropics. The reference state is constructed by “zonalizing” the PV contours conservatively (preserving area) on the isobaric surface and by inverting the resultant PV gradient for the mean flow. The adjustment associated with wave activity depends on the assumed surface boundary condition for the reference state. With a no-slip condition, the observed zonal-mean temperature is on average ∼33 (90) K higher than the reference state in the troposphere (stratosphere) of the Arctic winter, while the zonal-mean zonal wind is ∼30 m s−1 slower in the upper stratosphere. Since the reference state filters out the advective eddy–mean flow interaction, it fluctuates less than the zonal-mean state, potentially improving the signal-to-noise ratio for climate diagnosis.

Recovery of stratospheric ozone in calculations by the Center for Climate System Research/National Institute for Environmental Studies chemistry‐climate model under the CCMVal‐REF2 scenario and a no‐climate‐change run

Return times to 1980 values of ozone and the halogen concentrations in the future atmosphere were examined using the outputs of the Center for Climate System Research/National Institute for Environmental Studies chemistry‐climate model (CCM). The CCM calculation for the future atmosphere was performed for the period of 1975–2100 under a future scenario of time‐varying halogens, greenhouse gases, and sea surface temperatures for chemistry‐climate model validation (CCMVal‐REF2 scenario). A sensitivity test of a no‐climate‐change run for the future atmosphere was also performed, in which the concentrations of CO2, CH4, and N2O were fixed at those for 1975, and the sea surface temperature was fixed at that for the 1970–1979 mean, while the halogen concentrations were changed following the REF2 scenario. A comparison of the return time to 1980 values in these two runs was made. The return times of the halogen concentrations and column ozone in the extratropics for the REF2 run are earlier than those for the no‐climate‐change run, influenced by an enhancement of the meridional circulation and stratospheric cooling in the atmosphere. In the tropics, the column ozone of the REF2 run shows a second decrease after 2050, influenced by a stronger upward motion in the future atmosphere than in the present atmosphere. Trends in the zonal mean ozone concentration, zonal mean temperature, zonal mean zonal wind, Eliassen‐Palm (EP) flux, EP flux divergence, vertical component of the residual mean meridional circulation, and chemical forcing to the ozone concentration are examined in the periods of 2000–2050 and 2051–2100. These trends in the dynamical and chemical factors explain the earlier return times of halogens and ozone in the extratropics of the REF2 run than in the no‐climate‐change run. In the no‐climate‐change run, a better correlation between the return time of column ozone to the 1980 level and that of the lower stratospheric halogen concentration is calculated for the Southern Hemisphere corresponding to the facts that the ozone concentrations in the Southern Hemisphere lower stratosphere are more controlled by chlorine and bromine chemistries than those in the Northern Hemisphere.

The Roles of Equatorial Trapped Waves and Internal Inertia–Gravity Waves in Driving the Quasi-Biennial Oscillation. Part II: Three-Dimensional Distribution of Wave Forcing

Abstract Three-dimensional wave forcing of simulated quasi-biennial oscillation (QBO) is investigated using a high-resolution atmospheric general circulation model with T213L256 resolution (60-km horizontal and 300-m vertical resolution). In both the eastward and westward wind shear phases of the QBO, nearly all Eliassen–Palm flux (EP flux) divergence due to internal inertia–gravity waves (defined as fluctuations with zonal wavenumber ≥12) results from the divergence of the vertical component of the flux. On the other hand, EP flux divergence due to equatorial trapped waves (EQWs) results from both the meridional and vertical components of the flux in regions of strong vertical wind shear. Longitudinal dependence of wave forcing is also investigated by three-dimensional wave activity flux applicable to gravity waves. Near the top of the Walker circulation, strong eastward (westward) wave forcing occurs in the Eastern (Western) Hemisphere due to internal inertia–gravity waves with small horizontal phase speed. In the eastward wind shear zone associated with the QBO, the eastward wave forcing due to internal inertia–gravity waves in the Eastern Hemisphere is much larger than that in the Western Hemisphere, whereas in the westward wind shear zone, westward wave forcing does not vary much in the zonal direction. Zonal variation of wave forcing in the stratosphere results from (i) zonal variation of wave sources, (ii) the vertically sheared zonal winds associated with the Walker circulation, and (iii) the phase of the QBO.

Correction to “Nonlinearity of the combined warm ENSO and QBO effects on the Northern Hemisphere polar vortex in MAECHAM5 simulations”

Figure 3. Vertical profiles for the ENW–CTW (black) and ENE–CTE (gray) ensemble differences of the vertical (dashed) and meridional (dashed-dotted) components of the EP flux averaged from 60 N to 80 N and the Eliassen Palm (EP) flux divergence (solid) averaged from 50 N to 80 N. Values plotted are 10 Fy (N m ), 10 Fz (N m ), and 0.5 div(EP) in m s 1 d . JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 114, D20117, doi:10.1029/2009JD013257, 2009

Kelvin waves in stratosphere, mesosphere and lower thermosphere temperatures as observed by TIMED/SABER during 2002–2006

Abstract Temperature measurements from the SABER instrument on the TIMED spacecraft are used to elucidate the properties of Kelvin waves and other equatorial oscillations over the altitude range 20–120 km during 2002–2006. The dominant Kelvin waves transition from long periods (52–10 days) and short wavelengths (9–13 km) in the stratosphere, to shorter periods (2–3 days) and longer wavelengths (35–45 km) in the 80–120 km height region. Ultra-Fast Kelvin Waves (UFKW) with periods of 2.5–4.5 days intermittently exist at amplitudes of order 3–10 K between 80–120 km during all months of the year, with variability at periods typically in the 20–60 day range. An Intra-seasonal oscillation (ISO) of zonal mean temperatures also exists with periods 20–60 days that may be driven by Eliassen-Palm Flux Divergences (EPFD) due, at least in part, to UFKW and migrating diurnal tides.

Diagnostic Analysis of Tidal Winds and the Eliassen–Palm Flux Divergence in the Mesosphere and Lower Thermosphere from TIMED/SABER Temperatures

Abstract For migrating tides or fast-moving planetary waves, polarization relations derived from the linear wave equations are required to accurately derive the wind components from the temperature field. A common problem in diagnosing winds from the measured temperature is the error amplification associated with apparent singularities in the wave polarization relations. The authors have developed a spectral module that accurately derives tidal winds from the measured tidal temperature field and effectively eliminates the error amplification near the apparent singularities. The algorithm is used to perform a diagnostic analysis of tidal winds and the Eliassen–Palm (EP) flux divergence in the mesosphere and lower thermosphere (MLT) based on the zonal mean and tidal temperature fields derived from 6 yr of temperature measurements made by the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) instrument onboard the Thermosphere–Ionosphere–Mesosphere Energetics and Dynamics (TIMED) satellite. The derived zonal mean wind and diurnal tidal amplitude reveal new insights into the mesospheric biennial oscillation (MBO) that exists in the MLT at both equatorial and midlatitude regions. The equatorial MBO in the zonal mean wind is present in the entire mesosphere from 50 to 90 km. The equatorial MBO in the temperature amplitude of the diurnal tide occurs near the mesopause region between 80 and 90 km and is largely coincident with the downward phase propagation of the equatorial MBO in the zonal mean wind, indicating a possible mechanism of wave–mean flow interaction between the two. On the other hand, the newly discovered midlatitude MBOs in zonal mean wind and the meridional wind in diurnal tide occur at different altitudes, suggesting possibly a remote forcing–response relationship. The acceleration or deceleration of the zonal mean wind due to EP flux divergence that is contributed by the migrating tides peaks at midlatitudes with a typical value of 10–20 m s−1 day−1 around 95 km.

The Angular Momentum Budget of the Transformed Eulerian Mean Equations

The axial angular momentum (AAM) budget of zonal atmospheric annuli extending from the surface to a given height and over meridional belts is discussed within the framework of conventional and transformed Eulerian mean (TEM) theory. Conventionally, it is only fluxes of AAM through the boundaries and/or torques at the surface that are able to change the AAM of an annulus. TEM theory introduces new torques in the budget related to the vertically integrated Eliassen–Palm flux divergence and also new AAM fluxes of the residual difference circulation. Some of these torques are displayed for various annuli. In particular, the application of TEM theory generates a large positive torque at tropospheric upper boundaries in the global case. This torque is much larger than the global mountain and friction torques but is cancelled exactly by the new vertical AAM fluxes through the upper boundary. It is concluded that the TEM approach complicates the analysis of AAM budgets but does not provide additional insight. Isentropic pressure torques are believed to be similar to the TEM torques at the upper boundary of an annulus. The isentropic pressure torques are evaluated from data and found to differ in several respects from the TEM torques.

Forcing mechanism of the seasonally asymmetric quasi‐biennial oscillation secondary circulation in ERA‐40 and MAECHAM5

The seasonality of the quasi‐biennial oscillation (QBO) and its secondary circulation is analyzed in the European Reanalysis (ERA‐40) and Middle Atmosphere European Centre Hamburg Model (MAECHAM5) general circulation model data sets through the multitaper method‐singular value decomposition (MTM‐SVD). In agreement with previous studies, the results reveal a strong seasonal dependence of the QBO secondary circulation. This is characterized by a two‐cell structure symmetric about the equator during autumn and spring. However, anomalies strongly weaken in the summer hemisphere and strengthen in the winter hemisphere, leading to an asymmetric QBO secondary circulation characterized by a single‐cell structure displaced into the winter hemisphere during the solstices. In ERA‐40, this asymmetry is more pronounced during the northern than during the southern winter. These results provide the first observation of the QBO secondary circulation asymmetries in the ERA‐40 reanalysis data set across the full stratosphere and the lower mesosphere, up to 0.1 hPa. The MTM‐SVD reconstruction of the seasonal QBO signals in the residual circulation and the QBO signals in Eliassen Palm (EP) flux divergences suggest a particular mechanism for the seasonal asymmetries of the QBO secondary circulation and its extension across the midlatitudes. The analysis shows that the QBO modulates the EP flux in the winter hemispheric surf zone poleward of the QBO jets. The zonal wind forcing by EP flux divergence is transformed by the Coriolis effect into a meridional wind signal. The seasonality in the stratospheric EP flux and the hemispheric differences in planetary wave forcing cause the observed seasonality in the QBO secondary circulation and its hemispheric differences.

The effect of the 11-year solar cycle on the temperature in the lower stratosphere

Two temperature datasets are analyzed for quantifying the 11-year solar cycle effect in the lower stratosphere. The analysis is based on a regression linear model that takes into account volcanic, Arctic Oscillation (AO), Quasi-Biennial Oscillation (QBO) and El Nino Southern Oscillation (ENSO) effects. Under solar maximum conditions, temperatures are generally warmer for low- and mid-latitudes than under solar minimum, with the effect being the strongest in northern summer. At high latitudes, the vortex is generally stronger under solar maximum conditions, with the exception of February and to a lesser extent March in the Northern Hemisphere; associated with this positive signal at high latitudes, there is a significant negative signal at the equator. Observations also suggest that contrary to the beginning of the winter, in February–March, the residual circulation in the Northern Hemisphere is enhanced. A better understanding of the mechanisms at work comes from further investigations using the ERA-40 reanalysis dataset. First, a consistent response in terms of temperature and wind is obtained. Moreover, considering Eliassen-Palm (EP) flux divergence and residual circulation stream functions, we found an increased circulation in the Northern Hemisphere in February during solar maxima, which results in more adiabatic warming at high latitudes and more adiabatic cooling at low latitudes, thus demonstrating the dynamical origin of the response of the low stratosphere to the solar cycle.

Multiple Equilibrium States Appearing in a Venus-Like Atmospheric General Circulation Model

We investigated the existence of multiple equilibrium states in the Venusian atmospheric general circulation suggested by Matsuda (1980, 1982) using a Venus-like atmospheric general circulation model. We ran the model from two initial conditions: a state with large zonal wind increasing with height and a motionless state. For the large zonal wind initial state, a strong zonal wind with weak meridional circulation, i.e., super-rotation, appears. However for the motionless initial condition, slow zonal wind with strong meridional circulation relative to the farmer state appears. Each circulation reached a quasi-steady state. These results have the same features suggested by Matsuda. The presence of multiple equilibrium states was sensitive to the horizontal eddy viscosity parameter. For the strong zonal wind state, the acceleration of the zonal mean zonal wind results from the horizontal EP flux divergence from the wavenumber one component on the equator, which is mainly maintained by the Gierasch mechanism (1975). The presence of multiple equilibrium states suggests that an alternative slow zonal wind state could appear in the Venusian atmosphere if an appropriate initial condition or drastic fluctuation is assigned.

Simulations of anthropogenic change in the strength of the Brewer–Dobson circulation

The effect of climate change on the Brewer–Dobson circulation and, in particular, the large-scale seasonal-mean transport between the troposphere and stratosphere is compared in a number of middle atmosphere general circulation models. All the models reproduce the observed upwelling across the tropical tropopause balanced by downwelling in the extra tropics, though the seasonal cycle in upwelling in some models is more semi-annual than annual. All the models also consistently predict an increase in the mass exchange rate in response to growing greenhouse gas concentrations, irrespective of whether or not the model includes interactive ozone chemistry. The mean trend is 11 kt s−1 year−1 or about 2% per decade but varies considerably between models. In all but one of the models the increase in mass exchange occurs throughout the year though, generally, the trend is larger during the boreal winter. On average, more than 60% of the mean mass fluxes can be explained by the EP-flux divergence using the downward control principle. Trends in the annual mean mass fluxes derived from the EP-flux divergence also explain about 60% of the trend in the troposphere-to-stratosphere mass exchange rate when averaged over all the models. Apart from two models the interannual variability in the downward control derived and actual mass fluxes were generally well correlated, for the annual mean.

Temporal variation of nonmigrating diurnal tide and its relation with the moist convective activity

By using a general circulation model, behaviors of the eastward propagating diurnal tide with zonal wavenumber 3 (DE3) and their relation with the convective activity in the tropical troposphere are examined. The amplitude of the DE3 is significant in the equatorial mesosphere and lower thermosphere (MLT). Day‐to‐day variations of the DE3 amplitude are evident in the MLT region, and are closely related with temporal variations of the rainfall rate in the tropics. Estimates of the EP flux divergence due to the DE3 indicate that this convective modulated DE3 activity induces day‐to‐day variations of the wave‐induced driving of the zonal mean zonal wind in the MLT region. Our results show that temporal variations of the zonal mean zonal wind in the equatorial MLT region are influenced by the moist convective activity in the tropics.

Tropical Wave Driving of the Annual Cycle in Tropical Tropopause Temperatures. Part I: ECMWF Analyses

Abstract A quantitative examination of the annual cycle in the tropical tropopause temperatures, tropical ascent, momentum balance, and wave driving is performed using ECMWF analyses to determine how the annual cycle in tropical tropopause temperatures arises. Results show that the annual cycle in tropical tropopause temperatures is driven by the annual variation in ascent and consequent dynamical (adiabatic) cooling at the tropical tropopause. Mass divergence local to the tropical tropopause has the dominant contribution to ascent near the tropical tropopause. The annual cycle in mass divergence, and the associated meridional flow, near the tropical tropopause is driven by Eliassen–Palm (EP) flux divergence, that is, wave dissipation. The EP flux divergence near the tropical tropopause is dominated by stationary waves with both the horizontal and vertical components of the EP flux contributing. However, the largest annual cycle is in the divergence of the vertical EP flux and in particular from the contribution in the vertical flux of zonal momentum. These results do not match the existing theory that the annual cycle is driven by the wave dissipation in the extratropical stratosphere, that is, the stratospheric pump. It is suggested that the annual cycle is linked to equatorial Rossby waves forced by convective heating in the tropical troposphere.

Tropical Wave Driving of the Annual Cycle in Tropical Tropopause Temperatures. Part II: Model Results

Abstract The atmospheric response to a localized distribution of tropical heating is examined in terms of the stationary waves excited and how these impact the mean flow near the tropical tropopause. This is done by examining nonlinear simulations of the Gill model with a primitive equation model that extends from the surface up into the stratosphere. The model produces strong cooling of zonal mean temperatures near the tropical tropopause when the heating is on the equator but weaker cooling with the heating at 15°N. The model shows that equatorial Rossby waves that penetrate the lower stratosphere and changes in EP flux divergence that correspond to the observed changes between December and August. It is suggested that ascent in the upper tropical troposphere is driven by vorticity advection or equivalently potential vorticity fluxes due to these equatorial Rossby waves, particularly when the heating is close to the equator. The model results provide support to the hypothesis that the annual cycle in tropical tropopause temperatures is a result of the annual variation in latitude of tropical heating and that equatorial Rossby waves are key in producing the response in the upper troposphere and lower stratosphere.

On the forcing of the Mesospheric Semi‐Annual Oscillation in the Whole Atmosphere Community Climate Model

The Mesospheric Semi‐Annual Oscillation in the mean zonal winds (MSAO) is a prominent feature of the tropical middle atmosphere. This oscillation is thought to be driven by the selective transmission of small‐scale gravity waves through the wind system of the stratospheric Semi‐Annual Oscillation (SSAO). However, due to limited observations, the driving mechanisms of this oscillation are not completely understood. Here we examine the driving of the MSAO using the Whole Atmosphere Community Climate Model, version 2 (WACCM2). We show that near the solstices, in addition to gravity waves, meridional advection and EP flux divergence from resolved waves (in particular from the quasi two‐day wave) contribute significantly to the momentum budget of the MSAO. Near the equinoxes, all forcing terms are weak suggesting that the easterly phase of the MSAO is driven by a process not well represented in WACCM2.