sign in sign up
<<
Home , Scale height

Aeolian Research 3 (2011) 145–156

Contents lists available at ScienceDirect

Aeolian Research

journal homepage: www.elsevier.com/locate/aeolia

Review Article Influence of dust on the dynamics of the martian above the first scale height ⇑ Alexander S. Medvedev a, , Takeshi Kuroda a,b, Paul Hartogh a a Max Planck Institute for Solar System Research, Katlenburg-Lindau D-37191, Germany b Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, Sagamihara 252-5210, Japan article info abstract

Article history: Dust suspended in the martian atmosphere strongly affects the radiative transfer. Diabatic heating and Received 30 November 2010 cooling it creates are prominent factors that drive the atmosphere at various scales. This paper provides Revised 16 May 2011 a review of dust influence on the large-scale dynamics in the atmosphere of above approximately Accepted 16 May 2011 10 km. We outline the established properties of dust that influence the diabatic heating/cooling rates, and summarize the current knowledge of dust-related effects on the zonal-mean circulation and zonally asymmetric disturbances: planetary waves and tides. Keywords: Ó 2011 Elsevier B.V. All rights reserved. Mars atmosphere Atmospheric aerosol Dust storms General circulation

Contents

1. Introduction ...... 145 2. Basic facts about the martian atmosphere ...... 146 3. Properties of the airborne dust ...... 146 3.1. Seasonal and spatial distributions ...... 146 3.2. Vertical distributions...... 147 3.3. Particle size distributions ...... 148 3.4. Refractive indices ...... 148 4. Approach for analyzing the global circulation ...... 149 5. Influence of dust on the zonal-mean circulation ...... 149 5.1. Dust storm effects on ...... 149 5.2. Global meridional circulation ...... 150 5.3. Winter polar warming ...... 151 5.4. Atmosphere superrotation ...... 152 5.5. Equatorial semiannual oscillations...... 153 6. Influence of dust on non-zonal disturbances ...... 153 6.1. Thermal tides...... 153 6.2. Traveling planetary waves ...... 153 7. Summary and conclusions ...... 154 Acknowledgments ...... 154 References ...... 155

1. Introduction strongly radiatively active, dust absorbs solar radiation and emits in the infrared. The created local heating and cooling affects the Airborne dust suspended in the dry plays a atmospheric dynamics at various scales: from synoptic and role in many ways similar to the role of water on . Being meso-scale weather systems (martian meteorology) to the large- scale circulation and climate. A ‘‘dust cycle’’ on Mars with lifting from the surface, transport, and sedimentation back to reservoirs ⇑ Corresponding author. Tel.: +49 5556 979 314; fax: +49 5556 979 240. E-mail address: [email protected] (A.S. Medvedev). resembles the ‘‘water cycle’’ on Earth. Many martian weather

1875-9637/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.aeolia.2011.05.001 146 A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156 phenomena are associated with dust: storms from local to regional the diurnal and seasonal variations are similar as well. Because and global ones, dust devils (Balme and Greeley, 2006), ‘‘cells’’ martian atmosphere is less dense, the difference of temperature (Cantor et al., 2002), clouds (Clancy et al., 2003), and plumes (Fuer- between day and night is significantly larger. For instance, the stenau, 2006). These similarities with the terrestrial hydrometeo- atmospheric temperature near the surface varies between approx- rology prompted Heavens et al. (2011b) to introduce the term imately 200 and 260 K at the summer subtropics, according to the ‘‘coniometeorology’’ (from the Greek word konios for ‘‘dust’’) to Mars Pathfinder observations (Schofield et al., 1997). Due to the characterize the martian weather. colder , the atmospheric component of CO2 con- This paper is a review of the current knowledge of dust effects denses at the polar regions during winters (e.g., James et al., on the global-scale atmospheric dynamics of Mars. Most of meteo- 1992), and the averaged surface pressure on Mars varies annually rological phenomena are associated with the lower atmosphere within the range of 25%. Although the atmospheric air is thin, where the influence of the surface is strong. In the dynamical winds on Mars are strong. They can lift in the air millions of tons meteorology of Earth, the definition of the ‘‘lower’’ atmosphere is of dust particles during planetary-scale dust storms, which make vague, and most often means the troposphere. The martian tropo- the atmosphere opaque. sphere, an atmospheric layer with vertically decreasing tempera- Martian dust storms have been observed since the end of the ture, extends much higher, to 40–60 km. The martian middle 19th century. A comprehensive list of such activities from 1873 atmosphere lies above, and covers the altitudes to 110–130 km. to 1990 is presented in Table 1 of Martin and Zurek (1993). Most In this survey, the emphasis is placed on the layer which is little of the global-scale dust storms occurred near or before southern directly affected from below by convection, and, from above, by summer solstice, which is close to the perihelion. They started as heating due to absorption of extreme ultraviolet and energetic regional storms at elevated plateaus located in low- and mid-lati- particles. It corresponds to the altitudes between 10 and 110– tudes, and then expanded in the east-west direction to encircle the 130 km, that is, coincides with the troposphere above approxi- planet in 10–20 days, and to last for 50–100 days before decaying. mately one scale height and the middle atmosphere of Mars. Many regional dust storms have been observed at the edge of Therefore, we leave out of scope a large body of studies, which receding south polar cap (Briggs et al., 1979), and in northern concern mechanisms of dust lifting and interactions with synop- mid-latitudes during equinoxes (James et al., 1999), all of them tic-scale disturbances, and focus on effects of the dust already being associated with frontal systems. Various mechanisms of dust injected in the air. expansion processes, such as ‘‘dust hurricanes’’ (Gierasch and The most spectacular phenomena are the dust storms that reg- Goody, 1973), diurnal Kelvin tidal mode (Zurek and Leovy, 1981), ularly reach planetary scales. They sometimes obscure almost all have been discussed. In the recent decade, the processes of dust the surface of the planet. Earlier observations of these storms on lifting and transport have been studied extensively using martian Mars started in the beginning of the 20th century with ground- general circulation models (GCMs) (Newman et al., 2002a,b; Wang based telescopes (Briggs et al., 1979, and references therein). Since et al., 2003, 2005; Basu et al., 2004, 2006; Kahre et al., 2006, 2008). the 1970s, landers and spacecraft inserted into the Mars orbit pro- vided much more details. Orbital cameras and infrared spectrome- 3. Properties of the airborne dust ters mapped the spatial and seasonal changes of the dust opacity (Thorpe, 1979, 1981; Martin, 1986). It was found that the storms 3.1. Seasonal and spatial distributions occur systematically when Mars is near perihelion. Even during the clearest parts of the year, the dust optical depth remains rela- Martian atmospheric dust consists of mineral particles eroded tively high: s = 0.2–0.4 in visible wavelengths. Absorption of solar from rocks. It is continuously supplied from the martian surface radiation by dust is comparable with that of CO gas, which the 2 owing to two physical processes: the near-surface wind stress atmosphere of Mars mainly consists of. Thus, while heating/cooling (e.g., Greeley and Iversen, 1985), and small-scale convective vorti- by CO maintains the martian climate, atmospheric dust deter- 2 ces (‘‘dust devils’’) clearly observed from the Mars Exploration Ro- mines a variability of the circulation and weather, very much like ver Spirit (Greeley et al., 2006). The amount of airborne aerosol is the water on Earth. highly variable with most of the dust storm activity occurring in We begin with the overview of main features of Mars and mar- southern springs and summers. Because of the large orbit eccen- tian atmosphere in Section 2, and then consider dust properties and tricity (0.0934), the planet receives 1.3 times more of solar radia- observational constrains that control the dust-induced diabatic tion near the perihelion, which almost coincides with the heating and cooling of the air in Section 3. An approach to analyzing southern summer solstice (aerocentric longitude L = 270°). There- the global-scale circulation is outlined in Section 4. The influence of s fore, the lower atmospheric convection and meteorological pro- dust on the zonal-mean circulation and zonally-asymmetric eddies cesses are more violent during this season compared to northern are discussed in Sections 5 and 6, correspondingly. springs and summers. The scales of dust storms that occur in this period vary from year to year, and range from regional to planet- wide. In the southern spring and summer, the observed maxima 2. Basic facts about the martian atmosphere of the mean dust opacity over the equator are between 0.25 and

Mars is the second best-studied planet after the Earth. It is often Table 1 called a ‘‘terrestrial-like’’ planet. Comparison of main physical Comparison of physical parameters of Mars and Earth. parameters for Mars and Earth are given in Table 1. Martian radius is about a half, and acceleration is 0.38 that of Earth. Parameters Mars Earth Unit However, its surface area is only slightly less than the total area Radius of planet 3397 6378 km 2 of Earth’s dry land. The martian atmosphere is very thin (the sur- Acceleration of gravity 3.72 9.8 m s Rotational period 88,775 86,400 s face pressure is about 6 mbar, which is more than 150 times smal- Amount of solar radiation 589.2 1367.6 W m2 ler than on Earth), and very dry (there is virtually no water except Sols per a year 669 365 sol in subtropics in the northern summer and in polar regions). It con- Eccentricity 0.0934 0.0167 – Angle of equator inclination 25.19 23.45 ° sists mainly of CO2 (95.3%) with other important constituents being N (2.7%), Ar (1.6%), O (0.15%) and H O (0.03%). Since the Mean surface pressure 6 1013 mbar 2 2 2 Mean surface temperature 230 288 K rotation period and orbit inclination of Mars and Earth are similar, A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156 147

1 Fig. 1. Latitude–time cross-sections of 1075 cm zonal-mean column dust opacity: (top panel) MGS–TES observations from Ls = 120° of MY24 through Ls =80° of MY27, after

Smith (2006); (bottom panel) THEMIS observations from Ls = 330° of MY25 through Ls = 190° of MY29, after Smith (2009).

1.4 in infrared (9 lm) wavelength (Liu et al., 2003), depending on with p0 = 6 mbar being the global mean surface pressure. Both the scales and strength of the storms. In contrast, a relatively low expressions have been extensively used by GCM modelers for pre- dust opacity and temperature persist in northern springs and scribing dust vertical profiles and calculating the corresponding summers, and exhibit high year-to-year repeatability. The heating rates (e.g., Pollack et al., 1990; Wilson and Hamilton, measurements from Mariner 9 Infrared Interferometer Spectrome- 1996; Moudden and McConnell, 2005). The preset values of m varied ter (IRIS), Viking Infrared Thermal Mapper (IRTM), Thermal in their simulations from 0.007 in Forget et al. (1999) to 0.01 in Emission Spectrometer onboard Mars Global Surveyor (MGS–TES) Wilson and Hamilton (1996) and Moudden and McConnell (2005) and Thermal Emission Imaging System onboard Mars Odyssey to 0.03 in Pollack et al. (1990). Observations, however, indicate that (ODY–THEMIS) show that zonal-mean 9 lm dust opacities at the dust mixing ratios change with height in a more complex manner. equator were around 0.05 throughout the aphelion season at all In particular, aerosols always extend higher in the equatorial zone, years of observations (Liu et al., 2003; Smith, 2004, 2009). A mul- and decay more abruptly towards the poles in both hemispheres ti-year record of zonally-averaged column dust opacity is shown in (Anderson and Leovy, 1978; Jaquin et al., 1986). To account for this Fig. 1. It presents a compilation of data obtained with two instru- behavior, Forget et al. (1999) adopted a modified version of (3), ments: MGS–TES (Smith, 2006) and ODY–THEMIS (Smith, 2009). which was then utilized in many other GCMs (Lewis and Read, The record illustrates a wide interannual variability of the dust 2003; Kuroda et al., 2005; Hartogh et al., 2005): load during the perihelion season. Dust events differ by the timing ()"# 70 of their onsets, duration, strength, latitudinal extent. Particular p zmax Q ðzÞ¼Q exp m 1 0 ; p < p ; ð4Þ strong dust storms are seen in MY25 and MY28. Note also a local d 0d p 0 but consistent dust activity around the edge of the southern polar cap throughout almost the whole martian year. where the altitude (in km) of the top of the dust layer, zmax, varies with latitude, /, and season, Ls as 3.2. Vertical distributions 2 zmaxð/; LsÞ¼60 þ 18 sinðLs 160 Þ22 sin / ð5Þ The first vertical profile of atmospheric dust on Mars based on Mariner 9 spacecraft limb observations has been reported by Leovy Recently, a new wealth of data on vertical dust distribution has et al. (1972). Seasonal and latitudinal changes of cut-off heights of been obtained from Mars Climate Sounder onboard Mars Recon- aerosol distributions have also been obtained from Mariner 9 naissance Orbiter (MRO–MCS) (Kleinböhl et al., 2009). It has been (Anderson and Leovy, 1978). Jaquin et al. (1986) and Korablev observing the martian atmosphere since 2006 by performing limb et al. (1993) presented the vertical profiles measured from Viking measurements with global coverage from the surface up to 80 km. Orbiter and Phobos spacecraft, respectively. These distributions Its data significantly complement the limb sounding data of MGS– have been approximated by an analytical formula suggested by TES (McConnochie and Smith, 2008; McConnochie et al., 2009). Conrath (1975), who assumed that the vertical mixing of particles Fig. 2 presents zonally averaged height–latitude cross-sections of is determined by the effective diffusivity and gravitational settling: dust for different seasons from McCleese et al. (2010). It demon- nohi strates that the vertical distributions are more complex, and differ z Q ðzÞ¼Q exp m 1 exp ; ð1Þ from those previously observed. The mixing ratios have peaks at d 0d H 15–25 km over the tropics during much of northern springs and where Qd(z) is the dust mass mixing ratio as a function of height z, summers, a so-called ‘‘high altitude tropical dust maximum’’ (HAT- Q0d is the value of Qd at z =0,H is the scale height, and m is the ratio DM) (Heavens et al., 2011a,b). This HATDM structure neither is of the characteristic diffusion and gravitational settling times. The consistent with the parameterizations used to drive GCM simula- latter controls the dust cut-off altitude. This formula in pressure tions, nor has been reproduced by models which include interac- coordinates, p, has the form tive lifting and transport of dust (e.g., Richardson and Wilson, 2002; Kahre et al., 2006). The mechanism of the ATDM formation p0 Q dðzÞ¼Q 0d exp m 1 ; p < p ð2Þ in non-dusty seasons is not yet understood. Heavens et al. p 0 (2011b) discussed possible roles of topography, convective effects Q dðzÞ¼Q 0d; p P p0; ð3Þ associated with dust devils penetrating across the boundary layer, 148 A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156

2 1 Fig. 2. Log10 of the zonal-mean dust -scaled opacity (m kg ) observed by the MRO–MCS. After McCleese et al. (2010). and scavenging of dust by water ice. To date, recognizing the origin with shorter wavelengths. Thus, the ratio of opacities, an important of the HATDM, and its possible dynamical importance remain chal- parameter that affects the radiative heating/cooling, is a function lenging topics for investigations of the dust cycle. of particle sizes and their distributions.

3.3. Particle size distributions 3.4. Refractive indices The size of dust particles is a key optical parameter that deter- Dust refractive indices are another set of optical parameters mines absorption and scattering of solar radiation, infrared emis- that are important for radiative calculations. Presently, most mar- sion, and, therefore, atmospheric heating/cooling rates. Measur- tian GCMs adopt the set of indices obtained by Ockert-Bell et al. ements of dust particle sizes have been performed with Mariner 9 (1997) (for wavelength shorter than 5 lm), Toon et al. (1977) (be- IRIS (Toon et al., 1977), Viking Lander cameras (Pollack et al., 1977, tween 5 and 17 lm), and Forget (1998) (longer than 17 lm). 1979, 1995), Phobos spacecraft (Drossart et al., 1991; Korablev Ockert-Bell et al. (1997) inferred the refractive properties at four et al., 1993), and the Imager for Mars Pathfinder (IMP) (Tomasko visible wavelengths (0.5–0.86 lm) from the particle size distribu- et al., 1999; Markiewics et al., 1999). These distributions have been tion, shape and single-scattering properties reported by Pollack analytically approximated either by the modified gamma or log– et al. (1995). They also extended the coverage to all solar wave- normal functions. An example of the former type has the form (Han- lengths (0.2–4.2 lm) using ground-based telescope data (Owen sen and Travis, 1974): and Sagan, 1972; Bell et al., 1990; Rousch et al., 1992), and 13meff Phobos-2 ISM data (Mustard et al., 1993) at 0.77–2.9 lm. At infra- m r nðrÞ¼Cr eff exp ; ð6Þ red wavelengths, Toon et al. (1977) compared the Mariner 9 IRIS reff meff spectra at 5–50 lm with the terrestrial mineralogical samples where n(r) is the distribution as a function of the particle radius, r; and found that montmorillone 219b, a clay mineral sample with reff and meff are the effective radius and effective variance, respec- at least 60% of SiO2, has the best fit for the wavelengths shorter tively, and C is the normalization constant for adjusting the total than 15 lm. For longer waves, Forget (1998) introduced a ‘‘syn- size distribution to the number . Tomasko et al. (1999) thetic’’ model to fit the Mariner 9 IRIS spectra. determined reff as 1.6 ± 0.15 lm and meff as 0.2–0.5, which are con- Wolff and Clancy (2003) suggested a further update for wave- sistent with the earlier measurements. lengths between 0.2 and 135 lm based on the set described above Wolff and Clancy (2003) and Clancy et al. (2003) reported the and on iterative adjustments using a variety of the MGS–TES mea- spatial and time variations of reff estimated from the MGS–TES surements. Their distribution was mainly based on a Hawaiian pal- spectral data set. They showed that the values of reff were between agonite sample (Clancy et al., 1995), including a power-law 1.3 and 3 lm at different latitudes and longitudes near the peak of extrapolation in the wavelengths longer than 35 lm, and on fitting the global dust storm in 2001 (Ls = 213° of the Martian Year 25, to the estimates from the IMP observations (Tomasko et al., 1999). MY25), and were smaller (1 lm) in the northern spring and sum- Wolff et al. (2006) updated the refractive indices in infrared wave- mer. The effective particle radius tends to be larger when the dust lengths using the upward-viewing geometry of the Miniature Ther- opacity is high, and the ratio of opacities in solar and infrared mal Emission Spectrometer (Mini-TES) on Mars Exploration wavelengths (a ‘‘visible-to-infrared ratio of opacities’’) decreases Rovers. Recently, Wolff et al. (2009) updated the refractive indices when reff grows (Clancy et al., 2003, 2010; Kahre et al., 2008; Elteto in visible to near-infrared wavelengths using the spectra observed and Toon, 2010). This behavior can be explained from the Mie by the Compact Reconnaissance Imaging Spectrometer (CRISM) theory using the Viking Lander and Viking Orbiter data (Toigo onboard Mars Reconnaissance Orbiter during a planet-encircling and Richardson, 2000), and with GCM simulations employing dust storm in 2007. Fig. 3 shows the refractive indices from interactive dust transport for particles with several sizes. Larger- (Ockert-Bell et al., 1997; Toon et al., 1977; Forget, 1998) (denoted size particles produce stronger forward scattering, or weaker as ‘‘Refractive A’’), and from (Wolff and Clancy, 2003; Wolff et al., extinction by scattering, and this tendency increases for radiation 2006, 2009). Imaginary parts of the refractive index represent the A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156 149

Fig. 3. (a) Real and (b) imaginary parts of refractive indices of martian dust as functions of the wavelength. The values of Wolff et al. (2009) are for reff = 1.6 lm. effects of absorption. It is seen that they are larger for the ‘‘Refrac- ‘‘non-migrating’’) tidal modes. Particularly strong on Mars are east- tive A’’ set in ultraviolet and visible wavelengths (0.2–0.5 lm), and, ward propagating diurnal (s = 1) Kelvin waves, which arise from thus, produce higher daytime temperatures when used in radiative interactions of the diurnal tide with s = 2 topography variations calculations. (Wilson, 2000; Forbes and Hagan, 2000). Flow over topography with non-uniform surface thermal inertia 4. Approach for analyzing the global circulation generates stationary planetary waves (SPW). These disturbances with observed phase velocities close to zero (but with non-zero Having discussed the main properties of the dust suspended in speed with respect to the mean flow) can penetrate high into the the air, we now turn to the effects it produces in the atmosphere of middle atmosphere if propagation conditions, determined mainly Mars. A convenient way of analyzing dynamics of planetary atmo- by the mean zonal wind, are favorable. As on Earth, larger-scale spheres is the concept of wave–mean flow interactions. According martian SPWs with zonal wavenumbers s = 1 and 2 propagate ver- to it, all field variables (temperature, pressure, density, wind) are tically, whereas smaller-scale ones are mostly trapped near the subdivided into the ‘‘mean’’ (spatial, temporal, or combination of surface (Banfield et al., 2003; Hinson et al., 2003). both), and deviations. For fast rotating planets, like Earth or Mars, Traveling planetary waves (PWs) are generated by baroclinic an obvious choice is the separation of the variables into symmetric and/or barotropic instabilities of the mean zonal flow. They have with respect to the rotational axis (or zonal-mean) ones, and non- a broad nomenclature with variety of periods longer than one zonal disturbances (or eddies). Global-scale distribution of axisym- sol, horizontal wavelengths, and non-zero phase velocities. The metric fields is often called the ‘‘general circulation’’. Due to the spectrum of traveling PWs consists of various well-defined har- non-linearity of the Navier–Stokes equations governing the monics with certain dominating modes at each time and the tran- dynamics, the general circulation is then driven not only by zonally sition between them (Hinson, 2006). In the lower atmosphere, they averaged diabatic heating and cooling, but by covariances of eddy are associated with baroclinic eddies or weather patterns, which fields as well. The latter is called an ‘‘eddy’’ or ‘‘mechanical’’ forcing can give rise to flushing dust storms (Wang et al., 2003; Basu of the circulation. et al., 2006; Hinson and Wang, 2010). Under appropriate condi- In convectively stable , many disturbances behave tions, traveling PWs can propagate into the upper atmosphere like waves. In other words, they (1) are oscillatory motions that (Seth and Rao, 2008), where they contribute to the mechanical possess a high degree of coherence, (2) propagate horizontally forcing of the mean circulation. and/or vertically with time, (3) transport energy and momentum, and (4) deposit them to the mean flow upon dissipation. The most 5. Influence of dust on the zonal-mean circulation prominent large-scale atmospheric waves on Mars are thermal tides and planetary waves. 5.1. Dust storm effects on temperature Solar thermal tides are oscillations of winds, temperature, den- sity, and pressure induced by diurnally varying absorption of solar In their pioneering work, Gierasch and Goody (1972) first indi- radiation in the atmosphere and/or surface. They are particularly cated that the observed vertical temperature profiles could not be strong on Mars due to lower than on Earth density, and larger reproduced in a radiative–convective model without taking into heating rate per unit mass. The solar forcing of tides increases account the absorption of solar radiation by dust. The theoretical when the amount of dust in the air rises. Westward propagating studies that followed (Moriyama, 1974, 1975; Zurek, 1978) have Sun-synchronous components carrying westward momentum are demonstrated that the radiative effects of dust are important even the direct dynamical response to this forcing. Due to the non-line- when the dust layer is optically thin. The extent to which dust arity of atmospheric motions, tidal harmonics with various zonal storms modify the atmospheric temperature is illustrated in wavenumbers, s, are generated. Vertical and meridional (South– Fig. 4. It compares the height–latitude cross-sections of the North) propagation of the tides is very sensitive to the mean wind. zonal-mean daytime temperature measured by MGS–TES and

The diurnal component with s = 1 can propagate vertically in averaged between Ls = 205° and 210° during Martian Years 24 tropics and poleward in mid-latitudes aloft, but tends to be and 25 (MY24 and MY25). A global dust storm occurred in MY25 trapped at low levels. The semi-diurnal Sun-synchronous compo- at this time (see Fig. 1). As a result, the temperature has risen by nent (s = 2) has longer vertical wavelength, little vertical phase 40 K at pressure levels 0.2–0.5 mb (15–25 km) over the equator, propagation, and depends less on seasons. Variations of topogra- and more than 60 K over the Sun-lit South pole. This warming phy and surface thermal properties can interact non-linearly with has been traced up to 0.01 mb (60 km) (Gurwell et al., 2005). the solar forcing to produce non-Sun synchronous (or Conversely, the amplified dust content blocks solar radiation, and 150 A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156

Fig. 4. Mean daytime (2 pm) temperature measured by MGS–TES between Ls = 205° and 210° during Martian Years (a) 24 and (b) MY25, and (c) the corresponding temperature difference. A major dust storm occurred around this time in MY25. the near-surface temperature drops by 10 K. Such temperature Wilson (1997) compared the zonally-averaged meridional mass changes alter the convective, baroclinic and barotropic stabilities transport stream function (similar to the one in Fig. 5) with sur- of the atmosphere, transform wind shears related to meridional faces of the absolute angular momentum. He found that they are temperature gradients, and, thus, produce a dynamical feedback. roughly parallel over a large region in the atmosphere below 0.1 mb (or 40 km). This implies that the meridional circulation 5.2. Global meridional circulation is consistent with the almost inviscid Hadley circulation proposed by Schneider (1983) at these altitudes. In contrast, higher and clo- A typical solstitial distribution of zonal mean temperature sim- ser to the poles, significant deviations from the conservation of ulated with a GCM is plotted in Fig. 5. Since the subsolar point, and, angular momentum have been found. Together with the well doc- thus, the maximum of solar heating are shifted to the southern umented departure of the middle atmosphere from the radiative (summer) hemisphere, the temperature decreases almost mono- equilibrium, especially over the winter pole, this means that a tonically toward the north. In an absence of eddies, such differential dynamical forcing by zonally-asymmetric disturbances (eddies) heating allows for a development of a thermally direct meridional must be invoked. cell known as a ‘‘solstitial Hadley circulation’’ (Schneider, 1983). It The effect of eddies on the mean circulation enters the mean includes rising and sinking air over the warmer and colder hemi- momentum equation in the Transformed Eulerian Mean formula- spheres, correspondingly, a poleward flow due to the conservation tion through the divergence of the Eliassen–Palm (EP) fluxes, F, of mass, and a return flow near the surface (trade winds). This kind (Andrews et al., 1987) in the right-hand part: of transport cell, although confined to tropics, is observed in the ter- 1 1 restrial troposphere. The reason for this confinement is that there is ut þ v ½ða cos /Þ ðu cos /Þ/ f þw uz ¼ðq0a cos /Þ r F: ð7Þ no global summer-to-winter-hemisphere temperature gradient be- cause of the very high heat capacity of the oceans. The axisymmet- In (7) and everywhere in this paper, overbars and primes denote ric Hadley circulation is based on the conservation of absolute longitudinal averages and deviations from the mean, correspond- angular momentum. This implies a negligible role of non-zonal dis- ingly; ðu; v; w Þ are the components of the zonal-mean wind in turbances and their effects, and, therefore, the circulation is some- (x,y,z) direction, respectively; v and w are the residual meridional 1 0 0  times called ‘‘almost inviscid’’. Being purely thermally driven, the and vertical velocities defined as v ¼ v q0 ðq0v h =hzÞz and 1 0 0  Hadley cell is very sensitive to details of heating. It has been argued w ¼ w þða cos /Þ ðcos /v h =hzÞ/. It is seen, that the net transport (Wilson, 1997; Forget et al., 1999) that on Mars, the circulation of represented by the residual velocities is the sum of the Eulerian this kind may extend higher and across the globe due to the large mean ðv; w Þ and the Stokes drift induced by eddies. In the rest of pole-to-pole thermal contrast. the notations: / is the latitude, h = Texp(Rz/cpH) is the potential The mass-weighted stream function presented in Fig. 5a and b temperature, T is the temperature, R is the gas constant, cp is the (taken from GCM simulations of Kuroda (2006)) illustrates the specific heat, f =2Xsin/ is the Coriolis parameter, where X is angu- changes in the meridional circulation caused by an increased lar velocity, 2p divided by the martian rotational period. It is seen amount of airborne dust. During a relatively dustless season of that the divergence of the EP flux (or wave action flux) r F repre- MY24 (TES2 scenario with s 0.2), the direct thermal cell extends sents a forcing for the zonally averaged components of the wind. F from 30°Sto30°N in the lower atmosphere, and from pole to pole arises entirely due to the non-linearity of the Navier–Stokes equa- above 20 km. When the dust load dramatically increases to tion when the dynamical variables are split into mean and devia- almost s = 4.2 in visible (VIK1 scenario, MY12), the meridional cir- tions. In particular, the EP flux depends on covariations between culation intensifies and broadens. The transport cell expands high- wind and temperature fluctuations and their gradients. It follows er into the atmosphere and extends farther to the winter pole from (7) that the eddies not only contribute to the acceleration or (Fig. 5b). The corresponding zonal-mean temperature distributions deceleration of u, but force the meridional transport in the time– are shown in Fig. 5c and d with color shades. It is seen that more average sense (when @/@t ? 0). dust in the air produces more heating aloft. Consequently, less Using GCM simulations, Wilson (1997) pointed out to a very heating and lower temperature occur near the surface, where solar strong EP flux divergence in the middle atmosphere where the radiation is weakened being absorbed by the dust. The meridional meridional stream functions deviate from absolute momentum temperature gradient increases in the Sun-lit summer hemisphere, surfaces, especially, over the winter pole and during dust storms. thus maintaining stronger Hadley circulation. In turn, the axisym- Hartogh et al. (2005) and Medvedev and Hartogh (2007) showed metric meridional winds induce zonal mean westerlies in the win- that the residual meridional velocity v is determined largely by ter hemisphere via the Coriolis force, and easterlies in the summer the EP flux divergence at all seasons between 10 and 70 km, at hemisphere and over the tropics (Fig. 5c and d, contours). least outside the tropics (30° to +30°). This suggests that the A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156 151

Fig. 5. (a and b) Mass stream functions simulated for the ‘‘low dust’’ (s = 0.2) and ‘‘dust storm’’ (s = 4.2) scenarios, respectively. Air parcels move along the isopleth so that smaller values remain on the right. (c and d) The corresponding temperature (shaded) and mean zonal wind (contours). After Kuroda (2006).

meridional circulation on Mars, at least in extratropics, is forced however, apparently, underestimated their magnitudes. Wilson primarily by eddies (planetary waves, tides, and, perhaps, smal- (1997) was the first to demonstrate that an increase of diabatic ler-scale gravity waves) similarly to the so-called ‘‘extratropical heating during dust storms enhances the circulation, and leads to pump’’ mechanism (Holton et al., 1995), as on Earth. significant warmings. Further modeling studies corroborated this result (Forget et al., 1999; Moudden and McConnell, 2005). 5.3. Winter polar warming Medvedev and Hartogh (2007) explored the dependence between resolved non-zonal disturbances in the GCM, and the simulated A remarkable manifestation of the eddy-driven meridional residual meridional velocity and winter polar temperature. They transport in the middle atmosphere of Mars are temperature inver- argued that the maxima can be 10–30 K warmer compared to sions and related warmings over the winter poles. They are persis- the values predicted by other models (e.g., Forget et al., 1999; tent features in both hemispheres, which usually occur between 50 Lewis et al., 1999), if simulated tides and planetary waves are con- and 80 km, and have magnitudes from several to tens Kelvin de- sistent with observations and not artificially inhibited in GCMs. Re- grees (Deming et al., 1986; Théodore et al., 1993; Santee and Crisp, cently, temperature retrievals extending to 80 km have been 1993). Earlier observations have pointed out to a coincidence be- performed using the MRO–MCS instrument (McCleese et al., tween the onset of dust storms and polar warmings (Jakosky and 2008). They revealed intense polar temperature inversions in the Martin, 1987). Since there is no or little of insolation in high winter middle atmosphere (Fig. 6) with magnitudes, which are in a good latitudes, diabatic heating alone cannot cause this effect, even if agreement with the model predictions of Hartogh et al. (2005) the dust load is high. It was recognized (e.g., Barnes and Haberle, and Medvedev and Hartogh (2007). This implies more vigorous 1996; Wilson, 1997, and references therein) that polar warmings than previously expected meridional transport, v, and, therefore, are associated with the global transport cell, and produced by stronger eddy forcing. the descending air entering denser atmosphere. Strong downward What kind of atmospheric waves contribute most to the EP flux motions require strong poleward flow aloft. Thus, the temperature divergences, and why they enhance during global dust storms? To in the middle atmosphere above the polar cap can help to estimate date, the MCS measurements were made over the seasons with rel- the intensity of the meridional transport, but until recently, most atively low atmospheric dust amount (the data for the 2007 major such measurements were restricted to lower altitudes. dust storm were unfortunately missing except for very small Many martian GCMs with sufficiently high model tops repro- regions near the North pole), and, therefore, cannot constrain duced these temperature reversals in high winter latitudes, and/or validate GCM simulations for dust storms. Barnes and 152 A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156

Fig. 6. A cross-section of the atmospheric temperature in the southern hemisphere of Mars for mid-southern winter (Ls = 136°) (shaded) composed from successive MRO– MCS limb retrievals that extend from the nightside to the dayside (left-to-right). After McCleese et al. (2008).

Hollingsworth (1987) discussed the potential causation of sudden the EP flux divergence that drives the lower thermosphere polar polar warmings by breaking/dissipation of upward propagating warming. stationary planetary waves. This mechanism is prominent in the Earth stratosphere, and is usually accompanied by a complete 5.4. Atmosphere superrotation breakdown of the winter polar vortex (Andrews et al., 1987). Small-scale gravity waves (GWs) can penetrate high into the mid- We discussed above the main features of the solstitial circula- dle atmosphere and impose a significant wave drag on the zonal tion in the atmosphere of Mars. Over equinoxes, the subsolar point flow upon their breaking and/or saturation. GWs alone are respon- moves to the tropics giving rise to almost symmetric with respect sible for the temperature inversion and the zonal wind reversal in to the equator temperature distribution in the lower atmosphere. the terrestrial mesosphere. Effects of orographically generated Being in transition between two solstices, the equinoctial zonally GWs in the martian atmosphere were studied by Barnes (1990), averaged circulation consists of two meridional cells with rising Joshi et al. (1995), Collins et al. (1997). In simulations of Wilson motions over the equatorial region and the downward flow over (1997), the EP flux divergence was dominated by the contribution the poles. This structure reproduced by all martian GCMs extends of thermal tides. With the increase of solar radiation absorption up to at least 70 km or higher. As a result of the Coriolis force, two during dust storms, the diurnal variations of temperature are westerly (prograde) jets are maintained by the meridional cells in stronger in the atmosphere and surface, and tidal amplitudes are mid- and high-latitudes in both hemispheres. In low-latitudes, larger (Wilson and Hamilton, 1996). Kuroda et al. (2009) analyzed where the Coriolis force and the cross-equatorial flow are weak, the eddy forcing of the vigorous poleward and downward trans- the zonal wind is more influenced by eddies (see Eq. (7)). In the ab- port during a planet-encircling dust storm in detail. They showed sence of forcing associated with non-axisymmetric disturbances that the warming is caused by dissipating thermal tides, planetary zonal wind remains purely easterly (retrograde) (Hide, 1969). and resolved-scale GWs in almost equal degree. In their simula- Thus, the superrotation, defined as a ratio of the positive (pro- tions, one third of the EP flux divergence was contributed by solar grade) angular momentum of the atmosphere and that of the tides excited in the summer hemisphere, especially around the atmosphere at rest, depends on the eddy forcing. subsolar latitude. The other two thirds of the required forcing were It has been noticed that the atmosphere of Mars can exhibit a created by the stationary planetary wave (SPW) with the zonal local superrotation under dusty conditions (Hamilton, 1982; Zurek, wavenumber s = 1, the transient planetary wave (s = 1, period 1986; Wilson and Hamilton, 1996). Lewis and Read (2003) have 5 sol), and the resolved GWs. found that the superrotation is most pronounced in the lower A similar polar warming but in the lower thermosphere at equatorial atmosphere (one half to one scale height above the sur- 100–130 km have been found during the ODY aerobraking phase face) at equinoxes. They showed that the equatorial westerly jet around Ls = 270° (Keating et al., 2003). GCM simulations of Bou- can extend above 40 km during strong dust storms. In agreement gher et al. (2006) demonstrated that the mechanism of the with the work of Wilson and Hamilton (1996), solar tides have warming formation is analogous to the one in the middle atmo- been identified as the main driving force of the low-level westerly sphere. Using a lower-to-middle atmosphere GCM coupled with jet. When the dust opacity is large and the absorption of solar radi- a thermosphere model, Bell et al. 2007) and McDunn et al. ation increases, temperature swings between day and night be- (2010) have shown that the meridional transport in the thermo- come particularly strong. They amplify the excitation of solar sphere intensifies, and the polar temperature grows when the tides, mainly of sun-synchronous diurnal and semi-diurnal modes. total amount of dust increases and/or it is mixed to higher alti- These tides provide a superrotation torque at places of their gener- tudes. Examples of lofting of lower atmospheric aerosols to alti- ation, in accordance with the mechanism proposed by Fels and tudes as high as 80 km during 2001 strong dust storm event Lindzen (1974). This can be illustrated with the generalized version have been observed with the MGS–TES (Clancy et al., 2010). In of the Eliassen and Palm theorem: their simulations with the first ‘‘ground-to-’’ martian GCM, González-Galindo et al. (2009) emphasized the contribu- d jq ðq u0w0Þ¼ 0 DJ; ð8Þ tion of both migrating and non-migrating tidal components to dz 0 u c A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156 153

a semiannual periodicity in the difference between the day- (local time 2:00 pm) and night-time (local time 2:00 am) temperature retrieved from the MGS–TES limb measurements during MY24 and 25. This difference provides a good measure of twice the amplitude of the diurnal Sun-synchronous tide (Banfield et al., 2003). The latter is very sensitive to variations of the background wind. Further GCM simulations proved that the semiannual modulation of tidal ampli- tudes is indeed caused by the equatorial SAO on Mars. Fig. 7 shows the seasonal variation of the mean zonal wind, u, over the tropics. The equinoctial westerly (prograde) wind is maintained by the tor- que supplied by stationary planetary waves and tides in addition to the weak positive angular momentum transported from mid-lat- itudes by the meridional circulation. The annual cycle simulated with the dust opacity varying seasonally and latitudinally in accor- dance with the MGS–TES retrievals for MY24 (a relatively ‘‘minor’’

dust storm lasting from Ls = 210–270° with the opacity increasing from s = 0.2 to 0.7–1.5 in visible wavelengths) is plotted in Fig. 7a. A strong difference between the simulation for the seasonally and spa- tially uniform dust with s = 0.2 (Fig. 7b) is seen for the northern win- ter: the easterly (retrograde) wind increases from 30 to 120 m s1. It is caused by the combined effect of the enhanced meridional circulation and excitation of tides during dust storms.

6. Influence of dust on non-zonal disturbances

6.1. Thermal tides

A significant fraction of the natural variability in the Mars mid- dle atmosphere is associated with the thermal tide activity. They are composed of in-situ generated tides, and harmonics propagat- ing from below. Diurnal and semidiurnal Sun-synchronous (‘‘migrating’’) tides are the two main modes of the atmospheric re- sponse to the daily cyclic absorption of solar radiation. The semidi- urnal tide has a much longer vertical wavelength than its diurnal counterpart, and therefore, is less subjected to vertical damping by the background zonal winds. If heating becomes distributed over Fig. 7. (a) The composite annual cycle of the simulated mean zonal wind averaged between 10S and 10N from the run with the dust load prescribed according to the greater depths in the atmosphere, the efficiency of exciting semidi- MGS–TES measurements. The contour interval is 10 m s1, westerly (prograde) urnal tides increases. Therefore, during dust storms, the semidiur- wind is shaded with yellow. (b) Same as Fig. 7a, except from the run with the nal tide has noticeably larger amplitudes, extends well above the seasonally uniform dust opacity in visible wavelengths s = 0.2. After Kuroda et al. region of excitation, and dominates the diurnal component almost (2008). everywhere in the middle and upper atmosphere. This behavior has where c is the phase velocity of the disturbance, j =(c 1)/c, c been well reproduced in GCMs (Wilson and Hamilton, 1996; Wilson 0 being the ratio of specific heats cp=cv ; D ¼ðf=HÞþðp =pÞ is the and Richardson, 2000) and quasi-linear tidal models (Forbes and measure of a vertical displacement in the wave, f is the vertical dis- Miyahara, 2006). However, diurnal temperature perturbations can placement of a parcel, and J is the heating rate. The divergence of be quite large too (20–45 K) within the heating region during pla- the eddy momentum flux in the left-hand side of (8) enters the net-encircling dust storms (Smith et al., 2002). These variations are right-hand side of (7), and represents a forcing for the mean zonal composed of vertically trapped diurnal modes with very long verti- wind. It is seen from Eq. (8) that it acts to accelerate the flow in a cal scales. They contribute to the enhancement and poleward direction opposite to c. Thus, westward-propagating migrating tides extension of the meridional circulation, as was discussed above. produce eastward accelerations in regions where heating takes Above 100 km, both molecular viscosity and convective insta- place, extract westward angular moment from the mean flow, bility compete for limiting exponentially growing (due to density transport it upward, and deposit it back upon their dissipation. Both stratification) amplitudes of vertically propagating tides. Forbes the low-level westerly flow and easterlies aloft the equatorial re- and Miyahara (2006) have estimated that, above 150 km, ampli- gion are directly related to the thermal excitation of tides. Lewis tudes of semidiurnal temperature fluctuations, although being and Read (2003) found that the superrotation of the martian atmo- large, do not vary significantly (from 50 to 70 K) when the dust sphere increases slowly with dust loading in integral sense, and opacity rises from low (s = 0.5) to very high (s = 5.0). This occurs strongly depends on the latter locally. due to the dissipation in the thermosphere, and amplitude satura- tion. Between 50 and 100 km, the amplitudes are more sensitive to the amount of airborne aerosol. 5.5. Equatorial semiannual oscillations

Another manifestation of the wave-induced superrotation in the 6.2. Traveling planetary waves martian atmosphere is the semiannual oscillation (SAO) of the mean zonal wind in tropics. The SAO is well documented in the strato- Traveling planetary waves with periods from 2 to 10 sol were sphere of Earth (Garcia et al., 1997), and has been recently found observed in the northern hemisphere of Mars from autumn to on as well (Orton et al., 2008). Kuroda et al. (2008) detected spring. Occurrences of dust storms seriously modify excitation 154 A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156

Fig. 8. Viking Lander 2 observations of the daily-averaged surface pressure variations at 47.97°N, 225.74°W from Ls =90° of MY12 through Ls =0° of MY14. A major dust storm occurred around Ls = 280° in MY12, and lasted for 30–40 sol. Superimposed are the corresponding GCM simulations for the ‘‘dusty’’ MY12 and ‘‘dustless’’ MY13. After Kuroda et al. (2007). and propagation characteristics of traveling PW, and, thus, affect How high the response to dust storms is transmitted in the their periods, zonal wavenumbers, and amplitudes. In particular, atmosphere? There are indications that strong dust events in ejections of large quantities of aerosol to the atmosphere are the lower atmosphere may affect the ionosphere (Haider accompanied by a significant reduction of wave activity. Fig. 8 et al., 2010). What are the coupling mechanisms? Are they pure (red line) shows the time series of pressure variations at the sur- dynamical or include interactions with chemical and electrody- face (47°N) measured by Viking Lander 2. They are the result of namic processes? density oscillations created by traveling PW in the atmosphere Non-migrating tidal signatures are persistent in the upper above. A noticeable decrease of the wave activity (left panel) has atmosphere (Forbes, 2004), and have large amplitudes. The been reported in observations (Barnes, 1980) and GCM simulations associated density fluctuations in the thermosphere are strong (Kuroda et al., 2007) after the onset of a planet-encircling dust enough to affect aerobraking of spacecraft. How does atmo- storm with global mean visible dust opacity up to 4.2 at spheric dust influence non-migrating tides?

Ls 280°, which lasted for 30–40 sol. These changes are caused All martian GCMs consider that dust consists of particles with by the baroclinic stabilization of the flow. Fig. 5d shows that the one or few sizes only. As observations show, this is not true. low-level vertical wind shear in mid-latitudes of the northern What are the dynamical consequences of accounting for the (winter) hemisphere decreases as a result of the meridional tem- variable dust size? perature gradient flattening. The weaker wind shear stabilizes We did not consider in this review the dynamical influence of the flow with respect to baroclinic perturbations, especially for gravity waves, and its relations with a variable dust load. The longer-scale harmonics. This mechanism serves as a negative only reason for this omission is that this topic is still underde- dynamical feedback for dust storm developments. The enhanced veloped. However, there are indications that GWs are extremely mean meridional circulation and reduced baroclinic wave genera- important in the thermosphere (Medvedev et al., 2011). Since tion cause a decline of dust particle lifting from the surface (Basu propagation of GWs is very sensitive to the mean zonal wind, et al., 2006; Kahre et al., 2006). what is their response to the dust-modified background?

7. Summary and conclusions Although this review does not touch upon the role of aeolian processes near the martian surface, the pertaining research is ac- Diabatic heating due to absorption of solar radiation by airborne tively advancing (Basu et al., 2006; Kahre et al., 2006, 2008; dust particles, and the associated cooling due to emissions in infra- Newman et al., 2002a; Wang et al., 2005). The most challenging red are the prominent factors driving the atmosphere of Mars at goal is to understand why some of localized dust lifting events de- various scales. Our paper provides a survey of relatively recent velop into local-area dust storms, and how they strengthen to investigations on the influence of dust on global circulation and large-scale and planet-encircling events. This also includes an large-scale motions. We focus on properties of the dust which explanation for the quasi-regularity of dust storms and their initi- determine the diabatic heating/cooling, and on dynamical feed- ation in certain areas. The mentioned above fundamental science backs to it. question is becoming a practical as well: an accurate weather pre- There is a variety of meteorological phenomena in the lower diction is required for planning numerous martian missions that atmosphere which involve dust. Following the term introduced include landers and rovers. Most likely, a combination of geologi- by Heavens et al. (2011b), vigorous coniometeorological (or ‘‘dust cal, aeolian, meteorological and circulation processes are responsi- meteorological’’) processes take place within the lower scale ble for the dust storm phenomena. This is a fast growing field of height (10 km), which lift the dust from the surface. As a result, martian atmosphere research where the aeolian science commu- suspended particles may remain in the martian atmosphere for nity can greatly contribute. months. Observational data from the spacecraft in the last decade and broad modeling efforts provide an evidence that, although dust Acknowledgments heating/cooling is local, its effect is global. We outlined in this review the main dust-related dynamical ef- This work was partially supported by Research Fellowships of fects including the influence on the zonal-mean circulation, and on the Japan Society for the Promotion of Science (JSPS) for Young Sci- non-zonal disturbances: planetary waves and tides. There are entists, Grant-in-Aid 20-1761, Excellent Young Researchers Over- many open questions in this research area that still remain. Some seas Visit Program of JSPS, and German Science Foundation (DFG) immediate of them are: Grant HA 3261/4,5. A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156 155

References Gierasch, P.J., Goody, R.M., 1972. The effect of dust on the temperature of the Martian atmosphere. J. Atmos. Sci. 29, 400–401. Gierasch, P.J., Goody, R.M., 1973. A model of a Martian great dust storm. J. Atmos. Anderson, E., Leovy, C.B., 1978. Mariner 9 television limb observation of dust and ice Sci. 30, 169–179. hazes on Mars. J. Atmos. Sci. 35, 723–734. González-Galindo, F., Forget, F., López-Valverde, M.A., Angelats i Colli, M., 2009. A Andrews, D.G., Holton, J.R., Leovy, C.B., 1987. Middle Atmosphere Dynamics. ground-to-exosphere Martian general circulation model: 2. Atmosphere during Academic Press, San Diego, CA. solstice conditions – thermospheric polar warming. J. Geophys. Res. 114, Balme, M., Greeley, R., 2006. Dust devils on Earth and Mars. Rev. Geophys. 44, E08004. doi:10.1029/2008JE003277. RG3003. doi:10.1029/2005RG000188. Greeley, R., Iversen, J.D., 1985. Wind as a Geological Process on Earth, Mars, Banfield, D., Conrath, B., Smith, M.D., Christensen, P., Wilson, R.J., 2003. Forced and . Cambridge University Press, 333p. waves in the martian atmosphere from MGS TES nadir data. Icarus 161, 319– Greeley, R., Whelley, P.L., Arvidson, R.E., Cabrol, N.A., Foley, D.J., Franklin, B.J., 345. Geissler, P.G., Golombek, M.P., Kuzmin, R.O., Landis, G.A., Lemmon, M.T., Barnes, J.R., 1980. Time spectral analysis of midlatitude disturbances in the martian Neakrase, L.D., Squyres, S.W., Thompson, S.D., 2006. Active dust devils in atmosphere. J. Atmos. Sci. 37, 2002–2015. Gusev crater, Mars: observations from the Mars Exploration Rover Spirit. J. Barnes, J.R., 1990. Possible effects of breaking gravity waves on the circulation of the Geophys. Res. 111, E12S09. doi:10.1029/2006JE002743. middle atmosphere of Mars. J. Geophys. Res. 95, 1401–1421. Gurwell, M.A., Bergin, E.A., Melnick, G.J., Tolls, V., 2005. Mars surface and Barnes, J.R., Haberle, R.M., 1996. The Martian zonal-mean circulation: angular atmospheric temperature during the 2001 global dust storm. Icarus 175, 23–31. momentum and potential vorticity structure in GCM simulations. J. Atmos. Sci. Haider, S.A., Sheel, V., Smith, M.D., Maguire, W.C., Molina-Caberos, G.J., 2010. The 53, 3143–3156. effects of dust storms on the D region of the Martian ionosphere: atmospheric Barnes, J.R., Hollingsworth, J.L., 1987. Dynamical modeling of a planetary wave electricity. J. Geophys. Res. 115, A12336. doi:10.1029/2010JA016125. mechanism for a Martian polar warming. Icarus 71 (2), 313–334. Hamilton, K., 1982. The effect of solar tides on the general circulation of the Martian Basu, S., Richardson, M.I., Wilson, R.J., 2004. Simulation of the Martian dust cycle atmosphere. J. Atmos. Sci. 39, 481–485. with the GFDL Mars GCM. J. Geophys. Res. 109, E11006. doi:10.1029/ Hansen, J.E., Travis, L.D., 1974. Light scattering in planetary atmospheres. Space Sci. 2004JE002243. Rev. 16, 527–610. Basu, S., Wilson, R.J., Richardson, M.I., Ingersoll, A., 2006. Simulation of spontaneous Hartogh, P., Medvedev, A.S., Kuroda, T., Saito, R., Villanueva, G., Feofilov, A.G., and variable global dust storms with the GFDL Mars GCM. J. Geophys. Res. 111, Kutepov, A.A., Berger, U., 2005. Description and climatology of a new general E09004. doi:10.1029/2005JE002660. circulation model of the Martian atmosphere. J. Geophys. Res. 110, E11008. Bell III, J.F., McCordand, T.B., Owensby, P.D., 1990. Observational evidence of doi:10.1029/2005JE002498. crystalline iron oxides on Mars. J. Geophys. Res. 95, 14447–14461. Heavens, N.G., Richardson, M.I., Kleinböl, A., Kass, D.M., McCleese, D.J., Abdou, W., Bell, J.M., Bougher, S.W., Murphy, J.R., 2007. Vertical dust mixing and the Benson, J.L., Schofield, J.T., Shirley, J.H., Wolkenberg, P.M., 2011a. The vertical interannual variations in the Mars thermosphere. J. Geophys. Res. 112, distribution of dust in the martian atmosphere during northern spring and E12002. doi:10.1029/2006JE002856. summer: 1. Observations by the Mars Climate Sounder and analysis of zonal Bougher, S.W., Bell, J.M., Murphy, J.R., Lopez-Valverde, M.A., Withers, P.G., 2006. average vertical dust profiles. J. Geophys. Res. 116, E04003. doi:10.1029/ Polar warming in the Mars thermosphere: seasonal variations owing to 2010JE003691. changing insolation and dust distributions. Geophys. Res. Lett. 33, L02203. Heavens, N.G., Richardson, M.I., Kleinböl, A., Kass, D.M., McCleese, D.J., Abdou, W., doi:10.1029/2005GL024059. Benson, J.L., Schofield, J.T., Shirley, J.H., Wolkenberg, P.M., 2011b. The vertical Briggs, G.A., Baum, W.A., Barnes, J., 1979. Viking orbiter imaging observations of distribution of dust in the martian atmosphere during northern spring and dust in the Martian atmosphere. J. Geophys. Res. 84, 2795–2820. summer: 2. The high altitude tropical dust maximum. J. Geophys. Res. 116, Cantor, B.A., Malin, M., Edgett, K.S., 2002. Multiyear Mars Orbiter Camera (MOC) E01007. doi:10.1029/2010JE003692. observations of repeated martian weather phenomena during the northern Hide, R., 1969. Dynamics of the atmospheres of the major planets with an appendix summer season. J. Geophys. Res. 107, 5014. doi:10.1029/2001JE001588. on the viscous boundary layer at the rigid bounding surface of an electrically- Clancy, R.T., Lee, S.W., Gladstone, G.R., McMillan, W.W., Rousch, T., 1995. A new conducting rotating fluid in the presence of a magnetic field. J. Atmos. Sci. 26, model for Mars atmospheric dust based upon analysis of ultraviolet through 841–853. infrared observations from Mariner 9, Viking, and Phobos. J. Geophys. Res. 100, Hinson, D.P., 2006. Radio occultation measurements of transient eddies in the 5251–5263. northern hemisphere of Mars. J. Geophys. Res. 111, E05002. doi:10.1029/ Clancy, R.T., Wolff, M.J., Christensen, P.R., 2003. Mars aerosol studies with the MGS 2005JE002612. TES emission phase function observations: optical depths, particle sizes, and ice Hinson, D.P., Wang, H., 2010. Further observations of regional dust storms and cloud types versus latitude and solar longitude. J. Geophys. Res. 108 (E9). baroclinic eddies in the northern hemisphere of Mars. Icarus 206, 290–305. doi:10.1029/2003JE002058. Hinson, D.P., Wilson, R.J., Smith, M.D., Conrath, B.J., 2003. Stationary planetary Clancy, R.T., Wolff, M.J., Whitney, B.A., Cantor, B.A., Smith, M.D., McConnochie, T.H., waves in the atmosphere of Mars during southern winter. J. Geophys. Res. 108, 2010. Extension of atmospheric dust loading to high altitudes during the 2001 5004. doi:10.1029/2002JE001949. Mars dust storm: MGS TES limb observations. Icarus 207, 98–109. Holton, J.R., Haynes, P.H., McIntyre, M.E., Douglass, A.R., Rood, R.E., Pfister, L., 1995. Collins, M., Lewis, S.R., Read, P.L., 1997. Gravity wave drag in a global circulation Stratosphere–troposphere exchange. Rev. Geophys. 33, 403–439. model of the Martian atmosphere: parameterization and validation. Adv. Space Jakosky, B.M., Martin, T.Z., 1987. Mars: north-polar atmospheric temperatures Res. 19 (8), 1245–1254. during dust storms. Icarus 72, 528–534. Conrath, B.J., 1975. Thermal structure of the Martian atmosphere during the James, P.B., Kieffer, H.H., Paige, D.A., 1992. The seasonal cycle of carbon dioxide on dissipation of the dust storm of 1971. Icarus 24, 36–46. Mars. In: Kieffer, H.H. et al. (Eds.), Mars. University of Arizona Press, pp. 934–968. Deming, D., Mumma, M.J., Espenal, F., Kostiuk, T., Zipoy, D., 1986. Polar warming in James, P.B., Hollingsworth, J.L., Wolff, M.J., Lee, S.W., 1999. North polar dust storms the middle atmosphere of Mars. Icarus 66, 366–379. in early spring on Mars. Icarus 138, 64–73. Drossart, P., Rosenqvist, J., Erard, S., Langevin, Y., Bibring, J.P., Combes, M., 1991. Jaquin, F., Gierasch, P., Kahn, R., 1986. The vertical structure of limb hazes in the Martian aerosol properties from the Phobos/ISM experiment. Ann. Geophys. 9, Martian atmosphere. Icarus 68, 442–461. 754–760. Joshi, M.M., Lawrence, B.N., Lewis, S.R., 1995. Gravity wave drag in three- Elteto, A., Toon, O.B., 2010. The effects and characteristics of atmospheric dust dimensional atmospheric models of Mars. J. Geophys. Res. 100, 21235–21245. during martian global dust storm 2001A. Icarus 210, 589–611. Kahre, M.A., Murphy, J.R., Haberle, R.M., 2006. Modeling the martian dust cycle and Fels, S.B., Lindzen, R.S., 1974. The interaction of thermally excited gravity waves surface dust reservoirs with the NASA Ames general circulation model. J. with the mean flows. Geophys. Fluid Dyn. 6, 149–191. Geophys. Res. 111, E06008. doi:10.1029/2005JE002588. Forbes, J.M., 2004. Tides in the middle and upper atmospheres of Mars and Venus. Kahre, M.A., Hollingsworth, J.L., Haberle, R.M., Murphy, J.R., 2008. Investigations of Adv. Space Res. 33, 125–131. the variability of dust particle sizes in the martian atmosphere using the NASA Forbes, J.M., Hagan, M.E., 2000. Diurnal Kelvin wave in the atmosphere of Mars: Ames general circulation model. Icarus 195, 576–597. towards an understanding of ‘stationary’ density structures observed by the Keating, G.M. et al., 2003. Brief review on the results obtained with the MGS and MGS accelerometer. Geophys. Res. Lett. 27, 3563–3566. Mars Odyssey 2001 Accelerometer Experiments. Paper Presented at Forbes, J.M., Miyahara, S., 2006. Solar semidiurnal tide in the dusty atmosphere of International Workshop: Mars Atmosphere Modeling and Observations, Inst. Mars. J. Atmos. Sci. 69, 1798–1817. de Astrofis. de Andalucia, Granada, Spain, January 13–15. Forget, F., 1998. Improved optical properties of the atmospheric dust for radiative Kleinböhl, A., Schofield, J.T., Kass, D.M., Abdou, W.A., Backus, C.R., Sen, B., Shirley, transfer calculations in the infrared. Geophys. Res. Lett. 25, J.H., Lawson, W.G., Richardson, M.I., Taylor, F.W., Teanby, N.A., McCleese, D.J., 1105–1108. 2009. Mars Climate Sounder limb profile retrieval of atmospheric temperature, Forget, F., Hourdin, F., Fournier, R., Hourdin, C., Talagrand, O., Collins, M., Lewis, S.R., pressure, and dust and water ice opacity. J. Geophys. Res. 114, E10006. Read, P.L., Huot, J.-P., 1999. Improved general circulation models of the Martian doi:10.1029/2009JE003358. atmosphere from the surface to above 80 km. J. Geophys. Res. 104, 24155– Korablev, O.I., Kranopolsky, V.A., Rodin, A.V., Chassefiere, E., 1993. Vertical structure 24175. of Martian dust measured by solar infrared occultations from the Phobos Fuerstenau, S.D., 2006. Solar heating of suspended particles and the dynamics of spacecraft. Icarus 102, 76–87. Martian dust devils. Geophys. Res. Lett. 33, L19S03. doi:10.1029/ Kuroda, T., 2006. Study of the Effects of Dust in the Martian Meteorology Using a 2006GL026798. General Circulation Model. University of Tokyo. Garcia, R.R., Dunkerton, T.J., Lieberman, R.S., Vincent, R.A., 1997. Climatology of the Kuroda, T., Hashimoto, N., Sakai, D., Takahashi, M., 2005. Simulation of the Martian semiannual oscillation of the tropical middle atmosphere. J. Geophys. Res. 102, atmosphere using a CCSR/NIES AGCM. J. Meteorol. Soc. Jpn. 83, 1–19. 26019–26032. 156 A.S. Medvedev et al. / Aeolian Research 3 (2011) 145–156

Kuroda, T., Medvedev, A.S., Hartogh, P., Takahashi, M., 2007. Seasonal changes of the Pollack, J.B., Colbun, D.S., Flaser, F.S., Kahn, R., Carlson, C.E., Pidek, D.C., 1979. baroclinic wave activity in the northern hemisphere of Mars simulated with a Properties and effects of dust suspended in the Martian atmosphere. J. Geophys. GCM. Geophys. Res. Lett. 34, L09203. doi:10.1029/2006GL028816. Res. 84, 2929–2945. Kuroda, T., Medvedev, A.S., Hartogh, P., Takahashi, M., 2008. Semiannual oscillations Pollack, J.B., Haberle, R.M., Schaeffer, J., Lee, H., 1990. Simulations of the general in the atmosphere of Mars. Geophys. Res. Lett. 35, L23202. doi:10.1029/ circulation of the Martian atmosphere: 1. Polar processes. J. Geophys. Res. 95, 2008GL036061. 1447–1473. Kuroda, T., Medvedev, A.S., Hartogh, P., Takahashi, M., 2009. On forcing the winter Pollack, J.B., Ockert-Bell, M.E., Shepard, M.K., 1995. Viking Lander image analysis of polar warmings in the martian middle atmosphere during dust storms. J. Martian atmospheric dust. J. Geophys. Res. 100, 5235–5250. Meteorol. Soc. Jpn. 87, 913–921. doi:10.2151/jmsj.87.913. Richardson, M.I., Wilson, R.J., 2002. A topographically forced asymmetry in the Leovy, C.B., Briggs, G.A., Young, A.T., Smith, B.A., Pollack, J.B., Shipley, E.N., Wildey, martian circulation and climate. Nature 416, 98–301. R.L., 1972. The Martian atmosphere: Mariner 9 television experiment progress Rousch, T.L., Rousch, E.A., Singer, R.B., Lucey, P.G., 1992. Estimates of absolute flux report. Icarus 17, 373–393. and radiance factor of localized regions of Mars in 2–4 lm wavelength region. Lewis, S.R., Read, P.L., 2003. Equatorial jets in the dusty martian atmosphere. J. Icarus 99, 42–50. Geophys. Res. 108, 5034. doi:10.1029/2002JE001933. Santee, M., Crisp, D., 1993. Thermal structure and dust loading of the Martian Lewis, S.R., Collins, M., Read, P.L., Forget, F., Hourdin, F., Fournier, R., Hourdin, C., atmosphere during late southern summer – Mariner-9 revisited. J. Geophys. Talagrand, O., Huot, J.P., 1999. A climate data base for Mars. J. Geophys. Res. 104, Res. 98 (E2), 3261–3279. 24177–24194. Schneider, E.K., 1983. Martian great dust storms: interpretive axially symmetric Liu, J., Richardson, M.I., Wilson, R.J., 2003. An assessment of the global, seasonal, and models. Icarus 55, 302–331. interannual spacecraft record of Martian climate in the thermal infrared. J. Schofield, J.T., Barnes, J.R., Crisp, D., Haberle, R.M., Larsen, S., Magalhes, J.A., Murphy, Geophys. Res. 108. doi:10.1029/2002JE001921. J.R., Seiff, A., Wilson, G., 1997. The Mars Pathfinder Atmospheric Structure Markiewics, W.J., Sablotny, R.W., Keller, H.-U., Thomas, N., Titov, D., Smith, P.H., Investigation/Meteorology (ASI/MET) experiment. Science 278, 1752–1758. 1999. Optical properties of the Martian aerosols as derived from Imager for Seth, S.P., Rao, V.B., 2008. Evidence of baroclinic waves in the upper atmosphere of Mars Pathfinder midday sky brightness data. J. Geophys. Res. 104, 9009– Mars using the Mars Global Surveyor accelerometer data. J. Geophys. Res. 113, 9017. A10305. doi:10.1029/2008JA013165. Martin, T.Z., 1986. Thermal infrared opacity of the Mars atmosphere. Icarus 66, 2– Smith, M.D., 2004. Interannual variability in TES atmospheric observations of Mars 21. during 1999–2003. Icarus 167, 148–165. Martin, L.J., Zurek, R.W., 1993. An analysis of the history of dust activity on Mars. J. Smith, M.D., 2006. TES atmospheric temperature, aerosol optical depth, and water Geophys. Res. 98, 3221–3246. vapor observations 1999–2004. In: Second Workshop on Mars Atmos- McCleese, D.J. et al., 2008. Intense polar temperature inversion in the middle phere Modelling and Observations, Granada, Spain. Available from: . Irwin, P.G.J., Kass, D.M., Kleinböhl, A., Lewis, S.R., Paige, D.A., Read, P.L., Smith, M.D., 2009. THEMIS observations of Mars aerosol optical depth from 2002– Richardson, M.I., Shirley, J.H., Taylor, F.W., Teanby, N., Zurek, R.W., 2010. The 2008. Icarus 202, 444–452. structure and dynamics of the martian lower and middle atmosphere as Smith, M.D., Pearl, J.C., Conrath, B.J., Christensen, P.R., 2002. Thermal Emission observed by the mars climate sounder: 1. Seasonal variations in zonal mean Spectrometer observations of a planet-encircling dust storm. Icarus 157, 250– temperature, dust and water ice aerosols. J. Geophys. Res. 115, E12016. 263. doi:10.1029/2010JE003677. Théodore, B., Lellouch, E., Chassefiere, E., Hauchecorne, A., 1993. Solstitial McConnochie, T.H., Smith, M.D., 2008. Vertically resolved aerosol climatology from temperature inversions in the Martian middle atmosphere: observational Mars Global Survayor Thermal Emission Spectrometer (MGS–TES) Limb clues and 2-D modeling. Icarus 105, 512–528. Sounding, Abstract 9114. In: Third International Workshop on The Mars Thorpe, T.E., 1979. A history of Mars atmospheric opacity in the southern Atmosphere: Modeling and Observations Workshop, Williamsburg, VA. hemisphere during the Viking extended mission. J. Geophys. Res. 84, 6663– Available from: . Thorpe, T.E., 1981. Mars atmospheric opacity effects observed in the northern McConnochie, T.H., Wilson, R.J., Smith, M.D., 2009. Dust in the MGS–TES limb hemisphere by Viking Orbiter imaging. J. Geophys. Res. 86, 11419–11429. sounding data set: dust advection by thermal tides, and the dust-free winter Toigo, A.D., Richardson, M.I., 2000. Seasonal variation of aerosols in the Martian pole. In: Final Conference Proceeding of the Mars Dust Cycle Workshop, NASA/ atmosphere. J. Geophys. Res. 105, 4109–4121. Ames Research Center, CA, pp. 152–155. Available from: