Estimation of the escape of photoelectrons from Mars in 2004 liberated by the ionization of carbon dioxide and atomic oxygen

Photoelectron peaks in the atmosphere of Mars caused by the ionization of carbon dioxide and atomic oxygen by solar 30.4 nm photons have been observed by the Electron Spectrometer (ELS), a component of the Mars Express (MEx) Analyzer of Space Plasmas and Energetic Atoms (ASPERA-3) experiment. Ionization mostly occurs at the Mars exobase with the majority of the photoionized electron flux trapped in the remanent and induced magnetic field, with a portion of that flux escaping the planet down its tail. Since Mars is overall charge neutral, the number of electrons must be identical to the number of ion charges which escape the planet. An estimate of the fraction of the total number of escaping electrons is obtained for the year 2004, specifically those produced by the ionization of carbon dioxide and atomic oxygen by solar 30.4 nm photons. In achieving this process, an illustrative example pass is used to show how the electron spectrum is adjusted for the potential on the spacecraft; then the region of the electron spectrum which shows photoelectron peaks is integrated over energy, yielding a flux of 5.74 × 10⁶ electrons/(cm² s sr). This technique is then applied to a subset of 22 sample averaged spectra from the 2004 data (5 January 2004 through 25 January 2005), yielding an average result of 4.15 × 10⁶ electrons/(cm² s sr) for the 22 cases. The observation cone of 33.75° is used to integrate over solid angle (assuming the flux is constant), giving 4.39 × 10⁶ electrons/(cm² s). This average value was taken as representative of the full data interval. Frequency of occurrence statistics showing about a 6.2% occurrence rate for the 2004 data is applied to give an average escape flux from Mars of 2.72 × 10⁵ electrons/(cm² s) during 2004. By estimating the outflow area as 1.16 × 10¹⁸ cm² at X = −1.5 R_Mars the electron escape rate of 3.14 × 10²³ electrons/s is obtained. Thus about 9.92 × 10³⁰ electrons or 16.5 Mmole of electrons escaped Mars during 2004 due to the ionization of carbon dioxide and atomic oxygen by the He 30.4 nm line. Due to the caveats of the analysis, these derived escape rates should be considered lower limits on the total electron escape rate from Mars.     

Monte Carlo computations of the escape of atomic nitrogen from Mars

Monte Carlo simulations were carried out to compute the escape flux of atomic nitrogen for the low and high solar activity martian thermospheres. The total escape of atomic nitrogen at low and high solar activities was found to be 3.03×¹⁰⁵ and , respectively. The escape flux of atomic nitrogen at low and high solar activities from photodissociation of N₂ was found to be 2.75×¹⁰⁵ and , respectively. The remainder of the contribution is from dissociative recombination, which is only important at high solar activity were it comprises about 25% of the total escape. The relative contributions to the total N escape flux from thermal motion of the background atmosphere, winds and co-rotation, and photoionization and subsequent solar wind pickup are also considered here. We find that the total predicted escape fluxes are observed to increase by 20 and 25% at low and high solar activities owing to thermal motion of the background atmosphere. At low and high solar activities, we find that the co-rotation and wind velocities combined translate to a maximum transferable energy of ∼0.0103 and 0.0181 eV, respectively, and that the total escape flux contribution from winds and co-rotation is negligible. Photoionization was found to be a minor process only impacting those source atoms produced with energies close to the escape energy, between 1.5 and 2 eV. The contributions to the total escape fluxes at low and high solar activities from photoionization and subsequent solar wind pickup are found to be about 8 and 13%, respectively.     

Heavy ion escape from Mars, influence from solar wind conditions and crustal magnetic fields

We have used more than 4 years of Mars Express ion data to estimate the escape of heavy ions ( and ) from Mars. To take the limited field of view of the instrument into account, the data has been binned into spatial bins and angular bins to create average distribution functions for different positions in the near Mars space. The net escape flux for the studied low solar activity period, between May 2007 and May 2011, is 2.0 ± 0.2 × 10²⁴ s⁻¹. The escape has been calculated independently for four different quadrants in the Y_MSO − Z_MSO plane, south, dusk, north and dawn. Escape is highest from the northern and dusk quadrants, 0.6 ± 0.1 × 10²⁴ s⁻¹, and smallest from the south and dawn quadrants, 0.4 ± 0.1 × 10²⁴ s⁻¹. The flux ratio of molecular ( and ) to O⁺ ions is 0.9 ± 0.1, averaged over all quadrants. The flux difference between the north and south quadrants is statistically significant, and is presumed to be due to the presence of significant crustal magnetic fields in the southern hemisphere, reducing the outflow. The difference between the dawn and dusk quadrants is likely due to the magnetic tension associated with the nominal Parker angle spiral, which should lead to higher average magnetic tension on the dusk side. The escape increases during periods of high solar wind flux and during times when co-rotating interaction regions (CIR) affect Mars. In the latter case the increase is a factor 2.4-2.9 as compared to average conditions.     

Photochemical escape of oxygen from Mars: A comparison of the exobase approximation to a Monte Carlo method

The non-thermal escape of neutral O atoms from Mars at the current epoch is largely due to dissociative recombination of :

     

Rosetta-Alice observations of exospheric hydrogen and oxygen on Mars

The European Space Agency’s Rosetta spacecraft, en route to a 2014 encounter with comet 67P/Churyumov-Gerasimenko, made a gravity assist swing-by of Mars on 25 February 2007, closest approach being at 01:54 UT. The Alice instrument on board Rosetta, a lightweight far-ultraviolet imaging spectrograph optimized for in situ cometary spectroscopy in the 750-2000 Å spectral band, was used to study the daytime Mars upper atmosphere including emissions from exospheric hydrogen and oxygen. Offset pointing, obtained five hours before closest approach, enabled us to detect and map the H i Lyman-α and Lyman-β emissions from exospheric hydrogen out beyond 30,000 km from the planet’s center. These data are fit with a Chamberlain exospheric model from which we derive the hydrogen density at the 200 km exobase and the H escape flux. The results are comparable to those found from the Ultraviolet Spectrometer experiment on the Mariner 6 and 7 fly-bys of Mars in 1969. Atomic oxygen emission at 1304 Å is detected at altitudes of 400-1000 km above the limb during limb scans shortly after closest approach. However, the derived oxygen scale height is not consistent with recent models of oxygen escape based on the production of suprathermal oxygen atoms by the dissociative recombination of .     

A comparison of global models for the solar wind interaction with Mars

We present initial results from the first community-wide effort to compare global plasma interaction model results for Mars. Seven modeling groups participated in this activity, using MHD, multi-fluid, and hybrid assumptions in their simulations. Moderate solar wind and solar EUV conditions were chosen, and the conditions were implemented in the models and run to steady state. Model output was compared in three ways to determine how pressure was partitioned and conserved in each model, the location and asymmetry of plasma boundaries and pathways for planetary ion escape, and the total escape flux of planetary oxygen ions. The two participating MHD models provided similar results, while the five sets of multi-fluid and hybrid results were different in many ways. All hybrid results, however, showed two main channels for oxygen ion escape (a pickup ion ‘plume’ in the hemisphere toward which the solar wind convection electric field is directed, and a channel in the opposite hemisphere of the central magnetotail), while the MHD models showed one (a roughly symmetric channel in the central magnetotail). Most models showed a transition from an upstream region dominated by plasma dynamic pressure to a magnetosheath region dominated by thermal pressure to a low altitude region dominated by magnetic pressure. However, calculated escape rates for a single ion species varied by roughly an order of magnitude for similar input conditions, suggesting that the uncertainties in both the current and integrated escape over martian history as determined by models are large. These uncertainties are in addition to those associated with the evolution of the Sun, the martian dynamo, and the early atmosphere, highlighting the challenges we face in constructing Mars’ past using models.     

Curvilinear features in the southern hemisphere observed by Mars Global Surveyor Mars Orbiter Camera

We have used the complete set of Mars Global Surveyor (MGS) Mars Daily Global Maps (MDGMs) to study martian weather in the southern hemisphere, focusing on curvilinear features, including frontal events and streaks. “Frontal events” refer to visible events that are morphologically analogous to terrestrial baroclinic storms. MDGMs show that visible frontal events were mainly concentrated in the 210-300°E (60-150°W) sector and the 0-60°E sector around the southern polar cap during L_s = 140-250° and L_s = 340-60°. The non-uniform spatial and temporal distributions of activity were also shown by MGS Thermal Emission Spectrometer transient temperature variations near the surface. “Streaks” refer to long curvilinear features in the polar hood or over the polar cap. They are an indicator of the shape of the polar vortex. Streaks in late winter usually show wavy segments between the 180° meridian and Argyre. Model results suggest that the zonal wave number m = 3 eastward traveling waves are important for their formation.     

Parametric analysis of modeled ion escape from Mars

We develop a parametric fit to the results of a detailed magnetohydrodynamic (MHD) study of the response of ion escape rates (O⁺, and ) to strongly varied solar forcing factors, as a way to efficiently extend the MHD results to different conditions. We then use this to develop a second, evolutionary model of solar forced ion escape. We treat the escape fluxes of ion species at Mars as proportional to the product of power laws of four factors - that of the EUV flux R_euv, the solar wind particle density R_ρ, its velocity (squared) R_v², and the interplanetary magnetic field pressure R_B², where forcing factors are expressed in units of the current epoch-averaged values. Our parametric model is: , where ?(i) is the escape flux of ion i. We base our study on the results of just six provided MHD model runs employing large forcing factor variations, and thus construct a successful, first-order parametric model of the MHD program. We perform a five-dimensional least squares fit of this power law model to the MHD results to derive the flux normalizations and the indices of the solar forcing factors. For O⁺, we obtain the values, 1.73 × 10²⁴ s⁻¹, 0.782, 0.251, 0.382, and 0.214, for ?₀, α, β, γ, and δ, respectively. For , the corresponding values are 1.68 × 10²⁴ s⁻¹, −0.393, 0.798, 0.967, and 0.533. For , they are 8.66 × 10²² s⁻¹, −0.427, 1.083, 1.214, and 0.690. The fit reproduces the MHD results to an average error of about 5%, suggesting that the power laws are broadly representative of the MHD model results. Our analysis of the MHD model shows that by itself an increase in R_EUV enhances O⁺ loss, but suppresses the escape of and , whereas increases in solar wind (i.e., in , and R_B², with R_euv constant) favors the escape of heavier ions more than light ions. The ratios of escaping ions detectable at Mars today can be predicted by this parametric fit as a function of the solar forcing factors. We also use the parametric model to compute escape rates over martian history. This second parametric model expresses ion escape functions of one variable (per ion), ?(i) = ?₀(i)(t/t₀)^−ξ(i). The ξ(i) are linear combinations of the epoch-averaged ion escape sensitivities, which are seen to increase with ion mass. We integrate the and oxygen ion escape rates over time, and find that in the last 3.85 Gyr, Mars would have lost about mbars of , and of water (from O⁺ and ) from ion escape.     

Quasi-periodic dust events in the summertime south polar region of Mars

A Hovmöller diagram analysis of the dust optical depth measured by the Mars Global Surveyor Thermal Emission Spectrometer shows the occurrence of quasi-periodic westwardly-propagating disturbances with timescales of 10-20 sols during summer in the south polar region of Mars. Dust clouds emerge repeatedly around the region with a latitude of around 70-80°S and a longitude of 240-300°E, move westward at speeds of 3-6 m s⁻¹, reach the region with a longitude of 60-120°E, and finally disappear. This longitude range coincides with elevated terrains in the south polar region, and in this region an increase of dust optical depth encircling the south pole is also observed. This implies that the quasi-periodic dust events will contribute to the enhancement of the atmospheric dust loading in this region. These dust events might be related to baroclinic instability caused by the thermal contrast across the CO₂ cap edge, or the horizontal advection or vertical convection with radiative-dynamical feedback. The westward movement of the dust clouds suggests steady westward winds blowing in the near-surface layer, where the quasi-periodic dust lifting is expected to occur. Such a westward cap-edge flow will be created by the Coriolis force acting on the flow from the ice side to the regolith side.     

First ENA observations at Mars: ENA emissions from the martian upper atmosphere

The neutral particle detector (NPD) on board Mars Express has observed energetic neutral atoms (ENAs) from a broad region on the dayside of the martian upper atmosphere. We show one such example for which the observation was conducted at an altitude of 570 km, just above the induced magnetosphere boundary (IMB). The time of flight spectra of these ENAs show that they had energies of 0.2-2 keV/amu, with an average energy of ∼1.1 keV/amu. Both the spatial distribution and the energy of these ENAs are consistent with the backscattered ENAs, produced by an ENA albedo process. This is the first observation of backscattered ENAs from the martian upper atmosphere. The origin of these ENAs is considered to be the solar wind ENAs that are scattered back by collision processes in the martian upper atmosphere. The particle flux and energy flux of the backscattered ENAs are and , respectively.     

Ion escape at Mars: Comparison of a 3-D hybrid simulation with Mars Express IMA/ASPERA-3 measurements

We have analysed ion escape at Mars by comparing ASPERA-3/Mars Express ion measurements and a 3-D quasi-neutral hybrid model. As Mars Express does not have a magnetometer onboard, the analysed IMA data are from an orbit when the IMF clock angle was possible to determine from the magnetic field measurements of Mars Global Surveyor. We found that fast escaping planetary ions were observed at the place which, according to the 3-D model, is anticipated to contain accelerated heavy ions originating from the martian ionosphere. The direction of the interplanetary magnetic field was found to affect noticeably which regions can be magnetically connected to Mars Express and to the overall 3-D Mars-solar wind interaction.     

Further observations of regional dust storms and baroclinic eddies in the northern hemisphere of Mars

We have investigated the near-surface meteorology in the northern hemisphere of Mars through detailed analysis of data obtained with Mars Global Surveyor in January-August 2005. The season in the northern hemisphere ranged from midsummer through winter solstice of Mars Year (MY) 27. We examined composite, wide-angle images from the Mars Orbiter Camera and compiled a catalog of the dust storms that occurred in this interval. As in previous martian years, activity in the northern hemisphere was dominated by regional “flushing” dust storms that sweep southward through the major topographic basins, most frequently in Acidalia Planitia. We also used atmospheric profiles retrieved from radio occultation experiments to characterize eddy activity near the surface at high northern latitudes. There are strong correlations between the two sets of observations, which allowed us to identify three factors that influence the timing and location of the regional dust storms: (1) transitions among baroclinic wave modes, which strongly modulate the intensity of meridional winds near the surface, (2) storms zones, which impose strong zonal variations on the amplitude of some baroclinic eddies, and (3) stationary waves, which further modulate the wind field near the surface. The flushing dust storms ceased abruptly in midautumn, possibly in response to source depletion, CO₂ condensation, a shift in the period of the baroclinic eddies, and changes in the tidal wind field near the surface. Our results extend the meteorological record of the northern hemisphere, substantiate the findings of previous investigations, and further illuminate the climatic impact of baroclinic eddies.     

Solar activity variations of thermospheric temperatures on Mars and a problem of CO in the lower atmosphere

Long-term MGS drag density observations at 390 km reveal variations of the density with season L_S (by a factor of 2) and solar activity index F_10.7 (by a factor of 3 for F_10.7 = 40-100). According to Forbes et al. (Forbes, J.M., Lemoine, F.G., Bruinsma, S.L., Smith, M.D., Zhang, X. [2008]. Geophys. Res. Lett. 35, L01201, doi:10.1029/2007GL031904), the variation with F_10.7 reflects variations of the exospheric temperature from 192 to 284 K. However, the derived temperature range corresponds to variation of the density at 390 km by a factor of 8, far above the observed factor of 3. The recent thermospheric GCMs agree with the derived temperatures but do not prove their adequacy to the MGS densities at 390 km. A model used by Forbes et al. neglects effects of eddy diffusion, chemistry and escape on species densities above 138 km. We have made a 1D-model of neutral and ion composition at 80-400 km that treats selfconsistently chemistry and transport of species with F_10.7, T_∞, and [CO₂]_80 km as input parameters. Applying this model to the MGS densities at 390 km, we find variation of T_∞ from 240 to 280 K for F_10.7 = 40 and 100, respectively. The results are compared with other observations and models. Temperatures from some observations and the latest models disagree with the MGS densities at low and mean solar activity. Linear fits to the exospheric temperatures are T_∞ = 122 + 2.17F_10.7 for the observations, T_∞ = 131 + 1.46F_10.7 for the latest models, and T_∞ = 233 + 0.54F_10.7 for the MGS densities at 390 km. Maybe the observed MGS densities are overestimated near solar minimum when they are low and difficult to measure. Seasonal variations of Mars’ thermosphere corrected for the varying heliocentric distance are mostly due to the density variations in the lower and middle atmosphere and weakly affect thermospheric temperature. Nonthermal escape processes for H, D, H₂, HD, and He are calculated for the solar minimum and maximum conditions.Another problem considered here refers to Mars global photochemistry in the lower and middle atmosphere. The models gave too low abundances of CO, smaller by an order of magnitude than those observed. Our current work shows that modifications in the boundary conditions proposed by Zahnle et al. (Zahnle, K., Haberle, R.M., Catling, D.C., Kasting, J.F. [2008]. J. Geophys. Res. 113, E11004, doi:10.1029/2008JE003160) are reasonable but do not help to solve the problem.     

Annual survey of water vapor behavior from the OMEGA mapping spectrometer onboard Mars Express

We present here the annual behavior of atmospheric water vapor on Mars, as observed by the OMEGA spectrometer on board Mars Express during its first martian year. We consider all the different features of the cycle of water vapor: temporal evolution, both at a seasonal and at a diurnal scale; longitudinal distribution; and the vertical profile, through the variations in the saturation height. We put our results into the context of the current knowledge on the water cycle through a systematic comparison with the already published datasets. The seasonal behavior is in very good agreement with past and simultaneous retrievals both qualitatively and quantitatively, within the uncertainties. The average water vapor abundance during the year is ∼10 pr. μm, with an imbalance between northern and southern hemisphere, in favor of the first. The maximum of activity, up to 60 pr. μm, occurs at high northern latitudes during local summer and shows the dominance of the northern polar cap within the driving processes of the water cycle. A corresponding maximum at southern polar latitudes during the local summer is present, but less structured and intense. It reaches ∼25 pr. μm at its peak. Global circulation has some influence in shaping the water cycle, but it is less prominent than the results from previous instruments suggest. No significant correlation between water vapor column density and local hour is detected. We can constrain the amount of water vapor exchanged between the surface and the atmosphere to few pr. μm. This is consistent with recent results by OMEGA and PFS-LW. The action of the regolith layer on the global water cycle seems to be minor, but it cannot be precisely constrained. The distribution of water vapor on the planet, after removing the topography, shows the already known two-maxima system, over Tharsis and Arabia Terra. However, the Arabia Terra increase is quite fragmented compared with previous observations. A deep zone of minimum separates the two regions. The saturation height of water vapor is mainly governed by the variations of insolation during the year. It is confined within 5-15 km from the surface at aphelion, while in the perihelion season it stretches up to 55 km of altitude.     

Comment on the papers “Production of hot nitrogen atoms in the martian thermosphere” by F. Bakalian and “Monte Carlo computations of the escape of atomic nitrogen from Mars” by F. Bakalian and R.E. Hartle

In recent articles published in Icarus, Bakalian [2006. Icarus 183, 69-78] discusses and computes the production rates of hot nitrogen atoms in the martian thermosphere due to N₂ photodissociation and N⁺₂ dissociative recombination, and Bakalian and Hartle [2006. Icarus 183, 55-68] use a Monte Carlo code to compute the escape rates of nitrogen atoms from Mars due to photodissociation of N₂, dissociative recombination of N⁺₂, and pickup ion escape due to photoionization of N atoms above the ionopause. Bakalian concludes that “photodissociation of N₂ is the dominant escape mechanism in the martian atmosphere.” We will show that this conclusion is not supportable. In addition, both papers contain scientific errors, misrepresentations, inaccurate referencing, lack of proper attribution, and they fail to place these investigations into the existing extensive body of work on this subject.     

Carbon dioxide photoelectron energy peaks at Mars

The ELectron Spectrometer (ELS) from the Analyzer of Space Plasmas and Energetic Atoms (ASPERA-3) flown on the Mars Express spacecraft has an 8% energy resolution, combined with the capability to oversample the martian electron distribution. This makes possible the resolution and identification of electrons generated as a result of the He 304 Å ionization of CO₂ at the martian exobase on the dayside of the planet. Ionospheric photoelectrons were observed during almost every pass into the ionosphere and CO₂ photoelectron peaks were identified near the terminator. Atmospherically generated CO₂ photoelectrons are also observed at 10,000 km altitude in the martian tail near the inner magnetospheric boundary. Observations over a wide range of spacecraft orbits showed a consistent presence of photoelectrons at locations along the inner magnetospheric boundary and in the ionosphere, from an altitude of 250 to 10,000 km.     

Extension of atmospheric dust loading to high altitudes during the 2001 Mars dust storm: MGS TES limb observations

Mars Global Surveyor (MGS) visible (solarband bolometer) and thermal infrared (IR) spectral limb observations from the Thermal Emission Spectrometer (TES) support quantitative profile retrievals for dust opacity and particle sizes during the 2001 global dust event on Mars. The current analysis considers the behavior of dust lifted to altitudes above 30 km during the course of this storm; in terms of dust vertical mixing, particle sizes, and global distribution. TES global maps of visible (solarband) limb brightness at 60 km altitude indicate a global-scale, seasonally evolving (over 190-240° solar longitudes, L_S) longitudinal corridor of vertically extended dust loading (which may be associated with a retrograde propagating, wavenumber 1 Rossby wave). Spherical radiative transfer analysis of selected limb profiles for TES visible and thermal IR radiances provide quantitative vertical profiles of dust opacity, indicating regional conditions of altitude-increasing dust mixing ratios. Observed infrared spectral dependences and visible-to-infrared opacity ratios of dust scattering over 30-60 km altitudes indicate particle sizes characteristic of lower altitudes (cross-section weighted effective radius, ), during conditions of significant dust transport to these altitudes. Conditions of reduced dust loading at 30-60 km altitudes present smaller dust particle sizes . These observations suggest rapid meridional transport at 30-80 km altitudes, with substantial longitudinal variation, of dust lifted to these altitudes over southern hemisphere atmospheric regions characterized by extraordinary (m/s) vertical advection velocities. By L_S=230° dust loading above 50 km altitudes decreased markedly at southern latitudes, with a high altitude (60-80 km) haze of fine (likely) water ice particles appearing over 10°S-40°N latitudes.     

Dust devil sediment flux on Earth and Mars: Laboratory simulations

Laboratory simulations using the Arizona State University Vortex Generator (ASUVG) were run to simulate sediment flux in dust devils in terrestrial ambient and Mars-analog conditions. The objective of this study was to measure vortex sediment flux in the laboratory to yield estimations of natural dust devils on Earth and Mars, where all parameters may not be measured. These tests used particles ranging from 2 to 2000 μm in diameter and 1300 to 4800 kg m⁻³ in density, and the results were compared with data from natural dust devils on Earth and Mars. Typically, the cores of dust devils (regardless of planetary environment) have a pressure decrease of ∼0.1-1.5% of ambient atmospheric pressure, which enhances the lifting of particles from the surface. Core pressure decreases in our experiments ranged from ∼0.01% to 5.00% of ambient pressure (10 mbar Mars cases and 1000 mbar for Earth cases) corresponding to a few tenths of a millibar for Mars cases and a few millibars for Earth cases. Sediment flux experiments were run at vortex tangential wind velocities of 1-45 m s⁻¹, which typically correspond to ∼30-70% above vortex threshold values for the test particle sizes and densities. Sediment flux was determined by time-averaged measurements of mass loss for a given vortex size. Sediment fluxes of ∼10⁻⁶-10⁰ kg m⁻² s⁻¹ were obtained, similar to estimates and measurements for fluxes in dust devils on Earth and Mars. Sediment flux is closely related to the vortex intensity, which depends on the strength of the pressure decrease in the core (ΔP). This study found vortex size is less important for lifting materials because many different diameters can have the same ΔP. This finding is critical in scaling the laboratory results to natural dust devils that can be several orders of magnitude larger than the laboratory counterparts.     

Estimates of gravity wave drag on Mars: Indication of a possible lower thermospheric wind reversal

Spectral gravity wave parameterization suitable for planetary thermospheres applied to wind and temperature from Mars Climate Database reveals enormously strong drag incompatible with the wind distribution. It points out to a possible wind reversal in the 110-140 km layer similar to the one in the Earth’s lower thermosphere.     

The atmospheric temperatures over Olympus Mons on Mars: An atmospheric hot ring

We study the thermal fields over Olympus Mons separating seasons (northern spring and summer against southern spring and summer) and local time observations (day side versus night side). Temperature vertical profiles retrieved from Planetary Fourier Spectrometer on board Mars Express (PFS-MEX) data have been used. In many orbits (running north to south along a meridian) passing over the top of the volcano there is evidence of a hot air on top of the volcano, of two enhancement of the air temperature both north and south of it and in between a collar of air that is colder than nearby at low altitudes, and warmer than nearby at high altitudes. Mapping together many orbits passing over or near the volcano we find that the hot air has the tendency to form an hot ring around it. This hot structure occurs mostly between LT = 10.00 and 15.00 and during the northern summer. Distance of the hot structure from the top of the volcano is about 600 km (10° of latitude). The hot atmospheric region is 300-420 km (5-7°) wide. Hot ring temperature contrasts of about 40 K occur at 2 km above the surface and decrease to 20 K at 5 km and to 10 K at 10 km. The atmospheric circulation over an area of 40° × 40° (latitudes and longitudes) is affected by the topography and the presence of Olympus Mons (−133°W, 18°N). We discuss also the thermal stability of the atmosphere over the selected area using the potential temperatures. The temperature field over the top of the volcano shows unstable atmosphere within 10 km from the surface. Finally, we interpret the hot temperatures around volcano as an adiabatic compression of down-welling branch coming from over the top of volcano.Different air temperature profiles are observed in the same seasons during the night, or in different seasons. In northern spring-summer during the night the isothermal contours do not show the presence of the volcano until we reach close to the surface very much, where a thermal inversion is observed. The surface temperature shows higher values (by 10 K) in correspondence of the scarp (an abrupt altimetry variation of roughly 5 km) on south (6°N) and north (30°N) sides of volcano. During the southern spring-summer, on the contrary the isothermal curves run parallel to the surface even on top the volcano, just like the GCM have predicted.