Assimilation of thermal emission spectrometer atmospheric data during the Mars Global Surveyor aerobraking period

The Thermal Emission Spectrometer aboard the Mars Global Surveyor spacecraft has produced an extensive atmospheric data set, beginning during aerobraking and continuing throughout the extended scientific mapping phase. Temperature profiles for the atmosphere below about 40 km, surface temperatures and total dust and water ice opacities, can be retrieved from infrared spectra in nadir viewing mode. This paper describes assimilation of nadir retrievals from the spacecraft aerobraking period, L_S=190°–260°, northern hemisphere autumn to winter, into a Mars general circulation model. The assimilation scheme is able to combine information from temperature and dust optical depth retrievals, making use of a model forecast containing information from the assimilation of earlier observations, to obtain a global, time-dependent analysis. Given sufficient temperature retrievals, the assimilation procedure indicates errors in the a priori dust distribution assumptions even when lacking dust observations; in this case there are relatively cold regions above the poles compared to a model which assumes a horizontally-uniform dust distribution. One major reason for using assimilation techniques is in order to investigate the transient wave behavior on Mars. Whilst the data from the 2-h spacecraft mapping orbit phase is much more suitable for assimilation, even the longer (45–24 h) period aerobraking orbit data contain useful information about the three-dimensional synoptic-scale martian circulation which the assimilation procedure can reconstruct in a consistent way. Assimilations from the period of the Noachis regional dust storm demonstrate that the combined assimilation of temperature and dust retrievals has a beneficial impact on the atmospheric analysis.     

THEMIS observations of Mars aerosol optical depth from 2002-2008

We use infrared images obtained by the Thermal Emission Imaging System (THEMIS) instrument on-board Mars Odyssey to retrieve the optical depth of dust and water ice aerosols over more than 3.5 martian years between February 2002 (MY 25, Ls=330°) and December 2008 (MY 29, Ls=183°). These data provide an important bridge between earlier TES observations and recent observations from Mars Express and Mars Reconnaissance Orbiter. An improvement to our earlier retrieval [Smith, M.D., Bandfield, J.L., Christensen, P.R., Richardson, M.I., 2003. J. Geophys. Res. 108, doi:10.1029/2003JE002114] to include atmospheric temperature information from THEMIS Band 10 observations leads to much improved retrievals during the largest dust storms. The new retrievals show moderate dust storm activity during Mars Years 26 and 27, although details of the strength and timing of dust storms is different from year to year. A planet-encircling dust storm event was observed during Mars Year 28 near Southern Hemisphere Summer solstice. A belt of low-latitude water ice clouds was observed during the aphelion season during each year, Mars Years 26 through 29. The optical depth of water ice clouds is somewhat higher in the THEMIS retrievals at ∼5:00 PM local time than in the TES retrievals at ∼2:00 PM, suggestive of possible local time variation of clouds.     

Validation of martian meteorological data assimilation for MGS/TES using radio occultation measurements

We describe an assimilation of thermal profiles below about 40 km altitude and total dust opacities into a general circulation model (GCM) of the martian atmosphere. The data were provided by the Thermal Emission Spectrometer (TES) on board the Mars Global Surveyor (MGS) spacecraft. The results of the assimilation are verified against an independent source of contemporaneous data represented by radio occultation measurements with an ultra-stable radio oscillator, also aboard MGS. This paper describes a comparison between temperature profiles retrieved by the radio occultation experiments and the corresponding profiles given by both an independent, carefully tuned GCM simulation and by an assimilation of TES observations performed over the period of time from middle, northern summer in martian year 24, corresponding to May 1999, until late, northern spring in martian year 27, corresponding to August 2004. This study shows that the assimilation of TES measurements improves the overall agreement between radio occultation observations and the GCM analysis, in particular below 20 km altitude, where the radio occultation measurements are known to be most accurate. Discrepancies still remain, mostly during the global dust storm of year 2001 and at latitudes around 60° N in northern winter-early spring. These are the periods of time and locations, however, for which discrepancies between TES and radio occultation profiles are also shown to be the largest. Finally, a further direct validation is performed, comparing stationary waves at selected latitudes and time of year. Apart from biases at high latitudes in winter time, data assimilation is able to represent the correct wave behaviour, which is one major objective for martian assimilation.     

MGS TES observations of the water vapor above the seasonal and perennial ice caps during northern spring and summer

We report on new retrievals of water vapor column abundances from the Mars Global Surveyor (MGS) Thermal Emission Spectrometer (TES) data. The new retrievals are from the TES nadir data taken above the ‘cold’ surface areas in the North polar region (T_surf < 220 K, including seasonal frost and permanent ice cap) during spring and summer seasons, where retrievals were not performed initially. Retrievals are possible (with some modifications to the original algorithm) over cold surfaces overlaid by sufficiently warm atmosphere. The retrieved water vapor column abundances are compared to the column abundances observed by other spacecrafts in the Northern polar region during spring and summer and good agreement is found. We detect an annulus of water vapor growing above the edge of the retreating seasonal cap during spring. The formation of the vapor annulus is consistent with the previously proposed mechanism for water cycling in the polar region, according to which vapor released by frost sublimation during spring re-condenses on the retreating seasonal CO₂ cap. The source of the vapor in the vapor annulus, according to this model, is the water frost on the surface of the CO₂ at the retreating edge of the cap and the frost on the ground that is exposed by the retreating cap. Small contribution from regolith sources is possible too, but cannot be quantified based on the TES vapor data alone. Water vapor annulus exhibits interannual variability, which we attribute to variations in the atmospheric temperature. We propose that during spring and summer the water ice sublimation is retarded by high relative humidity of the local atmosphere, and that higher atmospheric temperatures lead to higher vapor column abundances by increasing the water holding capacity of the atmosphere. Since the atmospheric temperatures are strongly influenced by the atmospheric dust content, local dust storms may be controlling the release of vapor into the polar atmosphere. Water vapor abundances above the residual polar cap also exhibit noticeable interannual variability. In some years abundances above the cap are lower than the abundances outside of the cap, consistent with previous observations, while in the other years the abundances above the cap are higher or similar to abundances outside of the cap. We speculate that the differences may be due to weaker off-cap transport in the latter case, keeping more vapor closer to the source at the surface of the residual cap. Despite the large observed variability in water vapor column abundances in the Northern polar region during spring and summer, the latitudinal distribution of the vapor mass in the atmosphere is very similar during the summer season. If the variability in vapor abundances is caused by the variability of vapor sources across the residual cap then this would mean that they annually contribute relatively little vapor mass to significantly affect the vapor mass budget. Alternatively this may suggest that the vapor variability is caused by the variability of the polar atmospheric circulation. The new water vapor retrievals should be useful in tuning the Global Circulation Models of the martian water cycle.     

Convective instability in the martian middle atmosphere

Dry convective instabilities in Mars’s middle atmosphere are detected and mapped using temperature retrievals from Mars Climate Sounder observations spanning 1.5 martian years. The instabilities are moderately frequent in the winter extratropics. The frequency and strength of middle atmospheric convective instability in the northern extratropics is significantly higher in MY 28 than in MY 29. This may have coupled with changes to the northern hemisphere mid-latitude and tropical middle atmospheric temperatures and contributed to the development of the 2007 global dust storm. We interpret these instabilities to be the result of gravity waves saturating within regions of low stability created by the thermal tides. Gravity wave saturation in the winter extratropics has been proposed to provide the momentum lacking in general circulation models to produce the strong dynamically-maintained temperature maximum at 1-2 Pa over the winter pole, so these observations could be a partial control on modeling experiments.     

Propagation of tropospheric gravity waves into the upper atmosphere of Mars

We study the propagation of gravity waves in the martian atmosphere using a linearized one-dimensional full-wave model. Calculations are carried out for atmospheric parameters characteristic of Mars Orbiter Laser Altimeter (on Mars Global Surveyor MGS) observations of apparent gravity waves in high latitude clouds and MGS radio occultation measurements of temperature variations with height suggestive of gravity wave activity. Waves that reach the thermosphere produce fluctuations in density comparable in amplitude with the density variations detected in Mars Odyssey aerobraking data. Gravity waves of modest amplitude are found to deposit momentum and generate significant heating and cooling in the martian atmosphere. The largest heating and cooling effects occur in the thermosphere, at altitudes between about 130 and 150 km, with heating occurring at the lower altitudes and cooling taking place above.     

The effects of the martian regolith on GCM water cycle simulations

This paper describes General Circulation Model (GCM) simulations of the martian water cycle focusing on the effects of an adsorbing regolith. We describe the 10-layer regolith model used in this study which has been adapted from the 1-D model developed by Zent, A.P., Haberle, R.M., Houben, H.C., Jakosky, B.M. [1993. A coupled subsurface-boundary layer model of water on Mars. J. Geophys. Res. 98 (E2), 3319-3337, February]. Even with a 30-min timestep and taking into account the effect of surface water ice, our fully implicit scheme compares well with the results obtained by Zent, A.P., Haberle, R.M., Houben, H.C., Jakosky, B.M. [1993. A coupled subsurface-boundary layer model of water on Mars. J. Geophys. Res. 98 (E2), 3319-3337, February]. This means, however, that the regolith is not able to reproduce the diurnal variations in column water vapour abundance of up to a factor of 2-3 as seen in some observations, with only about 10% of the atmospheric water vapour column exchanging with the subsurface on a daily basis. In 3-D simulations we find that the regolith adsorbs water preferentially in high latitudes. This is especially true in the northern hemisphere, where perennial subsurface water ice builds up poleward of 60° N at depths which are comparable to the Odyssey observations. Much less ice forms in the southern high latitudes, which suggests that the water ice currently present in the martian subsurface is not stable under present conditions and is slowly subliming and being deposited in the northern hemisphere. When initialising the model with an Odyssey-like subsurface water ice distribution the model is capable of forcing the simulated water cycle from an arbitrary state close to the Mars Global Surveyor Thermal Emission Spectrometer observations. Without the actions of the adsorbing regolith the equilibrated water cycle is found to be a factor of 2-4 too wet. The process by which this occurs is by adsorption of water during northern hemisphere summer in northern mid and high latitudes where it remains locked in until northern spring when the seasonal CO₂ ice cap retreats. At this time the water diffuses out of the regolith in response to increased temperature and is returned to the residual water ice cap by eddie transport.     

MAWD observations revisited: seasonal behavior of water vapor in the martian atmosphere

Viking/MAWD experiment and more recent MGC/TES observations have provided to date the most detailed information about the annual atmospheric water cycle on Mars. Their data agree in major details but still reveal some disagreements. These disagreements turn out to be most significant in the perihelion season and especially during the major dust storms. We consider the potential influence of aerosol scattering on 1.38 μm water retrieval under various types of observation geometry. In order to obtain new retrievals of water vapor abundance from MAWD data, we apply radiative transfer calculations. The resulting seasonal and spatial distribution of water turns out to be more consistent with TES results, implying a remarkable stability of the martian seasonal water cycle. Mapping data corresponding to particular seasons reveals a distinct wave structure in the global distribution of the water column. We interpret it as a manifestation of a strong control over the water cycle on Mars from the atmospheric circulation.     

Evidence for subsurface water ice in Korolev crater, Mars

Following the work of Kieffer and Titus (2001, Icarus 154, 162-180), we present results of thermal IR observations of Korolev crater, located at ∼73° latitude in the martian northern polar region. Similar to techniques employed by Titus et al. (2003, Science 299, 1048-1050), we use infrared images from the Thermal Emission Imaging System (THEMIS) aboard Mars Odyssey to identify several regions within the crater basin with distinct thermal properties that correlate with topography. The THEMIS results show these regions exhibit temperature variations, spatially within the crater and throughout the martian year. In addition to the variations identified in the THEMIS observations, Mars Global Surveyor Thermal Emission Spectrometer (TES) observations show differences in albedo and temperature of these regions on both daily and seasonal cycles. Modeling annual temperature variations of the surface, we use TES observations to examine the thermal properties of these regions. This analysis reveals the crater interior deposits are likely thick layers (several meters) of high thermal inertia material (water ice, or extremely ice-rich regolith). Spatial variations of the physical properties of these regions are likely due to topography and possibly variations in the subsurface material itself. The nature of these deposits may help constrain polar processes, as well as provide context for the polar lander mission, Phoenix.     

Three-dimensional on-line chemical modeling in a Mars general circulation model

In this paper we present the design and some results of an on-line chemical model of the atmosphere of Mars using a three-dimensional dynamical-chemical model. A chemistry module has been added to the 3D Mars Global Multiscale Model (GM3), a general circulation model based on the Canadian weather forecast model. As water is a major fuel for chemistry on Mars, we have taken care to simulate the atmospheric water vapor content, taking into account evaporation from and condensation to the polar caps and in the atmosphere we allow for freezing and formation of a bulk cloud and sedimentation. The chemical module includes the main neutral carbon-oxygen-hydrogen-nitrogen species and associated relevant reactions. The outputs of the model are analyzed in the light of our current understanding of the chemistry on Mars. The results are in general accordance with 1D models and with two other 3D models with more limited chemistry. The model does not require elevated diffusion coefficients to fit observational data in the thermosphere as was suggested in some 1D studies. The large scale transport and the downward thermospheric transport in the winter pole appear to have a significant role in the concentrations of some trace species.     

Some problems related to the origin of methane on Mars

The following problems related to the origin of methane on Mars have been considered. (1) Laboratory simulations of the impact phenomena confirm effective heterogeneous chemistry between the products of the fireball. This chemistry lowers the fireball freezing temperature from 2000 to 750 K for methane and to 1100 K for CO/CO₂. Production of methane on Mars by cometary impacts is 0.8% of the total production. A probability that the observed methane on Mars came from impact of a single comet is 0.0011. (2) The PFS observations of variations of methane on Mars require a very effective heterogeneous loss of methane. Heterogeneous effect of dust is half that of the surface rocks. Thermochemical equilibrium requires production, not loss, of methane. Existing kinetic data show a very low efficiency of heterogeneous reactions of methane. Highly reactive superoxide ions generated by the solar UV photons on the martian rocks cannot remove methane. The required efficiency of heterogeneous loss of methane on Mars is higher than that on Earth by a factor of ?1000, although the expected efficiency on Earth is stronger than that on Mars because of the liquid ocean and the abundant oxygen. All these inconsistencies may be removed if variations of the rock reflectivity contribute to the PFS observations of methane on Mars. The PFS data on H₂CO, HCl, HF, and HBr also raise doubts. (3) Although geologic sources of methane are possible, the lack of current volcanism, hydrothermal activity, hot spots, and very low seepage of gases from the interior are not favorable for geologic methane. Any proposed geological source of methane on Mars should address these problems. Some weak points in the suggested geologic sources are discussed. (4) Measurements of ¹³C/¹²C and D/H in methane would be difficult because of the low methane abundance. These ratios are mostly sensitive to a temperature of methane formation and cannot distinguish between biogenic and low-temperature geologic sources. Their analysis requires the carbon isotope ratio in CO₂ on Mars, which is known with the insufficient accuracy, and D/H in water, which is different in the atmosphere, polar caps, regolith and interior. Therefore, the stable isotope ratios may not give a unique answer on the origin of methane. (5) Ethane and propane react with OH much faster than methane. If their production relative to methane is similar to that on Earth, then their expected abundances on Mars are of a few parts per trillion. (6) Loss of SO₂ in the reaction with peroxide on ice is smaller than its gas-phase loss by an order of magnitude. The overall results strengthen the biogenic origin of martian methane and its low variability.     

Investigations of the variability of dust particle sizes in the martian atmosphere using the NASA Ames General Circulation Model

We present a Mars General Circulation Model (GCM) numerical investigation of the physical processes (i.e., wind stress and dust devil dust lifting and atmospheric transport) responsible for temporal and spatial variability of suspended dust particle sizes. Measurements of spatial and temporal variations in airborne dust particles sizes in the martian atmosphere have been derived from Mars Global Surveyor (MGS) Thermal Emission Spectrometer (TES) spectral and emission phase function data [Wolff, M.J., Clancy, R.T., 2003. J. Geophys. Res. (Planets) 108 (E9), doi:10.1029/2003JE002057. 1-1; Clancy, R.T., Wolff, M.J., Christensen, P.R., 2003. J. Geophys. Res. (Planets) 108 (E9), doi:10.1029/2003JE002058. 2-1]. The range of dust particle sizes simulated by the NASA Ames GCM is qualitatively consistent with TES-derived observations of effective dust particle size variability. Model results suggest that the wind stress dust lifting scheme (which produces regionally confined dust lifting) is the process responsible for the majority of the dust particle size variability in the martian atmosphere. Additionally, model results suggest that atmospheric transport processes play an important role in the evolution of atmospheric dust particles sizes during substantial dust storms on Mars. Finally, we show that including the radiative effects of a spatially variable particle size distribution significantly influences thermal and dynamical fields during the dissipation phase of the simulated global dust storm.     

Climate, weather, and north polar observations from the Mars Reconnaissance Orbiter Mars Color Imager

The Mars Reconnaissance Orbiter observes Mars from a nearly circular, polar orbit. From this vantage point, the Mars Color Imager extends the ∼5 Mars years record of Mars Global Surveyor global, visible-wavelength multi-color observations of meteorological events and adds measurements at three additional visible and two ultraviolet wavelengths. Observations of the global distribution of ozone (which anti-correlates with water vapor) and water ice and dust clouds allow tracking of atmospheric circulation. Regional and local observations emphasize smaller scale atmospheric dynamics, especially those related to dust lifting and subsequent motion. Polar observations detail variations related to the polar heat budget, including changes in polar frosts and ices, and storms generated at high thermal contrast boundaries.     

Effect of atmospheric dust loading on martian albedo measurements

This work is devoted to the analysis of the variation of albedo measured by orbiting instruments with atmospheric opacity on Mars. The study has been conduced by analysing Mars Global Surveyor Thermal Emission Spectrometer (MGS-TES) data from martian regions with different surface albedo.In support of these data, synthetic spectra with different surface albedo and atmospheric opacities have been computed, so that a comparison has been performed. The synthetic spectra have been retrieved by using two different grain sizes for suspended dust (0.5 and 1.2 μm), allowing a comparison between the two models and the observations.Using the DCI, a parameter describing the quantity of dust deposited on the surface, the effectiveness of the single scattering approximation has been tested for low atmospheric opacity by analysing the quality of the linear fit up to different atmospheric opacity.For more opaque conditions two kinds of fits have been applied to the data, linear and second-order degree polynomial. In this case, we found that the polynomial fit better describes the observations.The analysis of these data made it possible to notice a peculiar trend, already reported by Christensen (1988), of the albedo over Syrtis Major after the occurrence of dust storms, but, differently from that work, now the study of DCI together with atmospheric opacity and albedo allowed us to robustly confirm the hypothesis made by Christensen.Finally, the comparison between observations and synthetic spectra computed with models with different particles grain sizes indicates that dust particles of 0.5 μm diameter are the most effective to change the aerosol atmospheric opacity on Mars.     

A spherical Monte-Carlo model of aerosols: Validation and first applications to Mars and Titan

The atmospheres of Mars and Titan are loaded with aerosols that impact remote sensing observations of their surface. Here we present the algorithm and the first applications of a radiative transfer model in spherical geometry designed for planetary data analysis. We first describe a fast Monte-Carlo code that takes advantage of symmetries and geometric redundancies. We then apply this model to observations of the surface of Mars and Titan at the terminator as acquired by OMEGA/Mars Express and VIMS/Cassini. These observations are used to probe the vertical distribution of aerosols down to the surface. On Mars, we find the scale height of dust particles to vary between 6 km and 12 km depending on season. Temporal variations in the vertical size distribution of aerosols are also highlighted. On Titan, an aerosols scale height of 80 ± 10 km is inferred, and the total optical depth is found to decrease with wavelength as a power-law with an exponent of −2.0 ± 0.4 from a value of 2.3 ± 0.5 at 1.08 μm. Once the aerosols properties have been constrained, the model is used to retrieve surface reflectance properties at high solar zenith angles and just after sunset.     

Properties of cryobrines on Mars

Brines, i.e. aqueous salty solutions, increasingly play a role in a better understanding of physics and chemistry (and eventually also putative biology) of the upper surface of Mars. Results of physico-chemical modeling and experimentally determined data to characterize properties of cryobrines of potential interest with respect to Mars are described. Eutectic diagrams, the related numerical eutectic values of composition and temperature, the water activity of Mars-relevant brines of sulfates, chlorides, perchlorides and carbonates, including related deliquescence relative humidity, are parameters and properties, which are described here in some detail. The results characterize conditions for liquid low-temperature brines (“cryobrines”) to evolve and to exist, at least temporarily, on present Mars.     

Methane adsorption on a martian soil analog: An abiogenic explanation for methane variability in the martian atmosphere

Recent observations suggest methane in the martian atmosphere is variable on short spatial and temporal scales. However, to explain the variability by loss reactions requires production rates much larger than expected. Here, we report results of laboratory studies of methane adsorption onto JSC-Mars-1, a martian soil simulant, and suggest that this process could explain the observations. Uptake coefficient (γ) values were measured as a function of temperature using a high-vacuum Knudsen cell able to simulate martian temperature and pressure conditions. Values of γ were measured from 115 to 135 K, and the data were extrapolated to higher temperatures with more relevance to Mars. Adsorptive uptake was found to increase at lower temperatures and larger methane partial pressures. Although only sub-monolayer methane surface coverage is likely to exist under martian conditions, a very large mineral surface area is available for adsorption as atmospheric methane can diffuse meters into the regolith. As a result, significant methane may be temporarily lost to the regolith on a seasonal time scale. As this weak adsorption is fully reversible, methane will be re-released into the atmosphere when surface and subsurface temperatures rise and so no net loss of methane occurs. Heterogeneous interaction of methane with martian soil grains is the only process proposed thus far which contains both rapid methane loss and rapid methane production mechanisms and is thus fully consistent with the reported variability of methane on Mars.     

Observing the martian surface albedo pattern: Comparing the AEOS and TES data sets

High spatial resolution images of Mars were acquired with the Advanced Electro-Optical System (AEOS) 3.63-meter telescope at the Maui Space Surveillance System (MSSS) during both the 2001 and 2003 Mars apparitions. Comparisons are made of the surface albedo patterns obtained from these AEOS images to the surface albedo maps constructed from the Mars Global Surveyor (MGS) Thermal Emission Spectrometer (TES) data taken during the same time periods. These comparisons demonstrate that the images provide albedo information in a limited area surrounding the sub-Earth point that is consistent with the TES-derived albedo field. Additionally, it is shown that by employing adaptive optics (AO), the typical ground-based observing season of Mars can be extended. This is the only known published AO data set of Mars with temporal coverage over an entire apparition. Changes in the surface albedo affect the local ground temperature, which impacts the depth of the planetary boundary layer (PBL) above the surface. Since it is the state of the PBL that controls surface/atmospheric interaction, albedo variations have the power to alter the amount of dust that is lifted. A one-dimensional radiative/convective version of the NASA Ames Mars General Circulation Model is used to demonstrate that the measured albedo variations can alter the daytime ground temperatures by as much as 5 K, which in turn alters the structure of the planetary boundary layer (PBL). Therefore, albedo changes are thermodynamically important, and the ability to characterize them, should orbital observations become unavailable, is a valuable capability.     

Comparison of HIPWAC and Mars Express SPICAM observations of ozone on Mars 2006-2008 and variation from 1993 IRHS observations

Ozone is a tracer of photochemistry in the atmosphere of Mars and an observable used to test predictions of photochemical models. We present a comparison of retrieved ozone abundances on Mars using ground-based infrared heterodyne measurements by NASA Goddard Space Flight Center’s Heterodyne Instrument for Planetary Wind And Composition (HIPWAC) and space-based Mars Express Spectroscopy for the Investigation of the Characteristics of the Atmosphere of Mars (SPICAM) ultraviolet measurements. Ozone retrievals from simultaneous measurements in February 2008 were very consistent (0.8 μm-atm), as were measurements made close in time (ranging from <1 to >8 μm-atm) during this period and during opportunities in October 2006 and February 2007. The consistency of retrievals from the two different observational techniques supports combining the measurements for testing photochemistry-coupled general circulation models and for investigating variability over the long-term between spacecraft missions. Quantitative comparison with ground-based measurements by NASA/GSFC’s Infrared Heterodyne Spectrometer (IRHS) in 1993 reveals 2-4 times more ozone at low latitudes than in 2008 at the same season, and such variability was not evident over the shorter period of the Mars Express mission. This variability may be due to cloud activity.