MEDEC Mesospheric Dynamics, Energetics and Chemistry
(Mesosphärische Dynamik, Energie und Chemie)
Final report
Förderkennzeichen: 07 ATF 10
Reporting period: January 1, 2001 to December 31, 2004
- 2 -
Partners in the Project:
• MPI-M: Max-Planck-Institut für Meteorologie, Hamburg • MPI-E : Max-Planck Institut für Extraterrestrische Physik, Garching • UNI-W: Fachbereich Mathematik und Naturwissenschaften, Bergische
Universität Wuppertal • UNI-L : Institut für Meteorologie, Universität Leipzig • IAP: Leibniz-Institut für Atmosphärenphysik, Kühlungsborn
Coordination of the Project:
Prof. Guy P. Brasseur
Max-Planck-Institut für Meteorologie
Bundesstr. 53
20146 Hamburg
phone: 040 41173421
fax: 040 41173430
For questions on this report, please contact:
Dr. Hauke Schmidt
Max-Planck-Institut für Meteorologie
Bundesstr. 53
20146 Hamburg
phone: 040 41173405
fax: 040 41173298
- 3 -
1. Scientific Objectives
The atmospheric region above the stratopause is in general much less understood than the region below. Or, as Paul Crutzen (1997) stated, a number of mysterious issues remain to be solved in the mesosphere.
The overall objectives of the MEDEC project were (1) to improve our understanding of the coupling between the dynamics, energetics and chemistry of the mesosphere, (2) to assess how these processes affect the global atmosphere, and (3) to estimate the vulnerability of the mesosphere to natural and human-induced perturbations.
More specifically, the consortium proposed:
• To quantify the mesospheric concentration, global budget and thermal impacts of chemical species, using recent space observations jointly with interactive multi-dimensional models of the middle atmosphere, and to identify the different processes (dynamics, radiation, chemistry) that control these budgets (MPI-M).
• To improve the representation of radiative transfer under non-LTE conditions, test this formulation versus observations from the CRISTA and MIPAS space experiments, and implement this code for the calculation of heating/cooling rates in middle atmosphere general circulation models (MPI-E).
• To use temperatures and assimilated observations of chemical compounds from the CRISTA experiment to study non-linear interactions between tides and planetary waves, and large scale mixing in the mesosphere (UNI-W).
• To use available ground-based and satellite measurements to evaluate numerical models and their ability to accurately represent dynamical and chemical processes in the mesosphere (IAP).
• To quantify the role of gravity waves and tides in the mesospheric large-scale circulation and assess the two-way dynamical connections between the lower and upper atmosphere, using a version of MAECHAM extended to the lower thermosphere (MPI-M).
• To investigate the coupling between the stratosphere, mesosphere and lower thermosphere through the quasi 2-day wave within the 11-year solar cycle (UNI-L).
• To identify the natural and human-induced processes that could potentially explain the observed trends in the temperature and concentration of chemical species (all partners).
These issues were tackled by collecting and analyzing ground-based, rocket-borne and satellite observations as well as by developing and applying complex three-dimensional numerical models of the entire atmosphere. Significant scientific progress has been achieved concerning all the tasks mentioned above. In the following sections, some of the progress obtained within MEDEC will be highlighted.
- 4 -
2. General Progress of the Project
Most of the partners could not start at the official beginning of the project (January 1, 2001) but during the course of the year 2001. This was mainly due to the fact that the project was officially approved only in the beginning of 2001 and several of the partners then had to search for appropriate staff. Several partners also had to cope with staff leaving during the course of the project (MPI-M, MPI-E, UNI-W) and subsequent problems to refill these positions without time-delay. Therefore, the original schedule could not be followed exactly by some of the partners. However, with the prolongation of the project duration by one year (without additional funding) almost all of the original goals have been reached.
At MPI-M, the coordinator of the subproject on gravity waves (Elisa Manzini) left and could not be replaced adequately. Therefore, a part of the money was used to invite visitors (e.g. from the National Center for Atmospheric Research in Boulder, Colorado, USA, and from the York University, Toronto, Canada) who contributed significantly to the development and evaluation of the Hamburg Model of the Neutral and Ionized Atmosphere (HAMMONIA).
A kick-off meeting for the project was held in Hamburg at the Max-Planck- Institute for Meteorology on March 12 and 13, 2001. This workshop was jointly organised with the AFO-projects KODYACS (Coupling of dynamics and atmospheric chemistry in the stratosphere, PI: Dameris) and ISOTROP (Integration of Satellite Observations with the global chemical transport model MOZART to study the chemical composition in the upper TROPosphere, PI: Rohrer), which concentrate on the stratosphere and the troposphere, respectively. With the development of the HAMMONIA model, MEDEC contributed also significantly to the COMMIT project (COMmunity Modelling InitiaTive, PI: Brasseur), which had the goal to develop a community state-of-the-art atmospheric model.
In the following years, annual project meetings and numerous small working meetings of two or three partners were organized to improve the communication and collaboration among the partners. MEDEC results were presented during a large number of international conferences, during the AFO 2000 Statusseminar in Schliersee from October 7 to 9, and during the final meeting of AFO 2000 in Bad Tölz (March 22 to 24, 2004). The success of the project may be evaluated from the scientific results presented in the following sections and in a large number of publications in international peer-reviewed journals (see “publications” section).
- 5 -
3. Temperature and Dynamics
During the summer months the upper mesosphere is characterized by the lowest temperatures in the Earths atmosphere leading to such spectacular phenomena as noctilucent clouds (NLC) and polar mesosphere summer echoes (PMSE). Different ground-based and rocket-borne measurements in the upper mesosphere and lower thermosphere performed by IAP enhance our knowledge of temperature and dynamics in this region. For example, in 2001, first in-situ and remote measurements of the thermal structure of the summer mesosphere at very high latitudes were performed from the north polar island Spitsbergen (78°N, Lübken and Müllemann, 2003). This database was complemented by intensive lidar measurements in 2002 and 2003. Now we have a quasi-permanent temperature coverage for the period from February to October (Figure 1). Continuous radar measurements in the last summers have shown nearly permanent PMSE occurrence over Spitsbergen from the begin of June to mid of August. This is significantly more frequent than PMSE occurrence in lower polar and mid latitudes and confirms constancy of the characteristic very low temperatures in the high polar mesosphere. The observations were used in an iterative process to evaluate and improve results of the newly developed chemistry and climate model HAMMONIA of MPI-M (see section 5).
Figure 1 : Preliminary results of seasonal temperature variation at 78°N from measurements at Spitsbergen by the mobile K-lidar of the IAP The comparison of temperatures simulated for the upper mesosphere at 69°N and 78°N and observed with the “falling sphere” technique are shown in the Figure 2. The current model version reproduces the seasonal variation of temperatures in the upper summer mesosphere provided that the model altitudes are corrected by ~3 km.
- 6 -
Figure 2 : Seasonal temperature variation at 69°N (upper pane ls) and 78°N (lower panels). Seven day mean values are displayed. They show very good conformities between the “Falling Sphere” data (red lines) and the HAMMONIA model output (blue lines), shifted by values of 3 km.
Temperatures at mid latitudes were measured for several years at Kühlungsborn (54°N) using a potassium lidar. Comparisons of observations and HAMMONIA model data show good agreements; however, the model temperatures are still too low in altitudes below ~85 km in summer and in winter (Figure 3).
- 7 -
Figure 3 : Height-time cross sections of temperatures at 54°N: HAMM ONIA model output (upper panel) and measurements by potassium lidar (lower panel) show good agreements in a wide height range.
The validation of the mesospheric circulation estimated by HAMMONIA with experimental results using long-term observations (MF radar Juliusruh, 55°N) carried out by the IAP shows a good agreement of the seasonal behaviour of the zonal prevailing winds in mid latitudes. The basic features of the mean zonal wind as the strong easterly wind jet of the mesosphere during summer and the reversal to weaker westerly winds during the winter months are nicely reflected by the model, especially at heights below 85 km. The thermospheric westerly winds in summer produced by the model however are stronger than the results derived from the observations. Observations of the zonal winds at high latitudes (MF radar Andenes, 69°N) show comparable properties to the results at mid latitudes characterized by dominating strong easterly winds in the mesosphere during the summer months. However these jets reproduced by the HAMMONIA model are still too weak. Comparisons between measurements and the model at low latitudes (MF radar Yamagawa, 31°N) show very good agreements of the zonal wind values and reversals (Figure 4). As already discussed for the temperature the model altitudes should corrected by ~3 km.
- 8 -
Figure 4 : Annual cycle of zonal winds (m/s) observed by MF-radar at Yamagawa (31°N, 131°E) in the year 2002 (upper panel), and si mulated for the respective location by the HAMMONIA model for an arbitrarily chosen year (lower panel).
Observations of turbulence parameters (e.g. Müllemann et al., 2002) complete the data sets. Process studies have been carried out to improve our knowledge of physical processes connected with the rapid change of the mesospheric thermal structure during the transition period from winter to summer, the interaction between the thermal structure, large scale circulation pattern and the corresponding turbulent activity using in situ rocket-borne and radar measurements 69°N.
The beginning of the summer period in the upper mesosphere is indicated by the change from eastward winds (u > 0 m/s) to westward winds (u<0 m/s) and could observed by continuous radar measurements. Two rocket launches in spring 2000 took place prior and after the change of the thermal and dynamical structure. Temperatures measured by the rocket soundings show a cooling of approximately 20 K at an altitude of 82 km and confirm the transition of the upper mesosphere from the winter to the summer state. Neutral air turbulence also shows a distinct seasonal variation with relatively low turbulent energy dissipation rates in the order of 10 mW/kg throughout the altitude range from 60 to 100 km in winter, whereas during the summer months the turbulent energy dissipation rates are restricted to a relatively small altitude range of ~80-95 km but they are reaching much higher rates of more than 100 mW/kg. This behaviour was confirmed using the estimated turbulent energy dissipation rates based on the measured relative density fluctuations by the rocket-borne instruments (Figure 5). Hence three observed atmosphere parameters display the prominent winter-summer-change at the same time.
50 100 150 200 250 300 350daynumber
75
80
85
90
heig
ht (
km)
-70-60-50-40-30-20-1001020304050
50 100 150 200 250 300 350
75
80
85
90
hei
gh
t (k
m)
Yamagawa MF Radar (31°N, 131°E) - zonal wind 2002
HAMMONIA (31° N, 131° E)
- 9 -
Figure 5 : Height-time cross section of the background mean zonal winds from the MF radar Andenes (69°N) for the period from 1.04.-30.06 .2000. The contour level of 0 m/s is emphasized by a thick black line. The times of the rocket launches are marked by white vertical lines (left panel, from Müllemann et al., 2002). Turbulence measurements for the rocket flights (MSMI03 and MSMI05), and the mean winter and summer profiles of the energy dissipation rate (right panel, from Müllemann et al., 2002) The global picture of mesospheric temperatures is improved by the analysis of infrared spectra observed during the CRISTA (CRyogenic Infrared Spectrometers and Tele-scopes for the Atmosphere) satellite missions in November 1994 and August 1997 by UNI-W (Grossmann et al., 2004). Figure 6 shows a polar view on temperatures at 82 km in August 1997. The lowest temperatuers are found above the northern sector of Canada/Alaska. Under these temperature conditions polar mesospheric clouds (PMC) can form. These were observed simultaneously by CRISTA through the thermal emission of the ice particles at 12 µm wavelength. The geographical cloud distribution
Figure 6 : Temperature at an altitude of 82 km around the North Pole (August 1997) observed by CRISTA.
- 10 -
over the course of the CRISTA mission agrees well with the thermal structure. A considerable part of the mesopause region water content is stored in these clouds (Grossmann et al., 2005). The zonal structures observed by CRISTA in the mesosphere indicate that the now commonly accepted dynamical control of the stratospheric trace gas distribution by planetary waves also governs the mesosphere. Respective structures were found in temperatures and ozone concentrations in the winter hemisphere as well as in the summer hemisphere in August 1997.
- 11 -
4. Chemical composition
CRISTA is a triple telescope cryogenically cooled infrared spectrometer which senses the earth limb from a Shuttle orbit. The geographical coverage was -57°/+68° and -74°/+74° during the two missions of slightly more than one week duration each. In the mesosphere and lower thermosphere, the CRISTA data allowed - beside temperatures - the analysis of pressure, winds (derived from temperatures), and a set of trace gases including ozone (Kaufmann et al., 2003), carbon dioxide (Kaufmann et al., 2002), carbon monoxide (Grossmann et al., 2005), methane, water vapour and atomic oxygen (Grossmann et al., 2000). At these altitudes carbon monoxide is of particular interest since it is an excellent tracer for dynamics due to its very long chemical lifetime in combination with a strong vertical gradient. The analysis yielded very large tidal variations at mid-latitudes with relative amplitudes of up to 50%. The overall latitudinal distribution of CO is governed by large scale inter-hemispheric transport with upward movement in the summer hemisphere and downward movement in the winter hemisphere. Over the course of the CRISTA missions vertical transport velocities of up to 600 m/day are derived (Grossmann et al., 2005). Horizontally CO is modulated by planetary waves and indications of a mesospheric surf zone are seen in the CO data of November 1994 (Oberheide et al., 2005).
An important finding regarding mesospheric trace gases was the discovery of the tertiary ozone maximum in the winter mesosphere at about 70 km in a study performed jointly with colleagues from NCAR (National Center for Atmospheric Research, Boulder, USA, Marsh et al., 2001). This feature is the result of a decrease in atomic oxygen loss by catalytic cycles involving odd hydrogen in the vicinity of the terminator where the atmosphere becomes optically thick for the radiation (Lyman- ) leading to water vapour photolysis. Figure 7 shows the tertiary maximum in both CRISTA observations and HAMMONIA simulations.
Another important feature in the atmospheric ozone field that has probably been described from HAMMONIA simulations for the first time (Schmidt et al., 2004) are
Figure 7 : Ozone volume mixing ratios as observed during August 1997 by CRISTA (from Marsh et al., 2001, data averages for local times of approximately 0000, 0400 and 0500 hours for 70S (black), 30S (red) and 2S (green), respectively) and simulated by HAMMONIA (monthly mean zonal average for August). Both data sets show the tertiary ozone maximum in the mesosphere at high southern latitudes.
- 12 -
the relatively high concentrations in the summer mesopause region. Figure 8 shows daytime zonal mean ozone mixing ratios for a given local time from HAMMONIA and from the SABER instrument on the TIMED satellite. The high concentrations close to the summer mesopause seem counterintuitive because they occur where atomic oxygen concentrations are very low due to the upwelling in the polar region. However, this feature is a consequence of the very low temperatures and can be explained from the ozone equation under photochemical equilibrium following Allen et al. (1984).
Figure 8 : Observed and simulated July daytime ozone mixing ratios [ppmv]. Left: Mean values for 1500 local time (LT) from SABER/TIMED observations between 6 and 14 July 2002 above 4e-4 hPa, and zonal mean July values from the UARS reference atmosphere project (URAP) below. Right: Simulated July mean values for 1500 LT.
- 13 -
5. The quasi two-day wave: investigations on the or igin and propagation in the middle atmosphere (Uni-L)
The quasi two-day wave (QTDW) is one of the most prominent features of the upper middle atmosphere. This planetary wave with a zonal wave number of 3 and a period of about two days is regularly observed in the summer hemisphere during solstice conditions when it exhibits strong amplitudes. This phenomenon lasts for few weeks, while during the rest of the year the wave amplitudes are small, but nevertheless often still observable. Two mechanisms of origin and excitation of the QTDW are in discussion. The wave can be identified as a solution of the Laplace tidal equations but this can not explain their amplification during each summer season. It is therefore assumed that the QTDW can be enhanced or even excited via instability processes. Likewise, interactions with gravity waves, solar tides and other planetary waves can play a role. The COMMA-LIM circulation model was updated (Jacobi et al., 2005) and used to study the propagation of the QTDW in the middle atmosphere, its impact on the mean flow and its interaction with manifold middle atmosphere waves. In this way, the knowledge about the wave could be summed up but revealed also the gaps in understanding and modelling the middle atmosphere. The quasi two-day wave was inserted into the model as a heating disturbance per unit mass at around the tropopause level. The forcing itself was smoothed in the vertical with an exponential factor F and the disturbance term h2dw was defined by the properties of the wave:
h2dw = A�
( � )F(z) cos(kx- � t) where F(z) = exp[-(z-10)2/25] with z in km, A is the amplitude scaled to produce the observed values,
�( � ) represents the latitudinal structure of the wave, obtained from
Hough-mode calculations for the Rossby-gravity wave (3,0) (Swarztrauber and Kasahara, 1985). The zonal wave number is given by k=3, x=2� � /360°. Within the angular frequency � =2 � /T the period T=52.5 h has been chosen, as this period gives the resonant response in COMMA-LIM. The numerical simulation of a QTDW forced as a steady Eigenmode shows a Rossby-gravity wave propagating upwards into the MLT region, predominantly in the summer hemisphere, see Figure 9. The maximum of the meridional wind disturbance is twice as large as the maximum of the zonal wind component and has a symmetric behaviour with respect to the equator in contrast to the other wave fields zonal wind, temperature and geopotential field. In the summer hemisphere the wave exerts a westward acceleration on the mean flow that intensifies the summer westward jet. Interaction between gravity waves and QTDW Gravity waves (GW) play a crucial role in the dynamics of the atmosphere. These waves are necessary for the QTDW to be able to propagate upwards into the MLT region. Gravity waves are responsible for the wind reversal at mesopause heights. The imposed eastward acceleration on the summer westward jet leads to a meridional circulation, which in turn causes upwelling at the summer mesopause and downwelling at the winter mesopause, leading to the temperature anomaly in polar mesopause heights. In this fundamental way, through decelerating the mesospheric jets and imposing a reverse circulation, gravity waves provide a vertical path for planetary waves. If the winds are slowed down the QTDW is able to propagate higher because
- 14 -
their critical line, where the zonal mean wind matches the wave's phase speed, rises or even vanishes. Further investigations into the influence of GW drag on both the mean flow and the QTDW used GW-amplitudes of different strengths. While the standard amplitudes of the gravity waves have on average a speed (vertical velocity) of ~2.1 cms-1 at the equator, the large-amplitude GW were forced with ~2.7 cms-1 (case 1)and the small-amplitude GW were set to a speed of ~1.5 cms-1 (case 2). The larger the GW amplitudes at the excitation level, the lower the heights of which the gravity waves reach their critical size and break. On the other hand, small GW-amplitudes cause very high propagating gravity waves that have a minimal effect on the mean flow in the mesosphere. The increased jets lead in turn to stronger acceleration if the waves break higher in the lower thermosphere. An increase in GW activity leads to weaker easterlies at low latitudes near the stratopause. These changes in the general circulation near the equator modify QTDW-propagation conditions into the northern hemisphere mesosphere-lower-thermosphere region. Figure 10 shows the zonal mean wind and the meridional wind amplitudes of the QTDW for the two experiments. Case 1 leads to strongly reduced jets allowing the QTDW to penetrate higher into the MLT region. The vertical propagation of the QTDW for case 2 stops at altitudes where the easterly jet core exceeds values of 60 ms-1 which is approximately the phase speed of the QTDW at mid-latitudes. This means, that due to the highly variable nature of their generation GW might be responsible for transient-like amplitudes of the QTDW. The experiments also revealed the still existing gaps of knowledge and numerical capability which are needed in order to realistically describe the gravity waves in the middle atmosphere. Interaction between the QTDW and other planetary waves The theory of a non-linear coupling between two planetary waves or a planetary wave and tides says that two ''primary'' waves can interact through the advection terms in the momentum equation and produce a family of ''secondary'' waves (Teitelbaum and Vial, 1991). For instance, if a signal consisting of two cosine waves with zonal wave numbers and frequencies (k1, 1) and (k2, 2) pass through a quadratic system, the output of this system will contain the secondary waves (2k1, 2 1),(2k2, 2 2), (k1+k2, 1+ 2), (k1-k2, 1-
2). The strongest secondary waves are those whose frequencies are the sum and difference of the frequencies of the primary waves. The secondary waves then beat with the primary waves and modulate the amplitude of the higher-frequency wave at the period of the lower-frequency wave. Modulation of QTDW by 16-day wave and 10-day wave Figure 11 shows the results of interaction between the QTDW and two prominent planetary waves. In the top panels the modulation in the meridional wind field of the QTDW by the 16-day wave (16DW) in the mesosphere and thermosphere is displayed for two different latitudes. This picture is obtained by taking the difference between the control run only with the QTDW and the experiment including both waves. The variation in the zonal wind field at 32° N accounts for 1.5 ms-1 (not shown here) representing an ~20% change, whereas the meridional component is modulated strongest at the equator with around 4 m s-1 (referring to an 8 % change). The strength of modulation decreases towards the north pole at mesospheric heights. Furthermore, increasing modulation in the course of July can be seen. Thus, these are mechanisms that could be responsible for transient-like behaviour of the QTDW if not for burst-like events. However, COMMA-LIM results showed a weaker modulation than that reported by Pancheva et al. (2000). In their study the variation of wave amplitudes exceeded 400
- 15 -
% between minimum and maximum. Additionally, in observations by Pancheva et al. (2000) the QTDW seemed to grow through the interaction with the 16DW, whereas in COMMA-LIM the 16DW-modulation results in smaller amplitude values than those for the controlled QTDW (positive differences between control and modulated QTDW). Similar results are obtained for the modulation of the QTDW by the 10DW. The modulation begins also at around 60 km height where the winter planetary waves penetrate into the summer hemisphere and is comparable in magnitude and latitudinal decrease of modulation. Interaction of QTDW and stationary planetary wave with zonal wave number 1 (SPW1) Strong secondary waves arised from interaction of these two waves: Especially the wind amplitudes of wave number k=2 show values that reach about 50% of the original QTDW, and the structure looks similar to the primary QTDW. It was proved that the k=2 QTDW arises from interaction with the stationary planetary wave and not as a subharmonic of the diurnal tide by carrying out a separate calculation without the SPW1. Then, such waves did not arise. Furthermore, the very weak response of the self-interacting diurnal tide exhibits a different shape from the k=2-QTDW. Wave number 4 is not as strong but still pronounced. A further interesting feature is that the QTDW forced without the presence of a SPW1 exhibits amplitudes of the same magnitude as in the control experiment. This means that an existing SPW1 delivers the main part of energy to the secondary two-day waves. Under this assumption transient events of SPW1 will increase the secondary wave amplitudes. The secondary QTDWs indicate that it seems necessary to take them into account when comparing the QTDW with radar measurements. Local measurements are only able to collect the frequency but cannot account for different wave numbers of the observed oscillations or divide them into parts of primary and secondary waves. By restoring the three waves back to the longitudinal grid the superposition of the waves with equal frequency but different wave number and amplitude shows a wave with the wave number of the strongest amplitude, see Figure 12. The longitude-latitude plots show that the wave number changes to 2 in the high-latitude winter hemisphere; however, there the QTDW amplitudes are smaller than 5 m s-1. Nevertheless, regarding this picture it seems possible that the wave number can change at specific latitudes or altitudes if one of the secondary waves grows stronger than the primary QTDW. The characteristic feature of the not equally distributed distance between two ridges over the latitudinal circle develops due to beating between the three waves. In the bottom panels of Figure 12 a clear example is given for the wind amplitudes at 32°N, 72 km height. Wave number estimation from radar stations at different longitudes would yield in this case a non-integer wave number as was the case in the papers of Poole and Harris (1995) and Meek et al. (1996),. These authors suggested that beating of waves might be responsible for wave number ambiguities. COMMA-LIM model investigations could confirm this assumption and explain it by taking into account the SPW1-QTDW interaction. Interaction with tides Considering the tides in relation with the QTDW it turned out that during QTDW events the tidal amplitudes reduce. This reduction seems to depend on the magnitude of the QTDW. Results of investigation on the secondary planetary waves and comparisons with Palo et al. (1999) indicate that the 16 h and 44 h-waves due to interaction with the
- 16 -
diurnal tide draw their energy dominantly from the tide, while the QTDW is more responsible for feeding the k=5, 9 h wave in interaction with the semidiurnal tide. Simulation of the 11-year solar cycle The question, if the QTDW is able to contribute in a remarkable way to solar induced wind and temperature changes, has to be answered in the negative. By comparing the numerical results of wind amplitudes at mid-latitude MLT region with Jacobi (1998) the results are not in agreement, because the QTDW-activity is smaller during solar maximum conditions. However, at mesospheric heights around the equator the QTDW propagates better for solar maximum conditions. Several reasons may account for this discrepancy: first, the missing QBO in COMMA-LIM, which provides another background circulation. Another possibility is the influence of the ozone climatology, and its variation in total and its latitudinal distribution throughout the solar cycle. Former studies gave remarkable different results if another data set for ozone or another variation between solar maximum and minimum were applied. Experiments with unstable stages of the atmosphere The numerical investigations on exciting an unstable growing QTDW from inertial or baroclinic instability failed when using the original COMMA-LIM model setup. COMMA-LIM can be tuned to exhibit similar summer conditions that are known to be sufficient for generating a QTDW in other models, however, still no growing QTDW could be observed then. It turned out furthermore, since the different model design ranging from linear and purely dynamical models up to GCMs does not prevent a wave from developing, this instability is a dynamical process existing in the mesosphere. The troposphere may play an indirect role as upward propagating disturbances alter the mesospheric jets but cannot penetrate the enhanced summer jet. Thus, the coarsely resolved troposphere is not the reason for the missing QTDW in the model. Therefore, the most important difference between model successfully and not successfully modelling instability seems to be the spectral design of the models that appears to be able to generate unstable waves. To investigate the possibility of wave generation via instability processes a spectral model based on the COMMA-LIM configuration was used (Merzlyakov and Jacobi, 2004). It could be shown that indeed a QTDW develops depending on an artificially increased summer mesospheric jet (Figure 13). These waves influence in turn the zonal mean flow. Strong QTDW's are also unstable and generate secondary waves with larger wavenumbers and longer periods. The interaction of the QTDW with wavenumber 4 and the 10-day wave with wavenumber 1 causes a new QTDW with a period of about 55-60 h and other secondary waves. This can be seen in Figure 12.
In summary it could be shown, that not only instability processes can lead to an increase in QTDW activity during summer. In particular, varying gravity waves or transient stationary planetary waves were found to be important in these processes. These results underline that the atmosphere allows several solutions to the phenomenon called the QTDW.
- 17 -
Figure 9 : Amplitudes of the quasi two-day wave with zonal wave number 3 and period T=52.2h for the zonal and meridional wind, temperature and geopotential height.
Figure 10 : Top panels: July mean of zonal mean wind for small- and high-amplitude gravity waves. Bottom panel: the subsequent QTDW amplitudes in the meridional wind field
- 18 -
Figure 11 : Top panels: time-height plot of modulation of the QTDW by the 16DW near the equator and at midlatitudes. Bottom panels: modulation by the 10DW.
Figure 12 : Superposition of wave number 2, 3 and 4 with period T=52,5h for zonal (left) and meridional wind (right). Upper panel show a longitude-latitude section at 72 km, whereas lower panels depict the sum at a specific latitude.
- 19 -
Figure 13 : Dependence of the amplitude (in m/s) of the 2-day wave (right panels) on the velocity of the easterly jet (left panels, values given in m/s). Axis labels are approximate altitude (in km) and colatitude.
Initial mean zonal wind
weak instability min=-81 period ~52.7h
medium instability min= -85 period ≥52h
strong instability min=-95 period ~51.5h
Min U0 = -64 m/s Stable distribution
���
����
��
�
- 20 -
6. The response of the mesopause region to natural and anthropogenic climate forcing
Before the start of this project coupled general circulation and chemistry models of the whole atmosphere did not exist. However, such a model is an essential tool for studying the mesosphere, its response to climate variability, and interactions with other height regions. The main task of MPI-M within the MEDEC framework was the development
Figure 16 : Zonal mean temperature change for three different “2*CO2” simulations with respect to the reference run with a CO2 concentration of 360 ppm. Top: Simulation where CO2 concentration and SSTs are changed. Lower left: Simulation where only the SSTs are changed. Lower right: Simulation where only the CO2 concantration is increased.
Figure 14 : Relative difference of ozone volume mixing ratio between the simulations for solar maximum and solar minimum conditions. Differences are given in % as zonal mean values for the month of July.
Figure 15: Temperature difference between the simulations for solar maximum and solar minimum conditions. Differences are given in K as zonal mean values for the month of July.
- 21 -
and application of such a model, HAMMONIA (HAMburg MOdel of the Neutral and Ionized Atmosphere), which covers the altitude range from the surface to the thermosphere. The model combines dynamics and physics from the ECHAM5 general circulation model (Roeckner et al., 2003) in a fully interactive way with the MOZART3 chemistry scheme (48 compounds, 152 reactions). The model physics were extended by parameterizations accounting for important processes in the upper atmosphere: e.g. heat conduction, solar heating in the UV and extreme UV, solar heating by CO2 in the near IR, non-LTE IR cooling, the ion drag and the gravity wave forcing (Charron and Manzini, 2002).
Model applications concentrate on the response of dynamics and trace gases in the MLT region to solar and anthropogenic climate forcing. Different simulations with HAMMONIA for low and high solar activity on the one hand, and for present day and doubled CO2 concentration on the other hand were performed (Schmidt and Brasseur, 2004, Schmidt et al., 2004). Results of the solar cycle experiments are e.g. an ozone increase for high solar activity up to 25% around the mesopause (Figure 14). The temperature is increased by 3 to 10 K (Figure 15) which corresponds well to observations listed by Beig et al. (2003). Together with the simulated decrease in water vapour this should have implications for the NLC formation. CO2 doubling (not shown here) leads to a temperature decrease everywhere above the tropopause but by only very small and sometimes even insignificant values in the mesopause region.
A sensitivity study on the atmospheric response to a doubling of the concentration of CO2 gives interesting insight on the role of dynamics in mesospheric climate change. Four experiments have been performed: a) a reference run with a CO2 concentration of 360 ppm fixed at the surface, b) a “2*CO2” simulation with a concentration of 720 ppm and the sea surface temperature (SST; and sea ice) changed according to a doubled CO2 tropospheric climate, c) a simulation with only CO2 changed to 720 ppm, and d) a simulation with only the SST changed. Figure 16 shows the zonal mean temperature change in the month of July for the latter three experiments with respect to the reference experiments. It can be seen that the atmosphere is expected to cool everywhere above the tropopause which is mainly a result of the increased infrared emissions by CO2. However, the cooling rates are smallest in the mesopause region. In the experiment with only the SST changed a warming occurs in this region. This indicates that the relatively small and sometimes even insignificant cooling near the mesopause (which is also a result from trend analyses of long-term observations) is only partly an effect of the small importance of CO2 cooling in this region but also influenced by dynamical feedbacks.
- 22 -
7. The role of gravity waves in mesospheric dynamic s
It is well known that gravity waves (which have their origin mainly in the troposphere) and in particular the momentum deposition due to the wave breaking are to a large extent responsible for the large scale dynamics of the mesosphere. Additionally, the diffusion caused by the wave breaking is important for the distribution of trace gases in this height region. The study of these phenomena in all their complexity requires a numerical model extending from the Earths surface to the thermosphere that includes all the relevant physical and chemical processes.
Therefore, the main task of this MEDEC subproject was the development of such a model and its application and analysis with respect to gravity wave processes. The HAMMONIA model was developed in close cooperation with the MEDEC group „Chemical Budgets“. And has been described above. In this section, we present the analysis of model simulations with respect to inter-hemispheric asymmetries and the effect of the 11-year solar cycle on the mesospheric dynamics. In our presentation we focus on the summer mesopause region.
It is under discussion to which extent inter-hemispheric asymmetries in the mesopause region exist. A comparison of falling sphere summer temperatures (Lübken et al., 1999; Lübken, 1999) shows no difference of the minimum temperatures at summer solstice but warmer temperatures thereafter for the southern hemisphere, while a comparison of
Figure 17 : Zonal mean temperatures as simulated by HAMMONIA for the months of January and July, and the temperature difference in the summer hemisphere for these months.
- 23 -
satellite observations suggests a warmer southern summer mesopause by about 6K and also differences in zonal winds (Liebermann et al., 2000).
For our model analysis we use mean values from a HAMMONIA simulation for present-day solar minimum conditions (see above). The model shows a stronger temperature difference (~11K, Fig. 17) and also stronger differences in lower zonal jets. From the model analysis it is found that gravity wave filtering may provide a mechanism for such asymmetries.
Concerning the solar cycle effect, as expected, a significant warming occurs in the regions from the upper mesosphere upwards with values of ~5K in high latitudes near the summer mesopause. Changes in residual vertical velocity support the possibility that not only a radiative but also a dynamical contribution of this heating exists. In the case of zonal winds, a significant decrease of the lower thermospheric and upper mesospheric jets can be observed (Fig. 18). Above 1.e-4 hPa, this is mainly due to the imposed stronger ion drag for solar maximum conditions. Between 0.01 hPa and 1.e-5 hPa a weaker gravity wave drag is observed which is consistent with weaker vertical wind shears. But in general, changes below the mesopause seem to be small and statistically not significant in most regions.
Figure 18 : Zonal mean zonal wind as calculated from HAMMONIA for the month of January under solar minimum conditions (right) and the difference between solar maximum and minimum conditions (left). Grey shading indicates regions where the statistical significance of the differences is higher than 95% (light) and 99% (dark), respectively.
- 24 -
8. A new method to compute non-LTE infrared cooling
A new efficient subroutine for estimating the total infrared radiative cooling/heating in the ro-vibrational bands of atmospheric gases in the mesosphere and lower thermosphere (MLT) was developed at MPI-E. It represents the optimized version of the exact non-LTE code described by (Gusev and Kutepov, 2003). The subroutine accounts for vibrational non-LTE and absorption and transformation of the near-infrared solar radiation and relies (a) on the accelerated lambda iteration (ALI) solution of the vibrational non-LTE problem and (b) on the “opacity distribution function” (ODF) technique used in the stellar radiative transfer (Mihalas, 1978). Compared to the correlated k approach widely used in atmospheric LTE radiative transfer, the ODF technique works for both LTE and non-LTE conditions. It allows treating each ro-vibrational band as a single line of a special shape whose variation with respect to pressure and temperature were tabulated and parameterized. Compared to the standard line-by-line (lbl) approach the ODF method provides an acceleration of the calculations by a factor close to the number of lines in the band (about 200 times for CO2 and about 1000 times for O3 and H2O bands).
Opposite to various parameterizations available the new routine treats precisely the exchange of radiation between various atmospheric layers and all variety of collisional energy transfers. As a result it accurately reproduces the non-LTE line-by-line radiative
cooling/heating reference calculations both for day and night conditions even for non-smooth profiles of atmospheric parameters when best parameterizations fail (Figure 11). Compared to the parameterizations new subroutine also (a) allows varying all input collisional rate and spectroscopic parameters, and (b) calculates cooling/heating with a prescribed accuracy by utilizing pre-selected optimized sets of molecular vibrational levels and bands. The subroutine has been used for the estimations of the infra-red
Figure 19 : Cooling/heating (K/day) in CO2 infrared bands calculated with the exact line-by-line technique (black), the parameterization of Fomichev et al. (1998, green) and the new “ODF” approach (red) for a particular temperature profile measured with the CRISTA instrument (right figure).
- 25 -
radiative cooling/heating during Solar Proton Event (Barabash et al., 2003). It was recently incorporated into the GCM of the Martian atmosphere developed in the framework of the DFG SPP-1115 MAOAM project in Katlenburg-Lindau (Hartogh et al, 2005).
- 26 -
9. Conclusions
The results of the project presented above prove that large progress was made in both analyzing observations and modeling of dynamics, radiation and chemistry in the mesosphere.
E.g., the HAMMONIA model is now worldwide one of only three general circulation models (together with the WACCM model from the National Center for Atmospheric Research in Boulder, USA, and the Extended Canadian Middle Atmospheric Model, which are both still under development) that resolve the atmosphere from the surface to the thermosphere.
New insight has been gained concerning the state of the mesosphere as well as the processes acting and interacting to produce this state. However, some of the ``mesospheric mysteries’’ remain, and the developed tools as well as the new observations and their analyses will probably be of great value also in future mesospheric research.
- 27 -
Publications ALPERS, M., EIXMANN , R., FRICKE-BEGEMANN, C., GERDING, M. AND J. HÖFFNER,
Temperature lidar measurements from 1 to 105 km altitude using resonance, rayleigh and rotational Raman scattering, Atmos. Chem. Phys., 4, 793-800, 2004.
BARABASH, V., S.KIRKWOOD, A.FEOFILOV, A.KUTEPOV. Polar Mesosphere Summer Echoes during July 2000 Solar Proton Event. Annales Geophysicae, 22, 759–771, 2004.
CHARRON, M., AND E. MANZINI , Gravity waves from fronts: Parameterization and middle atmosphere response in a general circulation model, J. Atm. Sc., 59, 923-941, 2002.
FRÖHLICH, K., A. POGORELTSEV, C. JACOBI: Numerical Simulations of Tides, Rossby and Kelvin Waves with the COMMA-LIM Model. Adv. Space Res. Vol.32, No.5, pp. 863-868, 2003.
FRÖHLICH, K., A. POGORELTSEV, C. JACOBI: The 48 Layer COMMA-LIM Model: Model description, new Aspects, and Climatology, Wissenschaftliche Mitteilungen aus dem Institut für Meteorologie der Universität Leipzig, Bd. 30, 157-185, 2003.
FRÖHLICH, K., AND CH. JACOBI: The solar cycle in the middle atmosphere: changes of the mean circulation and of propagation conditions for planetary waves, Rep. Inst. Meteorol. Univ. Leipzig, 34, 106-117, 2004.
FRÖHLICH, K., AND CH. JACOBI: Der solare Zyklus in der mittleren Atmosphäre: Änderungen in der mittleren Zirkulation und den Ausbreitungsbedingungen der planetaren Wellen, Terra Nostra 2003/6: 6. Deutsche Klimatagung, 155-160, 2003.
FRÖHLICH, K., CH. JACOBI, M. LANGE, AND A. POGORELTSEV: The quasi two-day wave - the results of numerical simulation with the COMMA-LIM model, Rep. Inst. Meteorol. Univ. Leipzig, 26, 122 - 134, 2002.
FRÖHLICH, K., A. POGORELTSEV, AND CH. JACOBI: Tides, Rossby and Kelvin waves simulated with the COMMA-LIM model, Rep. Inst. Meteorol. Univ. Leipzig, 30, 149-155, 2003.
GERDING, M., M. RAUTHE AND J. HÖFFNER, Temperature soundings from 1 to 105 km altitude by combination of co-located lidars, and its application for gravity wave examination, G. Pappalardo and A. Amodeo (Eds.), Reviewed and revised papers presented at the 22nd International Laser Radar Conference, ESA SP-561, 567-570, ESA-ESTEC, Nordwijk, Juni 2004.
GROSSMANN, K. U., M. KAUFMANN , AND E. GERSTNER, A global measurement of lower thermosphere atomic oxygen densities, Geophy. Res. Lett., 27 (9), 1387 – 1390, 2000.
GROSSMANN, K. U., O. GUSEV, M. KAUFMANN , A. KUTEPOV, AND P. KNIELING, A review of the scientific results from the CRISTA missions, Adv. Space Res., 34, 1715 – 1721, 2004.
GROSSMANN, K. U., O. GUSEV, AND P. KNIELING, The distribution of carbon monoxide in the upper mesosphere and lower thermosphere during CRISTA-1 and CRISTA-2, J. Atmos. Solar-Terr. Physics, submitted, 2005.
GROSSMANN, K. U., O. GUSEV, M. AIT ASSOU, AND G. WITT, Spectral emission measurements of Polar Mesospheric Clöouds by CRISTA-2, J. Atmos. Solar-Terr. Physics, submitted, 2005.
GUSEV, O.A. AND A. A. KUTEPOV, Non-LTE gas in planetary atmospheres, in book: Stellar Atmosphere Modeling, I. Hubeny, D. Mihalas, and K. Werner, eds., ASP Conference Series Vol. 288, 318, 2003.
HALL , C.M., ASO T., TSUTSUMI M., HÖFFNER, J., SIGERNES, F., Multi-instrument derivation of 90 km temperatures over Svalbard (78°N 16°E), Radio Science, accepted.
HARTOGH, P., MEDVEDEV, A. S., KURODA, T., SAITO, R., VILLANUEVA , G., FEOFILOV, A. G., KUTEPOV A. A., AND BERGER, U., Description and climatology of a new general circulation model of the Martian atmosphere, submitted to J. Geophys. Res., 2005.
HÖFFNER, J., LAUTENBACH, J., BEGEMANN, C.-F., AND P. MENZEL, Observation of temperature, NLC, PMSE and potassium at Svalbard, 78° N, in proc. of the 30th annual
- 28 -
europeanmeeting on atmospheric studies by optical methods, edited by F. Sigernes and D.A. Lorentzen, 65-67, 2003.
HOFFMANN, P., W. SINGER, AND D. KEUER, Variability of the mesospheric wind field at middle and Arctic latitudes , Journ. Atmos. and Solar Terr. Physics,64, 1229-1240, 2002.
JACOBI, CH., K. FRÖHLICH, AND A. POGORELTSEV: Quasi two-day-wave modulation of gravity wave flux and consequences for the planetary wave propagation in a simple circulation model, J. Atmos. Solar-Terr. Phys., accepted, 2005.
JACOBI, CH., AND D. KÜRSCHNER: Long-term trends of MLT region winds over Central Europe, Phys. Chem. Earth, accepted, 2005.
JACOBI, CH., AND D. KÜRSCHNER: Long-term measurements of nighttime LF radio wave reflection heights over Central Europe, Adv. Radio Sci., 3, 427-430, 2005.
JACOBI, CH., K. FRÖHLICH, AND A. POGORELTSEV: Gravity wave flux modulation by planetary waves in a circulation model, Rep. Inst. Meteorol. Univ. Leipzig, 36, 113-124, 2005.
JACOBI, CH., AND D. KÜRSCHNER: Variabilität der mittleren Atmosphäre über Mitteleuropa im Zeitbereich von 2 - 20 Jahren, Terra Nostra 2003/6: 6. Deutsche Klimatagung, 214-221, 2003.
KAUFMANN , M., O. A. GUSEV, K. U. GROSSMANN, R. ROBLE, M. HAGAN, C. HARTSOUGH, AND A. A. KUTEPOV: The vertical and horizontal distribution of CO2 densities in the upper mesosphere and lower thermosphere as measured by CRISTA, J. Geophys. Res., 107, 10.1029/2001JD000704, 2002.
KAUFMANN , M., O. A. GUSEV, K. U. GROSSMANN, F. J. MARTIN-TORRES, D. R. MARSH, AND
A. A. KUTEPOV, Satellite observations of day-and nighttime ozone in the mesosphere and lower thermosphere, J. Geophys. Res., 108, 10.1029/2002JD002800, 2003.
KÜRSCHNER, D., AND CH. JACOBI: The mesopause region wind field over Central Europe in 2003 and comparison with a long-term climatology, Adv. Space Res., in print, 2005.
KÜRSCHNER, D., AND CH. JACOBI: Unexpectedly small semidiurnal tidal wind amplitudes in the midlatitude mesopause region during September 2002, Ann. Geophysicae, 22, 701-704, 2004.
KÜRSCHNER, D., AND CH. JACOBI: Quasi-biennial and decadal variability obtained from long-term measurements of nighttime radio wave reflection heights over central Europe, Adv. Space Res., 32, 1701-1706, doi: 10.1016/S0273-1177(03)00773-2, 2003.
KÜRSCHNER, D., AND CH. JACOBI: A long-term record of sudden phase anomalies at Collm, Rep. Inst. Meteorol. Univ. Leipzig, 34, 89-96, 2004.
KÜRSCHNER, D., AND CH. JACOBI: Long-term behaviour of E-region nighttime LF reflection heights – long-term trend, solar cycle, and the QBO, Rep. Inst. Meteorol. Univ. Leipzig, 30, 127-135, 2003.
LÜBKEN, F.-J., HÖFFNER, J., BEGEMANN, C.-F., MÜLLEMANN , A., ZECHA, M., AND J. RÖTTGER, Mesospheric Layers and Temperatures at Spitsbergen, 78° N, Proc. of the mesospheric clouds workshop, Perth, Scotland, edited by M. Gadsden and N. James, 97-114, London, 2002.
LÜBKEN, F.-J., AND A. MÜLLEMANN , First in-situ temperature measurements in the summer mesosphere at very high latitudes (78°N), J. Geophys, Res., 108, D8, 10.1029/2002JD002414, 2003.
LÜBKEN, F.-J., M. ZECHA, J. HÖFFNER, AND J. RÖTTGER (2004), Temperatures, polar mesosphere summer echoes, and noctilucent clouds over Spitsbergen (78° N), J. Geophys. Res., 109, D11203, doi: 10.1029/2003JD004247.
LÜBKEN, F.-J., AND J. HÖFFNER, Experimental evidence for ice particle interaction with metal atoms at the high latitude summer mesopause region, Geophys. Res. Lett., 31, L08103, doi:10.1029/2004GL019586, 2004
MARSH, D. R., A. SMITH , G. BRASSEUR, M. KAUFMANN , AND K. U. GROSSMANN: The existence of a tertiary ozone maximum in the high-latitude middle mesosphere, Geophys. Res. Let., 28, 4531-4534, 2001.
- 29 -
MERZLYAKOV , E.G., AND CH. JACOBI: Quasi-two-day wave in an unstable summer atmosphere – some numerical results on excitation and propagation, Ann. Geophysicae, 22, 1917-1929, 2004.
MERZLYAKOV , E.G., YU.I. PORTNYAGIN, CH. JACOBI, I. FEDULINA, N.J. MITCHELL, B.L. KASHCHEYEV, A.N. OLEYNIKOV , AND A.H. MANSON: On the day-to-day wind and semidiurnal tide variations at heights of the mid-latitude summer mesopause: Zonal wavenumber estimations and its consequences, case-study in 1998, J. Atmos. Solar-Terr. Phys., 67, 535-551, 2005.
MERZLYAKOV , E., AND CH. JACOBI: Some numerical results on the quasi-two-day wave excitation and propagation in the unstable summer middle atmosphere, Rep. Inst. Meteorol. Univ. Leipzig, 30, 186-200, 2003.
MÜLLEMANN , A., M. RAPP, F.-J. LÜBKEN, AND P. HOFFMANN, In situ measurements of mesospheric turbulence during spring transition of the Arctic mesosphere, Geophys. Res. Lett., 29, 10.1029/2002GL014841, 115-1, 2002.
OBERHEIDE, J., H.-L. LIU, O.A. GUSEV, AND D. OFFERMANN, Mesospheric surfzone and temperature inversion layers in early November 1994, J. Atmos. Solar-Terr. Physics, submitted, 2005.
PANCHEVA, D., N.J. MITCHELL, A.H. MANSON, C.E. MEEK, CH. JACOBI, YU. PORTNYAGIN, E. MERZLYAKOV , W.K. HOCKING, J. MACDOUGALL, W. SINGER, K. IGARASHI, R.R. CLARK , D.M. RIGGIN, S.J. FRANKE, D. KÜRSCHNER, A.N. FAHRUTDINOVA, A.M. STEPANOV, B.L. KASHCHEYEV, A.N. OLEYNIKOV , AND H.G. MULLER: Variability of the quasi-2-day wave observed in the MLT region during the PSMOS campaign of June–August 1999, J. Atmos. Solar-Terr. Phys., 66, 539-565, 2004.
PREUSSE, P., M. Ern, S. D. Eckermann, C. D. Warner, R. H. Picard, P. Knieling, M. Krebsbach, J. M. Russell III, M. G. Mlynczak, C. J. Mertens, and M. Riese, Tropopause to mesopause gravity waves in August: measurement and modeling, J. Atmos. Solar-Terr. Physics, submitted, 2005.
SCHMIDT, H. AND G. P. BRASSEUR, Simulation of the mesospheric ozone response to natural and anthropogenic climate variability, Proceedings of the Quadrennial Ozone symposium, Kos, Greece, 2004.
SCHMIDT, H. ET AL., The HAMMONIA Chemistry Climate Model: Sensitivity of the Mesopause Region to the 11-year Solar Cycle and CO2 Doubling. submitted to Journal of Climate, 2004.
SINGER, W., J. BREMER, W.K. HOCKING, J. WEIß, R. LATTECK, AND M. ZECHA, Temperature and wind tides around the summer mesopause at middle and arctic latitudes, Adv. Space Res., Vol. 31, No. 9, 2055-2060, doi: 10.1016/S0273-1177(03)00228-X, 2003.
SINGER, W., J. BREMER, J. WEIß, W.K. HOCKING, J. HÖFFNER, M. DONNER, P. ESPY, Meteor radar observations at middle and Arctic latitudes Part 1. Mean temperatures, J. Atmos. Sol.-Terr. Phys., 66, 607-616, 2004.
Other References ALLEN, M., J. I. LUNNE, AND Y. L. YUNG, The vertical distribution of ozone in the mesosphere
and lower thermosphere. J. Geophys. Res., 89, 4841-4872, 1984.
BEIG, G. ET AL., Review of mesospheric temperature trends, Reviews of Geophysics, doi:10.1029/2002RG000121, 2003.
CRUTZEN, P. J., Mesospheric Mysteries, Science, 277, 1951-1952, 1997.
FOMICHEV, V. I., J.-P. BLANCHET, AND D. S. TURNER : Matrix parameterization of the 15 um CO2 band cooling in the middle and upper atmosphere for variable CO2 concentration, J. Geophys. Res.,103, 11505 – 11528, 1998.
- 30 -
FRITTS, J. ISLER, AND Y. PORTNYAGIN: Global study of northern hemisphere quasi-2-day wave events in recent summers near 90 km altitude. J. Atmos. Terr. Phys., 58(13), 1401 - 1411, 1996.
JACOBI, C.: On the solar cycle dependence of winds and planetary waves as seen from mid-latitude D1 LF mesopause region wind measurements. Ann. Geophysicae, 16, 1534 - 1543, 1998.
LÜBKEN, F.-J., Thermal structure of the Arctic summer mesosphere, J. Geophys. Res., 104, 9135 - 9149, 1999.
LÜBKEN, F.-J., M. J. JARVIS, AND G. O. L. JONES, First in situ temperature measurements at the Antarctic summer mesopause, Geophys. Res. Lett., 26, 3581 - 3584, 1999.
MIHALAS , D., Stellar Atmospheres, Second Edition, Freeman, San Francisco, 632pp., 1978.
MEEK, C., A. MANSON, S. FRANKE, W. SINGER, P. HOFFMAN, R. CLARK , T. TSUDA, T. NAKAMURA , M. TSUTSUMI, M. HAGAN, D. PALO, S.E., R. ROBLE, AND M. HAGAN: Middle atmosphere effects of the Quasi-two-day wave determined from a General Circulation Model. Earth Planets Space, 51, 629 - 647, 1999.
PANCHEVA, D., BEARD, A., MITCHELL, N., AND MULLER, H.: Nonlinear interactions between planetary waves in the mesosphere/lower-thermosphere region. J. Geophys. Res., 105(A1), 157 - 170, 2000.
POOLE, L., AND T. HARRIS: The propagation of the mesospheric two-day wave in the southern hemisphere. J. Atmos. Terr. Phys., 57, 1661 - 1666, 1995.
ROECKNER, E. ET AL., The atmospheric general circulation model ECHAM 5. PART I: Model description, Rep. 349, MPI for Meteorology, Hamburg, Germany, 2003.
SWARZTRAUBER, P. AND A. KASAHARA: The vector harmonic analysis of Laplace's tidal equations, SIAM J. Sci. Stat. Comput., 6, 464 - 491, 1985.
TEITELBAUM , H. AND F. VIAL : On tidal variability induced by nonlinear interaction with planetary waves, J. Geophys. Res., 96,14169 - 14178, 1991.
