SPARC logo (45 Ko)
S P A R C

Stratospheric Processes And their Role in Climate
A project of the World Climate Research Programme

Home Initiatives Organisation Publications Meetings Acronyms and Abbreviations Useful Links

 

22nd IUGG General Assembly, Birmingham, UK, 18-30 July 1999

Sessions relevant to SPARC

MC06. Quasi-biennial Oscillation and Internal Gravity Waves

Convener: Isamu Hirota. Report by Mark Baldwin (mark@nwra.com)

This session on the quasi-biennial oscillation (QBO) was held over an entire day, with two invited review talks and nine contributed presentations, covering the basics of equatorial dynamics, and extratropical dynamical effects, observations using rawinsondes, observations of QBO modulation of trace gases, modelling of the QBO, and modelling of its extratropical effects on circulation and chemical constituents such as ozone.

One current problem is the vertical coupling of the stratospheric QBO, both to the troposphere below and the mesosphere above. In the upper stratosphere, the amplitude of the QBO is much smaller than the semi-annual oscillation (SAO), but the stratospheric QBO wind regimes affect the vertically propagating waves that reach the mesosphere, and seem to cause a strong QBO near the mesopause. The upper equatorial stratosphere is also a region of ozone heating, which varies with the 11-year solar cycle. The possible role of the QBO in communicating variability of the QBO, the SAO, and the solar cycle to the extratropics is not yet understood, and is made more difficult by a lack of data above the highest (10hPa) radiosonde level. Rocketsonde wind data, which began in the early 1960s, is available at stations at 8°N and 8°S. These data extend well into the mesosphere, and may help to solve this problem.

The role of the QBO in stratosphere/troposphere coupling may take two forms. The QBO modulates the strength of the northern stratospheric vortex (and the timing of the break-up of the southern vortex), and high-latitude stratosphere-troposphere coupling (through the Arctic Oscillation) appears to communicate some effect from the QBO to the Earth’s surface. Although the QBO has only a very small amplitude in the equatorial upper troposphere, it may have an important effect on Atlantic hurricanes. This possible connection is not yet understood, and is based on a statistical relationship between the QBO phase and hurricane formation and intensity. Several mechanisms have been proposed to explain the effect of the QBO on Atlantic hurricanes, but the most promising mechanism may involve the QBO modulating vertically-propagating waves during hurricane formation.

Several GCMs have successfully simulated the QBO, or a "QBO-like oscillation." The QBO is generated spontaneously in these models, and it is challenging and time consuming to adjust the models to improve the QBO simulations. In general, high vertical resolution, as close as 500m, allows an improved simulation of gravity waves, which force the QBO. Small horizontal diffusion is also necessary. The simulation of the QBO in GCMs is a particularly challenging problem, but it is potentially an important one. It is becoming clear that realistic climate (and climate change) simulations require a realistic stratosphere. It is not yet known how important the global circulation changes associated with the QBO are to these simulations. Modelling studies have shown that increasing greenhouse gases may cause the northern polar vortex to become stronger and colder, and the presence of a QBO would tend to increase the variability of the vortex.

There have been many observational studies, using reanalysed data sets such as NCEP and ECMWF, as well as with satellite observations (e.g. HALOE) of trace constituents. These data sets allow the analysis of the QBO’s global effects on the zonally-averaged circulation, including the details of how the phase of the QBO interacts with the annual cycle in the tropics and subtropics, as well as ascent rates in the tropics. The analysis of zonally averaged trace constituent fields help to deduce the mean meridional circulation of the stratosphere. UARS/HRDI wind measurements provide a lower resolution replacement for rocketsonde measurements, which tapered off near the time that HRDI measurements began. Satellite observations of chemical constituents also allow analysis of both ozone feedback mechanisms and long-term variations.

In October the SPARC QBO Working Group submitted a paper, "The Quasi-biennial Oscillation," to Reviews of Geophysics.

JSM26. Chemistry and Transport in the Upper Troposphere and Lower Stratosphere (UT/LS)

M. Shiotani (shiotani@ees.hokudai.ac.jp), T. Shepherd and A. Ravishankara

This 2 1/2-day symposium focused on space-time variations in minor species and chemical-dynamical processes in the UT/LS. Special attention was given to stratosphere-troposphere exchange (STE) and to the recent ozone decrease in Arctic spring. A. Plumb (invited) opened the symposium with discussion mainly on the meridional circulation which controls STE. Factors controlling the circulation largely depend on the tropical angular momentum distribution and budget in the LS. A puzzle is why models, which are presumably too viscous, do not tend to have too strong diabatic circulations. Another topic was the "age" of air, especially how the age of mid-latitude air may change substantially in relation to the relative strengths of the meridional circulation and horizontal mixing. W. Sturges reported that C2F6 measured at Cape Grim seems to be an ideal age-of-air tracer in the stratosphere. On the basis of the Lightweight Airborne Chromatograph Experiment (LACE) observations, E. Ray investigated transport and mixing in the lowermost stratosphere. W. Grose showed results from a 3D GCM (IMPACT) including gas-phase and heterogeneous chemical reactions, with a focus on transport and mixing across the vortex-edge and subtropical barriers. L. Bruhwiler presented results obtained using the GFDL SKYHI on the effects of an imposed stratospheric QBO. The QBO was found to systematically modulate the eddy transport of ozone during the NH winter. P. Haynes (invited) focused on the extratropical tropopause in analogy with the edge of the stratospheric polar vortex. This analogy has two aspects. One is based on the fact that numerical simulations show that a quasi-realistic, statistical equilibrium tropopause may form in a simplified numerical model. Another aspect is based on transport properties of the isentropic velocity field obtained from observational data using the effective diffusivity diagnostics. T. Horinouchi used the NCAR CCM3 to investigate signatures of synoptic-scale Rossby waves in the UT/LS in connection with the Pacific westerly duct. J. Thuburn examined the tropopause height by taking into account the solar heating and dynamical warming or cooling, comparing results from simple models with GCMs. A. Gettelman compared several methods to diagnose STE from GEOS-1 assimilation data, and found that the instantaneous two-way exchange may be significantly exaggerated by Wei’s method. R. Rood used conditional probability distribution functions and showed that the GEOS model maintains a mix of air in the lowermost stratosphere similar to observations.

M. Nishida showed the tropopause statistics using the GPS/MET measurements. The tropopause altitude was found to be higher than the level of minimum temperature, especially in winter. A. Langford presented evidence from the Fritz Peak ozone DIAL (lidar) on the intrusion of stratospheric air into the troposphere in a fold. The intrusion itself contained tropospheric air, suggesting a return of tropospheric air back into the troposphere (reversible transport). C. Timmis addressed the question of the cause of low-ozone layers in the lowermost stratosphere over Aberystwyth, arguing that the layers already exist in the subtropics and are advected over Europe in NH winter. H. Teyssedre examined age of air with the Cambridge SLIMCAT and found it produces too old ages. P. Haynes, substituting for J.-R. Alisse, presented a combination of theory and balloon measurements to address the question of vertical diffusivity in the stratosphere finding values consistent with tracer-based estimates from aircraft measurements, but an order of magnitude smaller than estimates from radars. D. Jackson studied cross-tropopause transport in low latitudes using ERA analyses to drive off-line particle trajectories. D. Fahey gave an invited talk on photochemistry and transport in the high-latitude summer hemisphere based on the 1997 Polaris campaign. The summer is the time of the largest seasonal decrease in ozone due to increased sunlight and reduced transport. This period is relatively simple, because sulphate aerosol reactions are cut off and the chemistry is purely gas phase. The NO2/NO ratio exhibits far greater dynamic range than in winter; departures from theory are greatest for high solar exposure, for reasons that are not completely understood. S. Li studied the break-up of the Antarctic vortex using the CRC CTM and suggested that the vortex loses mass rapidly during October. S. Pawson examined the transport properties of assimilated data sets, using the GEOS-DAS, from the viewpoint of errors which can arise from deficiencies in the observations or in the assimilating GCM. S. Yoden studied chaotic mixing in the stratosphere with the flow generated by barotropic instability of a polar night jet.

D. Kley’s invited review on water vapour started with the earliest measurements by Brewer and the inference of tropical dehydration. The trends in water vapour abundance were most clearly shown by the data of CMDL over Boulder. The difference between the tropopause and hygropause was reviewed. The "tape recorder" picture semi-quantitatively explains the observed water vapour concentration and its temporal evolution in the LS. Some data from the recent CEPEX experiments were shown. I. Keramitsoglou examined the discrepancy between the Lyman-alpha and the frost point hygrometers, which might be due to the presence of extraneous vacuum UV radiation in the Lyman-alpha hygrometer. A. Dethof studied the isentropic transport of wate r vapour in the lowermost stratosphere and found that the cross-tropopause transport far exceeded descent from the overworld. S. Borrmann (invited) reviewed the role of condensed matter in the LS/UT. He showed evidence for potential halogen activation on so-called cirrus clouds. The repeated exposure of inactive halogen species to the "cirrus" leads to activation of chlorine. Since the deactivation is not very rapid, episodic repeated exposure leads to a higher level of active chlorine. He concluded that cirrus activation and its contribution to the ozone trend has to be considered unproven but likely. V. Dvortsov described how short-lived bromine containing species, such as bromoform, can provide a significant amount of bromine to the stratosphere via rapid convective transport in the tropics. L. Steffanutti, substituting for R. McKenzie, depicted the aims and preliminary results of the APE-THESEO. One of the findings was the presence of thin clouds almost continuously around 18km.

R. Cox (invited) examined the laboratory studies of some key reactions: ClO + HO2, BrO + HO2, IO + HO2, BrO + ClO, IO + BrO. The possible production of a hitherto-unreported species, OBrO, in the laboratory was described. It has been also reported to be present in the upper stratosphere. If true, it will require a complete alteration of our understanding of bromine chemistry. The recommended values of some of these reaction rate coefficients were presented.. H. Graf’s modelling aimed at investigating the relative role of ozone changes, greenhouse-gas changes, and natural variability in accounting for the observed changes in stratospheric climate. Tropospheric circulation changes resulting from a combination of greenhouse gas forcing and natural variability were argued to be responsible for the changes in stratospheric circulation. On the basis of results from the SOWER/Pacific mission over the Galapagos, M. Shiotani showed, contrary to expectations, clear evidence of freeze drying over a deep layer in March. H. Singh (invited) presented a summary of the findings of the SONEX mission over the north Atlantic: (1) the measured NOy abundance agrees with that deduced from individual components of NOy. (2) Enhanced nitrogen oxides in the aircraft corridor is clearly visible. (3) The measured ratio of HO2 to OH agrees very well with the computed value. The measured abundance of HOx is larger than predicted, the observed HO2 abundance increases linearly with the abundance of NO, and the measured abundances of CH2O are much larger than predictions. This mission is expected to yield a much better understanding of the chemistry of the UT.

Y. Orsolini used the SLIMCAT CTM to diagnose isentropic tracer transport in the LS, paying attention to the occurrence of laminae in N2O. C. Marquart examined quasi-isentropic transport in the LS by using high-resolution reverse domain filling calculations. M. Bourqui used one-year data from the ECMWF analysis and ozone measurements from the MOZAIC and NOXAR projects to infer the STE flux of air. D. Karoly focused on the zonally symmetric oscillation of the total column ozone and its variability in the Antarctic UT/LS, noting the contrast with the behaviour in the Arctic. W. Norton with the SLIMCAT CTM investigated interannual variability in isentropic transport and mixing in connection with the QBO, finding reduced mixing during the westerly phase. A. Iwi compared several UK models in view of the age of air, the tape recorder signal in the equatorial LS and its modulation by the QBO. O. Morgenstern investigated filamentation activity in the NH tropopause region using PV fields. R. Worthington presented results based on the MST radar at Aberystwyth and ozonesonde data. D. Fonteyn conducted 4D-VAR assimilation including stratospheric chemical species using the CRISTA observations.

JSM32. Small-scale and Layered Phenomena around the Summer mesopause

F.-J. Lubken and G. E. Thomas ( Thomas@iap-kborn.de)

A total of 17 oral papers were presented and 50-60 scientists attended the meeting. The main topic was the appearance of small scale layered phenomena, such as noctilucent clouds (NLC), polar mesosphere clouds (PMC), and polar mesosphere summer echoes (PMSE). Recent observations from modern rocket-borne and ground-based instruments, for example by steerable and multicolour lidars, were presented as well as new results from sophisticated modelling of fine structures in the neutrals and in the plasma.

A 22-year database derived from the MeteoSat revealed the presence of PMC over the sunlit polar regions. This data set opens up the possibility of determining solar cycle and longer-term changes of PMC. Southern hemisphere observations of PMC from the MSX spacecraft, of NLC from the ground at Davis, of PMSE from Peru's King George Station and of density/temperature from rocket soundings at Rothra provide further inter-hemispheric comparisons. Seasonal, diurnal and planetary-wave effects were reviewed, and indicated a strong causal connection between ice saturation and the various layered phenomena. Modelling of the relationship of turbulence, mixing, electric fields, and metallic species suggests that a complex combination of physical processes is responsible for the observed layered structures. Observations of PMSE at Svalbvard help determine layered phenomena in the deep polar region.

Several papers depicted hemispheric differences of these layers and their dependence on the background thermal structure. It is not clear yet how the obvious lack and weakness of PMSE echoes at southern latitudes is related to the thermal and dynamical structure of the background atmosphere. It seems as if differences in other geophysical processes (e.g. particle charging mechanisms) must be taken into account too.

The potential role of small scale layers in the upper mesosphere as indicators for long term trends was addressed. Although there seems to be no conclusive picture about the relationship between the obvious increase of NLC occurrence frequency and the thermal structure, there was general agreement that the high latitude mesosphere during summer is possibly the best place in the atmosphere to unambiguously detect the influence of anthropogenic activities.

The Layered Phenomena Working Group agreed to have the workshops every two years (the next in October 2000, possible venue Embry-Riddle Univ. at Daytona Beach, Florida; another possibility is Boulder, CO). At the next IAMAS meeting in Innsbruck (July 2001), layered phenomena will be covered at such sessions as MA chemistry, dynamics, long-term trends, etc.

MW01. Intercomparison of Troposphere-Stratosphere GCMs

Convener: Steven Pawson (pawson@polska.gsfc.nasa.gov), E. Drysdale and K. Krueger

Many of the invited presentations reported results from the SPARC-GRIPS initiative. The theme of the workshop was on the ability of climate-middle atmosphere models to simulate different aspects of the climate system and on their sensitivity to parameterisations and numerical schemes.

J. Koshyk's paper "The kinetic energy spectrum of horizontal motions of middle atmosphere GCMs" examined the simulated eddy kinetic energy spectra. All the models produced the familiar -3 gradient in the tropospheric wave energy spectra, where the flow is dominated by Rossby waves. In the stratosphere the lower wavenumbers had larger amplitude but the higher wavenumbers are smaller because of filtering. The mesosphere was dominated by gravity waves (GW), which form at ever smaller scales as resolution increases in the SKYHI model. K. Hamilton ("Sensitivity of model simulation of the middle atmospheric circulation to vertical and horizontal resolution") ran SKYHI at a variety of resolutions. Increasing the horizontal resolution (in the range N30-N270) improved many aspects of the simulation, especially the polar vortex structure. At high resolution, tropical cyclones were well modelled, having realistic depth, rainfall patterns and trajectories. With moderately high horizontal resolution, increased vertical resolution enabled the model to generate a QBO-like oscillation in the lower stratosphere (LS). Vertical resolution changes had little impact on the extra-tropics. In "Physically-derived dynamical cores in AGCMs," R. Rood and Shian-Jiann Lin presented a next-generation, flux-conserving dynamical core which has been coupled to a version of the CCM3 physics. The model is to be used for data assimilation, climate assessment and weather forecasting. As well as good climatologies, the model produces realistic frontal systems, manages a good simulation of a Gulf of Mexico hurricane and is numerically efficient. Overall, although limitations still exist (e.g. the model is hydrostatic), the new generation dynamical core is a feasible candidate for the next version of the NCAR CCM. T. Horinouchi ("Vertically propagating waves and the mean flow in the equatorial LS of GRIPS and other GCMs") examined the impact of vertically propagating waves in the tropics on simulating a realistic QBO and SAO. Compared were: MRI, FUB, SKYHI N30L40/N45L80, and CCM3MA, along with the AGCM5-HY98 model. The main differences between the tropical waves simulated in the models were found to be due to the space-time variability of cumulus convection influencing the excitation of waves, the model vertical resolution impacting the propagation mechanisms and the horizontal resolution affecting the oscillation mechanisms of these waves. The importance of GW for climate simulation was addressed by J. Alexander ("Parameterization of mean-flow forcing due to GW ") with several key areas highlighted: observations typically do not give the most useful parameters of the waves (the momentum flux) and, crucially, the nature of the wave sources is poorly known. Issues such as the source variability and uncertainty of the breaking processes can not be resolved without further observations.

In the poster presentations, R. Bannister et al. showed the impacts of upper boundary height on the simulation of tropospheric features. S. Pawson et al. presented a summary of the first model intercomparison performed for GRIPS. V. West and R. Harwood examined the transport properties of the unified model with a variety of advection schemes implemented, showing that choice of vertical advection scheme can severely impact the transport of trace gases. Wenyi Zhong et al. examined the sensitivity of the UGAMP model to the choice of radiation schemes, including a new version of the ECMWF scheme which includes a line-by-line LW transmission table (for CO2, O3 and H2O), an updated parameterisation for the water vapour continuum and a representation of oxygen absorption.

In "Selecting meteorological input for the Global Modelling Initiative (GMI) assessments," S. Strahan discussed the requirement to select an optimal meteorological data set for a GMI study of the impact of a stratospheric aircraft fleet. A set of objective criteria were established to test the quality of the data sets available (NCAR-CCM2 and GISS models and the GEOS-DAS assimilations). The criteria addressed the meteorological state and the transport characteristics of the data. Based on the results, the NCAR-CCM2 data set was chosen for the GMI assessment. G. Stenchikov and A. Robock split the presentation "Intercomparison of climate model simulations of the response to the 1991 Pinatubo volcanic eruption". The impact of the eruption on climate was simulated in several GCMs, using an ensemble of two-year integrations designed to study sensitivity to volcanic eruptions and SSTs. The largest response arises from radiative forcing due to the volcanic aerosol. This result is found to be insensitive to the SSTs used. The temperature increase of the tropical LS was larger than observations, apparently because they do not simulate a QBO. J. De Grandpre et al. ("Ozone study with the Canadian Middle Atmosphere Model") presented simulations of the pre-ozone hole SH column ozone performed with the Canadian MAM coupled to a fairly complex photochemistry scheme using a choice of semi-Lagrangian (SL) and spectral transport schemes. The spectral scheme proved to be numerical stable, non-dispersive, and produced a realistic climatology with no drift in trace gas concentrations. The present implementation of the SL transport remains diffusive, non-conservative and numerically unstable, making it unsuitable for interactive ozone-climate studies. U. Langematz et al. (presentation made by S. Pawson) discussed "First results from the radiation-intercomparison project for GRIPS." This study is motivated by the cold bias in most models. Offline calculations with identical input data revealed a considerable difference between the radiative codes investigated. In "The impact of two decades of ozone change on the stratosphere" S. Rosier et al. used an `Intermediate' GCM to study the effect of observed ozone changes on the stratosphere. The ozone changes were implemented in two ways. Initially, seasonal changes in ozone were applied to the column, and little signature was seen. Applying monthly mean ozone changes in the 150-35hPa layer led to a more realistic impact. The LS cooling reached -5K at 70°N and -15K at 86°S, rather stronger than values of -3K and -9K obtained from a FDH model experiment performed to isolate the radiative and dynamical contributions. J. Knight et al. ("The response of the stratospheric climate to changes in greenhouse gas concentrations from 1992-2051") examined two 70-year model runs with slightly different initial conditions, performed to study the impact of double CO2 concentrations with fixed SSTs on stratospheric climate. They found a linear cooling trend of -1.4K/dec at 10hPa, increasing PSC amounts, and a 2 day/dec delay in the timing of the SH vortex breakdown. The two simulations showed different responses, which might mean that climate models should be used to address the range of possible atmospheric changes, so that ensembles of predictions are needed.

MW-04. Atmospheric Tides in Global Models

K. Hamilton (kph@princeton.edu) and C. McLandress (mcland@stpl.cress.yorku.ca)

A one-day workshop was organised by ICMA to discuss the current state of tidal simulation in global models that include the middle atmosphere (MA). The workshop began with a number of talks summarising tidal observations, and, in many cases, comparing the observations to results from the Global Scale Wave Model (GSWM) developed by J. Forbes (U. Colorado) and M. Hagan (NCAR). The GSWM is a linear model of the tidal response to a prescribed diurnal or semidiurnal heating. It allows the inclusion of arbitrary mean wind and temperature structure. R. Vincent (U. Adelaide) reviewed tida l wind observations around 86km altitude from MF radars at 5 sites from Kauaii (22°N) to Davis (68°S). He noted significant seasonal and interannual variations in the diurnal tide. At Adelaide (35°S) the phase of the diurnal wind oscillation varies by 4 hours throughout the year, and the amplitudes have large year-to-year variability apparently related to the stratospheric QBO. Two radars very near the equator and separated by about 90 degrees longitude showed similar diurnal variations in the meridional wind, but a factor of two differences in the diurnal zonal wind variation. A. Manson (U. Saskatchewan) discussed MF observations from 7 stations from Christmas Is. (2°N) to Tromso (70°N) and compared them with results from two versions of the GSWM model, one (GSWM95) with mean winds based on CIRA and one (GSWM98) on UARS HRDI. Observed diurnal tides were generally better modelled by the GSWM than semidiurnal. Both the diurnal and semidiurnal tidal observations are in better agreement with the GSWM95 predictions than those from GSWM98. C. Jacobi (U. Leipzig) discussed semidiurnal tides observed from radars (MF, low frequency and meteor radars, all within 52°N-56°N). Significant longitudinal asymmetries were noted dependent on season. In January the semidiurnal wind amplitudes near 90km vary by a factor of 4 among the stations, with some success in relating variations in the longitudinal tidal asymmetries to patterns of stratospheric geopotential height. M. Riese (Wuppertal U.) discussed the diurnal temperature oscillation determined from CRISTA on the Space Shuttle in 1994 and 1997. The results were compared to the GSWM when run with mean winds computed with gradient wind balance from the CRISTA observed temperatures, and with winds from earlier HRDI observations. There was quite good agreement when the CRISTA-based winds, but much poorer when the HRDI winds were used.

C. McLandress (York U.) reviewed migrating tide observations from the UARS. Winds from WINDII and HRDI, and temperatures from MLS all exhibit a strong semi-annual variation in the propagating diurnal tide in the mesosphere. Interannual variations, which appear to be linked to the phase of the stratospheric wind QBO, are also evident in the HRDI data. Simulations of the diurnal tide using the Canadian MA GCM were also discussed. These were shown to exhibit a semi-annual variation of amplitude in the mesosphere, in good qualitative agreement with the observations. Resolved large-scale planetary waves were thought to be responsible for the amplitude modulation of the diurnal tide. These simulations did not employ a non-orographic GW drag parameterisation and so such drag was not involved in the seasonal tide variation. H. Mayr (GSFC) using a 3D non-linear mechanistic model with the Hines Doppler spread GW drag parameterisation found that the migrating diurnal tide exhibited a semi-annual variation in amplitude like that seen in observations. This was attributed to GW filtering by seasonally-varying zonal winds. Results showed that the GW scheme acted to increase the amplitude and modify the phase of the diurnal tide. J. Forbes (U. Colorado) discussed experiments with the GSWM incorporating parameterised GW effects. This incorporation produced major changes in the predicted tides, and both the parameterised momentum drag and eddy diffusion made important contributions to this effect. An attempt was further made to include geographical variability of the tropospheric latent heating excitation for the diurnal tide; this forcing results in predicted non migrating components of the diurnal and semidiurnal wind oscillations at 94km of order 10m/s. K. Hamilton (GFDL/Princeton U.) discussed the wind oscillations appearing in control simulations with a version of the GFDL SKYHI GCM with 54 levels up to about 130km. The model produces a realistic tidal surface pressure oscillation, but diurnal wind amplitudes in the lower thermosphere are somewhat weaker than indicated by UARS. The semidiurnal wind oscillation in the MA is quite strong and displays a pronounced seasonal cycle with largest amplitudes in the winter extratropics.

W. Norton (Oxford U.) used a 52-level version of the UGAMP spectral GCM, with an emphasis on seasonal variations in the mesopause-lower thermosphere (MLT) region. In low latitudes there is a strong semi-annual modulation of the diurnal tidal amplitude which is attributed to the interaction of the tide with planetary waves. S. Miyahara`s (Kyushu U.) 55-level spectral GCM produced a tidal simulation with quite significant non-migrating components in the lower thermosphere. Miyahara went on to show that in the MLT region the eddy stresses associated with the tide in the model significantly perturb the gradient wind balance normally expected for the zonal-mean zonal wind and zonal-mean pressure fields. This effect should be borne in mind when observed temperatures are used to infer zonal-mean wind at high altitudes. S. Oliver (Oxford U.) discussed simulations in a mechanistic version of the UKMO Unified Model (UM) with 36 levels up to about 96km. The diurnal thermal forcing for the experiment was taken from a run with the full UM. The mechanistic model produced a tidal simulation in the MLT region with higher amplitude and weaker zonal asymmetries than the full model. The diurnal wind variations in the mechanistic model at high latitudes were significantly stronger than those in radar observations. M. Geller (State U of NY at Stony Brook) showed linear model calculations of the seasonal and interannual variations in the diurnal tide. These used mean winds taken from HRDI above 55km and from unrelated GCM simulations below. The model included a GW drag parameterisation, and the variation of the GW-induced dissipation could account for the observed semi-annual modulation of the diurnal tide in the lower thermosphere. The model could also explain at least part of the interannual variability of the tides seen in the HRDI data. R. Swinbank (USRA/NASA Goddard) discussed the diurnal tide in the Goddard GEOS data assimilation scheme. The GEOS GCM, when run in control simulation mode, produced a reasonable representation of the tide in the stratosphere, but the assimilation products significantly misrepresented the tide. The problem was traced to inadequacies in the operational satellite temperature retrievals that are inserted into the model during the assimilation.

MW05. Radiative processes in the mesosphere and lower thermosphere

Conveners: G.M. Shved ( shved@lmupa.phys.spbu.ru) and V.I. Fomichev ( victor@nimbus.yorku.ca)

There were 10 presentations (including 3 invited and 2 poster) during this half-day workshop. Half of the papers were devoted to aspects of the energy budget of the mesosphere and lower thermosphere (MLT) region. V. Fomichev presented a summary of the current status of our understanding of the MLT energy balance and overviewed radiative parameterisations used in dynamical models. The main uncertainties in understanding the MLT energetics are primarily related to our knowledge of the quenching processes (especially for CO2-O collisions). In addition, important mechanisms such as the solar heating due to the absorption by the near IR CO2 bands and the thermal effects of the 9.6m m O3 band in the presence of chemical pumping, still need to be parameterised. V. Ogibalov presented first ever estimates of the effect of the two- and more n 2-quanta transitions during CO2-O collision on the 15m m CO2 band cooling and showed that the potential increase in the cooling rate could be as high as 15-25K/day near 130km. K. Koutoulaki et al presented estimates of changes induced by non-thermal emission in the 9.6m m O3 band cooling. X. Zhu and C. Fu et al. emphasised that in order to have a more realistic energy balance and atmospheric circulation, dynamical models should include fully-coupled chemical schemes.

K. Grossmann presented first results from the CRISTA measurements which showed longitudinal variations in O in the lower thermosphere and surprisingly small values of CO2 in the mesosphere. Ground-based nightglow measurements were used by S. Melo et al. to obtain variations in the vertical distribution of O.

Ogibalov and Shved considered the non-LTE CO2 problem and suggested some criteria allowing for the simplification of the CO2 vibrational level system, depending upon the accuracy required for the calculation. Kostcov et al. used the CRISTA measurements toretrieve the vibrational temperature of the CO2(0110) state for different CO2 isotopes in the 60-130km region. Ogibalov and Shved also demonstrated the possibility for the existence of an inversion population of the CO2(0001) and CO2(1000) vibrational states near 95-100km.

MW08: Quasi-Decadal Oscillation

Convener: Lon Hood (lon@pl.arizona.edu)

Recent observational studies of the quasi-decadal oscillation (QDO) in the middle and upper stratosphere (US) were discussed by L. Hood. Analyses of SBUV/SBUV2 and SAGE I/II data showed evidence for a decadal oscillation in upper stratospheric ozone (1-2hPa) that has been approximately in phase with the solar cycle during the 1979 to 1998 period. The amplitude of the observed US oscillation from solar minimum to maximum is roughly twice as large as is predicted by models that account for the observed solar cycle variation of solar UV spectral irradiance. A mechanism involving solar-induced changes in tropical upwelling, which affects the concentration of key trace constituents such as methane in the US, was proposed to explain the disagreement.

H. van Loon described global correlative studies of NCEP/NCAR reanalysis data below 10hPa at middle and lower latitudes with the solar 10.7 cm radio flux. The correlation patterns in each hemisphere are similar with maximum annual mean correlations at 30hPa in the subtropics. The correlations in the northern hemisphere (NH) confirm those derived from the 40-year Berlin data record. K. Labitzke discussed the influence of the equatorial wind QBO on the solar signal in the NH winter. Correlations between the solar 10.7 cm flux and geopotential heights below 10hPa are strong and positive in the Arctic when only QBO west phase years are considered. When only QBO east phase years are analysed, correlations are negative in the Arctic and are positive at lower latitudes. D. Shindell applied the GISS GCM to show how US ozone changes can induce circulation changes at lower levels extending into the troposphere. The model simulates several aspects of the observed QDO in the LS. M. Baldwin presented an overview of the dynamical effects of the equatorial QBO on the stratospheric extratropical circulation. Observed effects can be explained mechanistically by changes in the tropical zonal wind field which modifies the effective wave guide for upward and equatorward propagating planetary waves. During NH winter, the west QBO phase tends to excite the Arctic oscillation (AO), a deep, mostly zonally symmetric mode that couples the stratosphere and the troposphere. Downward propagation of the AO may be a general mechanism for downward communication of stratospheric circulation anomalies associated with the QBO or other (solar). Ch. Marquardt discussed possible mechanisms for the modulation of the extratropical winter LS by both the QBO and the QDO. During early and middle winter, modulation by the QBO alone dominates while during late winter, modulation by both the QBO and the QDO dominates. The occurrence of an additional biennial oscillation in the extratropical winter was also noted and was suggested to originate as a consequence of the interaction between the annual cycle, the tropical QBO, and the QDO. N. Balachandran and co-workers described further comparisons between the predictions of the GISS GCM and the observed LS oscillation. Using a version of the model in which ozone is held fixed but the solar UV flux is varied by large amounts, it was found that the subtropical maximum of geopotential height variation from solar minimum to maximum could be reasonably well simulated. Mechanisms involving downward propagation of changes in stratospheric circulation were discussed. The talk was presented by D. Shindell.

 

Back to SPARC Newsletter 14 Homepage