Stratospheric Semiannual Oscillation Research Articles

Infrasound Thunder Detections across 15 Years over Ivory Coast: Localization, Propagation, and Link with the Stratospheric Semi-Annual Oscillation

Every day, about one thousand thunderstorms occur around the world, producing about 45 lightning flashes per second. One prominent infrasound station of the International Monitoring System infrasound network of the Comprehensive Nuclear-Test-Ban Treaty Organization for studying lightning activity is in Ivory Coast, where the lightning rate of this region is relatively high. Infrasound defines acoustic waves with frequencies below 20 Hz, the lower limit of human hearing. Statistical results are presented in this paper based on infrasound measurements from 2004 to 2019. One-to-one association between infrasound detections from 0.5 to 5 Hz and lightning flashes detected by the World Wide Lightning Location Network within 500 km from the infrasound station is systematically investigated. Most of the infrasound signals detected at IS17 in this frequency band are due to thunder, even if the thunderstorms are located up to 500 km away from the station. A decay of the thunder amplitude with the flash distance, d, is found to scale as d−0.717 for flashes within 100 km from the station, which holds for direct propagation. Interestingly, the stratospheric detections reflect a pattern in the annual azimuth variation, which is consistent with the equatorial stratospheric semi-annual oscillation.

The equatorial stratospheric semiannual oscillation and time‐mean winds in QBOi models

AbstractThe Quasi‐Biennial Oscillation initiative (QBOi) is a model intercomparison programme that specifically targets simulation of the QBO in current global climate models. Eleven of the models or model versions that participated in a QBOi intercomparison study have upper boundaries in or above the mesosphere and therefore simulate the region where the stratopause semiannual oscillation (SAO) is the dominant mode of variability of zonal winds in the tropical upper stratosphere. Comparisons of the SAO simulations in these models are presented here. These show that the model simulations of the amplitudes and phases of the SAO in zonal‐mean zonal wind near the stratopause agree well with the information derived from available observations. However, most of the models simulate time‐average zonal winds that are more westward than determined from observations, in some cases by several tens of m·s–1. Validation of wave activity in the models is hampered by the limited observations of tropical waves in the upper stratosphere but suggests a deficit of eastward forcing either by large‐scale waves, such as Kelvin waves, or by gravity waves.

Effects of the Quasi-Biennial Oscillation and Stratospheric Semiannual Oscillation on Tracer Transport in the Upper Stratosphere

Abstract Using satellite observations together with a chemistry–climate model (CCM), the effect of the stratospheric semiannual oscillation (SAO) and quasi-biennial oscillation (QBO) on the equatorial double peak in observed CH4 and NO2 is reexamined. It is concluded that the lower-equatorial Halogen Occultation Experiment (HALOE) CH4 mixing ratio of the April double peak in 1993 and 1995 was associated with the prominent first cycle of the SAO westerlies, which causes local vertical downwelling in the upper equatorial stratosphere. The observational evidences imply that the strong westerlies of the first cycle of the stratospheric SAO in 1993 and 1995 were driven by enhanced lower-stratospheric gravity wave activity in the early parts of those years. The CCM simulations further verify that the gravity wave source strength has a large impact on the development and strength of the SAO westerlies. This result suggests that the equatorial long-lived tracer mixing ratio near the stratopause (which is associated with the strength of the SAO westerlies) was not only modulated by the QBO phase, but was also significantly influenced by interannual variation in the gravity waves. It is also found that the deeper equatorial trough of the double peak is unlikely to be always accompanied by the more prominent Northern Hemispheric lobe, and the Northern Hemispheric lobe of the double peak can be mainly attributed to subtropical upwelling. The altitude of greatest chemical destruction anomalies associated with the SAO and QBO is below the trough of the double peak, implying that the effect of the chemical process on the double peak is insignificant.

Meteor‐radar observed mesospheric semi‐annual oscillation (SAO) and quasi‐biennial oscillation (QBO) over Maui, Hawaii

We observed the variations of the semi‐annual oscillation (SAO) in the mesopause region of northern subtropics, and we describe its origin and forcing implications. Using the measurements of horizontal wind profiles made by the University of Illinois meteor radar in Maui, Hawaii (20.7°N, 156.3°W) from May 2002 to June 2007 and the European Centre for Medium‐Range Weather Forecasts (ECMWF) interim data set, we find that the mesospheric SAO, with the winter westerly near 80–90 km clearly stronger than the summer westerly, is out of phase with the stratospheric SAO near 1 hPa (∼50 km). The mesospheric SAO easterly is strong during the easterly phase and weak during the westerly phase of the stratospheric quasi‐biennial oscillation (QBO) near 10 hPa (∼30 km), suggesting the modulation of the mesospheric SAO by the stratospheric QBO. The mesospheric QBO has an amplitude of approximately 5 m/s near 80 km. It is in phase with the stratospheric QBO near 10 hPa and out of phase with the QBO‐like oscillation near 1 hPa. The correlation of the gravity wave (GW) and the quasi‐two‐day wave (QTDW) activities with the mesospheric SAO and QBO suggests that the GW drags and the QTDW Eliassen‐Palm flux divergences likely contribute to the QBO modulation of the mesospheric SAO. The winter easterly wind in the tropical upper stratosphere pushes further into the northern subtropics during the QBO westerly phase than during the easterly phase. This may have impacts on the upward propagation of westward‐propagating GWs originated from the middle latitudes, and thus the westward GW forcing in the upper mesosphere of northern subtropics.

Meteor radar observations of short-term variability of quasi 2 day waves and their interaction with tides and planetary waves in the mesosphere–lower thermosphere region over Thumba (8.5°N, 77°E)

[1] Meteor radar measurements of zonal and meridional winds over a low-latitude station, Thumba (8.5°N, 77°E), during March 2006 to March 2009 are used to study the quasi 2 day wave structure in the mesosphere–lower thermosphere (MLT) region. Emphasis is placed on the study of the day-to-day variations of quasi 2 day waves and their interaction with diurnal tides and planetary waves. The seasonal variation of quasi 2 day waves showed two maxima, one during summer and the other during winter, coinciding with the strong eastward winds in the MLT region along with a distinct peak during the month of October. Subsequent analysis of TIMED/SABER observations during October revealed the presence of a westward propagating quasi 2 day wave with zonal wave number 3 over the present latitude. The seasonal variation of quasi 2 day wave is also discussed with respect to the Stratospheric Semiannual Oscillation (SSAO), and we found that the quasi 2 day wave activity peaks in the MLT region during the phase transition of SSAO as well as during the eastward phase. Wavelet and subsequent bispectral analysis confirmed the wave-wave interactions of quasi 2 day wave and planetary waves with periods of 5–10 and 15–20 days. The 16 h oscillation, which is believed to be generated by interaction of quasi 2 day wave and diurnal tide, showed an excellent correlation with quasi 2 day wave activity. Apart from reporting day-to-day variation of quasi 2 day wave for the first time over this latitude, there are two new results from the present study: (1) a consistent peak in quasi 2 day activity during October, which has never been reported and has been observed for the first time over low latitudes, and (2) the relation between SSAO and quasi 2 day wave activity.

QBO modulation of the semiannual oscillation in MAECHAM5 and HAMMONIA

The quasi‐biennial oscillation (QBO) signature in the equatorial upper stratosphere and mesosphere is analyzed from MAECHAM5 and HAMMONIA general circulation models. Our results show that this region is significantly influenced by the stratospheric QBO. In the upper stratosphere the QBO modulates the altitude of maximum descent of the stratospheric semiannual oscillation (SSAO) westerly phases. Our results also suggest that the QBO modulates the altitude of maximum descent and also the strength of the SSAO easterly phase. We explore the role of large‐scale and small‐scale waves and also momentum advection in the forcing of the QBO signature in the SSAO domain. The results show how the vertical propagation of the QBO signature to the middle and upper mesosphere depends on the vertical phase structure of the SAO and consequently on the seasonal cycle. During the solstices when MSAO westerlies prevail in the middle and upper mesosphere no QBO signature can be detected above the stratopause region. However, during the equinoxes, when MSAO easterlies dominate in the middle and upper mesosphere, the QBO signature extends throughout the mesosphere and low thermosphere. The QBO directly modulates MSAO easterlies by modifying the altitude at which they are generated in the upper mesosphere. A QBO signature is also detected on the MSAO westerly phase occurring in the mesopause region during the equinoxes.

Seasonal and interannual variability of gravity wave activity revealed by long‐term lidar observations over Mauna Loa Observatory, Hawaii

The seasonal and interannual variability of gravity wave (GW) variance in the upper stratosphere (35–50 km) and lower mesosphere (48–63 km) has been studied using 10.5 years (January 1997 to June 2007) of temperature profile results obtained with the Jet Propulsion Laboratory Rayleigh lidar at Mauna Loa Observatory, Hawaii (19.5°N, 155.6°W). Seasonal variability with a maximum in winter and a minimum in summer was observed in the upper stratosphere, suggesting dominance of the annual oscillation. In the lower mesosphere the seasonal oscillations of GW variance were dominated by a semiannual oscillation (SAO), likely due to the selective filtering of GWs by the tropical upper stratospheric SAO wind. Modulation of GW variance by the quasi‐biennial oscillation was clearly present only for the long vertical wavelength band in the upper stratosphere, and not in the lower mesosphere. The United Kingdom Met Office zonal mean zonal wind further supports that enhanced GW activity in the upper stratosphere corresponds to the westerly shear phase of the zonal wind at 10 hPa (∼30 km), and suppressed activity corresponds to the easterly shear phase. During the strong El Niño event in the winter of 1997–1998, enhanced GW activity was observed only in the lower mesosphere, and not in the upper stratosphere. Additional enhancement of GW variance, especially clear in the upper stratosphere, was also found during 2001–2002 and winter 2005–2006.

Momentum Flux Spectrum of Convectively Forced Internal Gravity Waves and Its Application to Gravity Wave Drag Parameterization. Part II: Impacts in a GCM (WACCM)

Abstract Impacts of a spectral parameterization of gravity wave drag (GWD) induced by cumulus convection (GWDC) in the NCAR Whole Atmosphere Community Climate Model (WACCM1b) are investigated. In the spectral GWDC parameterization, reference wave momentum flux spectrum is launched at cloud top and analytically calculated based on the physical properties of convection and the large-scale flow. The cloud-top wave momentum flux is strong mainly in the Tropics and midlatitude storm-track regions, and exhibits anisotropy and spatiotemporal variability. The anisotropy and variability are determined by the distributions and variations of convective activities, the moving speed of convection, and horizontal wind and stability in convection regions. Zonal-mean zonal GWDC has a maximum of 13–27 (37–50) m s−1 day−1 in the mesosphere in January (July). Impacts of GWDC on zonal wind appear mainly in the low to midlatitudes of the upper stratosphere and mesosphere. In these regions, biases of zonal wind with respect to observation are reduced more than 50% through the GWDC process. In contrast to zonal wind, impacts of GWDC on temperature occur mainly in the mid- to high latitudes. Through the analysis of forcing terms in the zonal wind and temperature equations, it is found that impacts of GWDC result from interaction among wave forcing terms (resolved wave forcing, parameterized background GWD, and GWDC) and meridional circulations induced by the wave forcing terms. With regard to tropical variability, when GWDC is included, the model produces the stratospheric semiannual oscillation with more realistic amplitude and structure and stronger interannual variabilities in the lower stratosphere. These enhanced variabilities are caused by resolved wave forcing and meridional circulations.

A quantitative study on the role of gravity waves in driving the tropical Stratospheric Semiannual Oscillation

Using a coordinated experimental observation under Middle Atmospheric Dynamics (MIDAS) program, an extensive study is carried out to quantify the role of gravity waves in driving the tropical Stratospheric Semiannual Oscillation (SSAO). Rayleigh lidar observations of middle atmospheric temperature over Gadanki (13.5°N, 79.2°E) and rocketsonde wind measurements over Trivandrum (8.5°N, 76.9°E) during the period November 2002 to June 2005 are used for the present study. Gravity waves with periods ranging from 2–4 hours and 0.5–1 hour are found to be dominant in the middle atmosphere throughout the observational period. The altitude profiles of momentum fluxes of gravity waves having these time periods are estimated and their seasonal variations are studied, which showed semiannual variation with maximum around equinoxes and minimum around solstitial months. The mean flow acceleration estimated from the divergence of momentum flux of gravity waves is compared with the mean flow acceleration observed using rocket measured zonal winds during three different cycles of SSAO. This comparison provided an opportunity to quantify the contribution of gravity waves toward generation of SSAO, which is found to be ∼30–50% of the observed acceleration during the evolution of the westerly phase of the SSAO. The present observations showed that the contribution of the gravity waves toward the westerly phase of SSAO varies significantly from cycle to cycle. The significance of the present results lies in quantifying the gravity wave–mean flow interaction during both easterly and westerly phases of SSAO for the first time over this tropical latitude.

Gravity waves observed from the Equatorial Wave Studies (EWS) campaign during 1999 and 2000 and their role in the generation of stratospheric semiannual oscillations

Abstract. The altitude profiles of temperature fluctuations in the stratosphere and mesosphere observed with the Rayleigh Lidar at Gadanki (13.5° N, 79.2° E) on 30 nights during January to March 1999 and 21 nights during February to April 2000 were analysed to bring out the temporal and vertical propagation characteristics of gravity wave perturbations. The gravity wave perturbations showed periodicities in the 0.5–3-h range and attained large amplitudes (4–5 K) in the mesosphere. The phase propagation characteristics of gravity waves with different periods showed upward wave propagation with a vertical wavelength of 5–7 km. The mean flow acceleration computed from the divergence of momentum flux of gravity waves is compared with that calculated from monthly values of zonal wind obtained from RH-200 rockets flights. Thus, the contribution of gravity waves towards the generation of Stratospheric Semi Annual Oscillation (SSAO) is estimated.

Stratospheric and mesospheric temperature variations for the quasi-biennial and semiannual (QBO and SAO) oscillations based on measurements from SABER (TIMED) and MLS (UARS)

Abstract. We present the zonal mean temperature variations for the quasi-biennial oscillation (QBO) and the semiannual oscillation (SAO) based on data from SABER on the TIMED spacecraft (years 2002 to 2004) and from MLS on the UARS mission (1992 to 1994). The SABER measurements provide the rare opportunity to analyze data from one instrument over a wide altitude range (15 to 95 km), while MLS data were taken in the 16 to 55 km altitude range a decade earlier. The results are presented for latitudes from 48° S to 48° N. New results are obtained for the QBO, especially in the upper stratosphere and mesosphere, and at mid-latitudes. At Equatorial latitudes, the QBO amplitudes show local peaks, albeit small, that occur at different altitudes. From about 20 to 40 km, and within about 15° of the Equator, the amplitudes can approach 3.5° K for the stratospheric QBO (SQBO). For the mesospheric QBO (MQBO), we find peaks near 70 km, with temperature amplitudes reaching 3.5° K, and near 85 km, the amplitudes approach 2.5° K. Morphologically, the amplitude and phase variations derived from the SABER and MLS measurements are in qualitative agreement. As a function of latitude, the QBO amplitudes tend to peak at the Equator but then increase again pole-ward of about 15° to 20°. The phase progression with altitude varies more gradually at the Equator than at mid-latitudes. Many of the SAO results presented are also new, in part because measurements were not previously available or were more limited in nature. At lower altitudes near 45 km, within about 15° of the Equator, the temperature amplitudes for the stratospheric SAO (SSAO) reveal a local maximum of about 5° K. At higher altitudes close to the Equator, our results show separate peaks of about 7° K near 75 and 90 km for the mesospheric SAO (MSAO). In the SAO results, significant inter-annual differences are evident, with the amplitudes being largest in 2002 relative to 2003 and 2004. As in the case for the QBO, the SAO temperature amplitudes go through minima away from the Equator, and then increase towards mid latitudes, especially at altitudes above 55 km. We compare our findings with previously published empirical results, and with corresponding results from the numerical spectral model (NSM). Although not a focus of this study, we also show results for the inter-annual variations (which appear to be generated at least in part by the QBO) of the migrating diurnal tide. In the upper mesosphere, their amplitudes can approach 20° K, and they are derived jointly with the zonal-mean components.

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.

The semi-annual oscillation in upper stratospheric and mesospheric water vapour as observed by HALOE

In this paper observations made by the Halogen Occultation Experiment (HALOE) instrument are used to investigate the semi-annual oscillation (SAO) in water vapour mixing ratio in the upper stratosphere. We also present first observations of an SAO in water vapour in the mid to upper equatorial mesosphere. The HALOE data set, which spans a period of greater than four years, also facilitates the examination of interannual variability, and in particular the effects of quasi-biennial oscillation (QBO)/SAO interaction on the water vapour distribution. The data show an SAO in equatorial water vapour near 1 mb, with troughs in February and August and peaks in May and November. This SAO travels downwards in time to around the 5 mb level, approximately in phase with contemporaneous zonal winds calculated from United Kingdom Meteorological Office (UKMO) analyses. In addition, the UKMO winds show a seasonal variation in the strength of the SAO, related to which is an annual cycle in HALOE water vapour near 1 mb. There is also an SAO in the upper mesosphere, which has a maximum amplitude approximately 2 1/2 times larger than that in the upper stratosphere. This SAO propagates downwards in time, approximately in phase with the zonal winds observed by the High-Resolution Doppler Imager instrument. However, the pattern of the water vapour maxima between 0.04 and 0.1 mb does not appear to be consistent with the vertical transport that is induced by the SAO in zonal wind. There is also an annual cycle in mesospheric water vapour, which results from the modulation of the winds in the mesosphere by the seasonal cycle in the upper stratospheric SAO in zonal wind. Latitude/pressure sections show that the amplitude of both the upper stratospheric and mesospheric SAO in water vapour decreases away from the equator. However, between 30 and 35° S the SAO in the mesosphere is again large, possibly due to upward transport in winter associated with breaking gravity waves. There are also considerable bieninal cycles in upper stratospheric and mesospheric water vapour, which appear to result chiefly from the phase of the QBO modulating the SAO in zonal wind.

The semi‐annual oscillation in upper stratospheric and mesospheric water vapour as observed by HALOE

AbstractIn this paper observations made by the Halogen Occultation Experiment (HALOE) instrument are used to investigate the semi‐annual oscillation (SAO) in water vapour mixing ratio in the upper stratosphere. We also present first observations of an SAO in water vapour in the mid to upper equatorial mesosphere. The HALOE data set, which spans a period of greater than four years, also facilitates the examination of interannual variability, and in particular the effects of quasi‐biennial oscillation (QBO)/SAO interaction on the water vapour distribution. The data show an SAO in equatorial water vapour near 1 mb, with troughs in February and August and peaks in May and November. This SAO travels downwards in time to around the 5 mb level, approximately in phase with contemporaneous zonal winds calculated from United Kingdom Meteorological Office (UKMO) analyses. In addition, the UKMO winds show a seasonal variation in the strength of the SAO, related to which is an annual cycle in HALOE water vapour near 1 mb. There is also an SAO in the upper mesosphere, which has a maximum amplitude approximately 2 1/2 times larger than that in the upper stratosphere. This SAO propagates downwards in time, approximately in phase with the zonal winds observed by the High‐Resolution Doppler Imager instrument. However, the pattern of the water vapour maxima between 0.04 and 0.1 mb does not appear to be consistent with the vertical transport that is induced by the SAO in zonal wind. There is also an annual cycle in mesospheric water vapour, which results from the modulation of the winds in the mesosphere by the seasonal cycle in the upper stratospheric SAO in zonal wind. Latitude/pressure sections show that the amplitude of both the upper stratospheric and mesospheric SAO in water vapour decreases away from the equator. However, between 30 and 35° S the SAO in the mesosphere is again large, possibly due to upward transport in winter associated with breaking gravity waves. There are also considerable bieninal cycles in upper stratospheric and mesospheric water vapour, which appear to result chiefly from the phase of the QBO modulating the SAO in zonal wind.

Climatology of the semiannual oscillation of the tropical middle atmosphere

We have used a variety of satellite, ground‐based, and in situ observations to construct a climatology of the semiannual oscillation (SAO) of the tropical middle atmosphere. The sources of data include rocketsonde observations of winds and temperature, MF radar wind observations, and observations of winds and temperatures made from space by the High Resolution Doppler Imager (HRDI) and the Solar Mesosphere Explorer (SME). These data sets provide a generally consistent picture of the SAO, of the relationship between its stratospheric and mesospheric manifestations, and of its apparent modulation by the stratospheric quasi‐biennial oscillation (QBO). In agreement with earlier studies, we find that the first cycle of the stratospheric SAO (which begins with the stratopause easterly phase in northern winter) is stronger than the second cycle (beginning with the easterly phase in southern winter). Similar behavior is apparent in the mesosphere, where the easterly phase is stronger during the first cycle than during the second cycle. HRDI and MF radar are capable of observing the seasonal cycle well into the lower thermosphere. Data from these two sources indicate that a strong SAO is present up to about 90 km, giving way above this altitude to time mean easterly winds with a weak semiannual variation. Between 105 and 110 km, HRDI data indicate the presence of a westerly wind layer with almost no seasonal variation. Apparent modulation of the stratospheric SAO by the QBO is found in rocketsonde data, while HRDI and MF radar observations suggest a correlation between the QBO and the easterly phase of the mesospheric SAO. We discuss the implications of these observations for the wave processes that drive the SAO.

The tropical stratopause in the UKMO stratospheric analyses: Evidence for a 2-day wave and inertial circulations

Large-scale atmospheric motions near the low-latitude stratopause in the UK Meteorological Office stratospheric analyses during the northern hemisphere winter are investigated, focusing in particular on the 2-day wave and inertial eddies. The 2-day wave is a westward-propagating planetary wave which appears recurrently in the summer subtropical upper stratosphere and mesosphere. Space-time spectral analysis of the assimilated fields of temperature and winds, and high-resolution transport of water vapour, reveal large-amplitude breaking planetary waves in the summer subtropics, principally a 2-day wave 3, and a wave 1 of a period near 7 days. A region of anomalous zonal-mean meridional gradient of potential vorticity also exists, pointing to a barotropic instability mechanism for the origin of the 2-day wave. By contrast, the winter subtropics are characterized by mixing on less-than-planetary scales during the period considered, occurring during the easterly phase of the stratospheric semi-annual oscillation. Water vapour is mixed turbulently into gyres, and drawn in narrow filaments out of the southern hemisphere. The potential-vorticity field displays vertically layered, zonally-asymmetric variations. The behaviour of the model assimilated fields is supported by along-track water vapour distribution observed by the Microwave Limb Sounder aboard the Upper Atmosphere Research Satellite which shows evidences for intermittent filamentary mixing in the northern subtropics, and a 2-day wave signature in the summer subtropics. Based on these results, a possible mechanism connecting the initiation of the austral 2-day wave events to strong planetary-wave activity in the winter is proposed. The winter planetary waves provide the background conditions for development of inertial instabilities at low latitudes of the winter hemisphere, which in turn create conditions favourable for barotropic instability of the summer subtropical easterly jet.

The annual cycle of stratospheric water vapor in a general circulation model

The application of general circulation models (GCMs) to stratospheric chemistry and transport both permits and requires a thorough investigation of stratospheric water vapor. The National Center for Atmospheric Research has redesigned its GCM, the Community Climate Model (CCM2), to enable studies of the chemistry and transport of tracers including water vapor; the importance of water vapor to the climate and chemistry of the stratosphere requires that it be better understood in the atmosphere and well represented in the model. In this study, methane is carried as a tracer and converted to water; this simple chemistry provides an adequate representation of the upper stratospheric water vapor source. The cold temperature bias in the winter polar stratosphere, which the CCM2 shares with other GCMs, produces excessive dehydration in the southern hemisphere, but this dry bias can be ameliorated by setting a minimum vapor pressure. The CCM2's water vapor distribution and seasonality compare favorably with observations in many respects, though seasonal variations including the upper stratospheric semiannual oscillation are generally too small. Southern polar dehydration affects midlatitude water vapor mixing ratios by a few tenths of a part per million, mostly after the demise of the vortex. The annual cycle of water vapor in the tropical and northern midlatitude lower stratosphere is dominated by drying at the tropical tropopause. Water vapor has a longer adjustment time than methane and had not reached equilibrium at the end of the 9 years simulated here.