P11A-0943 0800h
An experimental crystallization study of a proposed high-Fe low-Al Martian parental liquid
Recent work has started to question the validity of the assumption of high-Fe, low-Al Martian magma compositions as parental to the SNC meteorites, in particular to the Chassigny meteorite. Minitti and Rutherford (2000 GCA Vol. 64) performed experiments on Johnson et al.'s (1991 GCA vol. 55) A* composition at low pressure and did not produce the assemblage of any of the SNC meteorites. In contrast, the crystallizing assemblages of hawaiite with greater the 0.4 wt% water and pressures above 4.3 kbar has been shown to replicate well the Chassigny assemblages (Nekvasil et al. 2004 LPSC 35). Experiments were conducted to investigate whether similar results could be obtained by a low-Al, high-Fe magma crystallizing at higher pressures and water contents than used in earlier work. The Chassigny cumulate assemblage is dominantly olivine with minor cpx, low-Ca pyroxene, and chromite. The polyphase melt inclusions contained within the olivine contain augite, low-Ca pyroxene, kaersutite, apatite, Ti-biotite, chrome rich spinel, and a feldspar glass. In contrast, Minitti and Rutherford's (2000 GCA Vol. 64) experimentally obtained low pressure assemblage for the A* composition of Johnson et al. (1991 GCA vol. 55) is olivine, sub-calcic augite, pigeonite, Ti-magnetite, ilmenite, and plagioclase. Our experimental results indicate that at 4.3 kbar, in the presence of water (0.2 wt% bulk water), the assemblage is pigeonite-dominated with olivine, plagioclase, and ilmenite as additional crystallizing phases. At 9.3 kbar the assemblage continues to be pigeonite dominated, with olivine, and plagioclase. Ongoing experiments are evaluating the direct role of water in affecting the mineral assemblages. These assemblages at 0.2 wt% water do not produce the cumulate assemblage seen in the Chassigny meteorite. Furthermore, since they do not contain kaersutite, apatite, biotite, or low-Ca pyroxene, they differ markedly from the melt-inclusion assemblages of the Chassigny meteorite. Thus, to date no experiments on any high-Fe, low-Al parental liquid have produced the Chassigny mineral assemblages.
P11A-0944 0800h
Vertical Extrapolation of Mars Magnetic Potentials
Mars Global Surveyor measured the most strongly magnetized crust in the heavily-cratered southern hemisphere of Mars. Our analysis concentrates on the magnetic lineations or patterns centered near latitude $40^\circ$S, longitude $180^\circ$W, with a range of values $\pm 40^\circ$, using a rotated Cartesian coordinate system. Using Fourier transforms the magnetic field components at satellite mapping altitude can be vertically extrapolated downward to aerbobraking altitudes and the results are in excellent agreement with measurements at $\sim$100 km and fill in data gaps. We also construct scalar and vector potentials and a streamline function which provide a more succinct and smoother representation of the field. A model constructed with just 8 vertical dipoles accounts for 80% of the variance of the scalar potential at 400 km over the region analyzed, but 14 dipoles can account for only 64% of the variance at 100 km. More than 20 dipoles are required to model the surface field, and the fit is not as good. At any altitude, as more centers are included they begin to overlap, and the improvement in fit slows and then worsens. No obvious relation of the dipole locations and surface geology is apparent. The magnetic field can be represented as the gradient of the scalar potential or the curl of the vector potential. The vector potential can be chosen so that the vertical component is identically zero, and the horizontal components can be represented as the curl of a streamline function. We reduce the vector potential to a two component horizontal field which can be plotted on a map as a set of arrows. The observed vector potential field shows abrupt changes in direction over the analyzed region, suggesting either different stages of magnetization or local demagnetization. In the southern hemisphere region analyzed, the strongest values of the magnetization lie adjacent to ancient multi-ring basins, notably Sirenum, though not actually within its rings.
P11A-0945 0800h
Magnetic Signatures of Martian Volcanoes: Evidence for a Second Dynamo Episode?
Using electron reflection magnetometry, we investigate the magnetic anomaly signatures of 20 Martian volcanoes. Most show magnetic lows characteristic of thermal demagnetization of crust in the presence of no significant global magnetic field. However, one of the oldest volcanoes, Hadriaca Patera, appears to be a magnetic source, suggesting thermoremanent magnetization following its last major period of magmatism. We investigate the likelihood that a global magnetic field was present at this time and find a distinct possibility that such a field was present. We compare the relative ages (according to the Hartmann & Neukum cratering time-scale) and magnetic signatures of these 20 volcanoes and 7 giant impact basins and conclude that, ~300 Myr after the cessation of an early dynamo, a second dynamo episode may have started around ~3.85 Gyr ago and lasted for 100-350 Myr. We speculate that the second episode may have been due to one or more large mantle plumes that, in removing heat from the lower mantle, temporarily increased the core-mantle thermal gradient to restart the core dynamo. We discuss possible geophysical causes for such plumes and consider the implications of a second dynamo, and its associated magnetic shielding, for astrobiology.
P11A-0946 0800h
Tectonic Extension and Bulging in Southern Alba Patera Region, Mars
We investigated the Alba Patera area, south of the Alba Patera Volcano, in order to obtain quantitative information on the tectonic extension affecting this area. Tectonic structures of Alba Patera area have been previously described using Viking images and interpreted as extensional structures. Digital Elevation Model made using altimetry data (MOLA) can be used to quantitatively investigate these structures with a better resolution. We proceeded by constructing E-W oriented profiles and used techniques commonly employed in Structural Geology to estimate the tectonic extension. We investigated the presence of mechanical discontinuities by calculating the maximum depths reached by the graben structures. This has been estimated using the lost-area balancing method (Groshong, 1996) and the crossing depth of the border faults assuming planar fault surfaces, which gave similar results. The extension calculated in the studied area (up to 12%) cannot be explained only with the observed doming of the topographic surface and require a significant amount of horizontal stretching.
P11A-0947 0800h
Fractal Mars: Global Equal-Area Mapping of 0.3 km to 1.2 km Slope Properties
We have completed a global equal-area mapping (sinusoidal projection) of Martian 0.3 km to 1.2 km slope properties using the Mars Global Surveyor (MGS) Mars Orbiter Laser Altimeter (MOLA) dataset. Under the assumption of self-affine surface roughness statistics we find the Hurst exponent that best fits variograms of surface heights measured by MOLA at 0.3, 0.6, 0.9, and 1.2 km intervals. This data type was one form of landing site characterization for the Mars Exploration Rover (MER) Project; the Hurst exponent can be used to extrapolate to 100 m slopes. For MER we carried out site-specific mapping and analysis, and now we have expaned our mapping globally. We have generated sinusoidal projection maps of the variogram values at 15 km resolution (quarter degree at the equator) for all of Mars. Small gaps exist where insufficient MOLA data were available. We compare our maps and technique with previous roughness analyses and mapping efforts that used MOLA data. Our maps will be released shortly to the Planetary Data System (PDS). This work was funded by the NASA's Mars Data Analysis Program, and the work was carried out at the Jet Propulsion Laboratory, California Institute of Technology.
P11A-0948 0800h
Long-Wavelength Shoreline Deformation on Mars
A number of geologic and topographic features within the northern plains of Mars have been interpreted as shorelines formed by ancient oceans. Several recent studies have challenged this interpretation, arguing that the present topographic profiles do not appear to originate from surfaces of equal gravitational potential. Elevations along the ``shorelines'' are especially variable at long wavelengths (thousands of km), with amplitudes of hundreds of meters to kilometers. To test the hypothesis that the features in the northern plains are deformed shorelines, we compare the long-wavelength topography (solid surface position relative to the areoid) of the two most prominent shorelines (the Arabia and Deuteronilus contacts of Clifford and Parker, Icarus, 2001) with the deformation expected for: (1) flexural response of the lithosphere to surface loading from the growth of Tharsis and ocean (i.e., sea level) redistribution, (2) true polar wander (TPW), and (3) dynamic topography linked to internal convective flow. We find that TPW and dynamic topography are both capable of reconciling the longest-wavelength variation in topography (the former is a purely degree two signal). The predicted TPW path that best fits the shoreline record is a function of the effective elastic thickness of the Martian lithosphere and it is consistent with recent inferences based on paleomagnetic evidence. The inference is also compatible with our recent re-analysis of the rotational stability of Mars subject to Tharsis and internal loading, which we briefly summarize.
P11A-0949 0800h
Lava Domes of the Arcadia Region of Mars
It is widely accepted that the volcanic rocks on the surface of Mars are basic to intermediate in composition. A few locations on Mars, however, have been considered by some workers as possibly consisting of silicic volcanic rocks, based primarily on morphologic considerations. If it could be demonstrated that silicic volcanism occurred on the surface of Mars then the poorly-differentiated, basic to intermediate crust model would have to be discarded in favor of a more magmatically diversified one. Past efforts at demonstrating the presence of silicic volcanic rocks at these locations, however, were thwarted by the lack of spatial and spectral resolution of the data available at the time. One of these locations, referred to in this study as the Tyndall Volcanic Field (TVF), in the western Arcadia Planitia region, possesses $>$ 600 small ($<$ 6 km diameter) edifices that morphologically resemble terrestrial lava domes. They consist of massive, steep-sided rock bodies, frequently partly covered by boulders, occasionally possessing what we interpret to be summit collapse features and crease structures and are usually surrounded by rock talus aprons. Curious, between-dome lineaments occur that are tentatively interpreted as being fissures. Our work uses Mars Orbiter Camera (MOC), Mars Orbiter Laser Altimeter (MOLA), Thermal Emission Spectrometer (TES) and Thermal Emission Imaging System (THEMIS) orbital imagery data to study the morphologic, morphometric and thermal IR spectral characteristics of the TVF features, comparing them with possible Earth analogue volcanic features of known lithologies. Work on deconvolving TES and THEMIS spectra is in progress. A main goal is to determine the identities of the lithologies that make up the material of the domes.
P11A-0950 0800h
Lava Flows on Olympus Mons, Mars: Estimates of Flow Speeds and Volume Fluxes from MOC, THEMIS and MOLA Data
This research aims at characterizing the eruptive behavior of Olympus Mons and other major volcanoes in the Tharsis region on Mars. We are particularly interested in estimating flow speeds and volume fluxes for flows emanating from vents at various locations between the summit and the basal scarp. Our methodology involves: (1) identifying lava tubes and channels that are relatively pristine and undegraded from high-resolution (3-10m/pixel) MOC images and lower-resolution, broader-coverage (18-19m/pixel) THEMIS visible images; (2) reprojecting the MOC images into a planetocentric, simple-cylindrical coordinate system to facilitate comparison with the Mars Orbiter Laser Altimeter (MOLA) digital elevation data; (3) measuring the width and depth of tubes and channels on the processed images; (4) constructing a global slope map of Mars from the interpolated MOLA 1/128$\deg$ DEM and using it to determine ground slopes along the lava tubes and channels of interest; (5) calculating flow speeds and total volume fluxes from the Jeffrey's equations based on assumed values of lava viscosity and density and a range of plausible values for lava depth (which has an upper bound of tube/channel depth); and (6) identifying correlations between flow speeds and/or total volume fluxes and the presence of tubes vs. channels. A key challenge lies in making accurate measurements of the depths of lava tubes and channels. Flow speed and total volume flux respectively have quadratic and cubic dependences on lava flow depth (which is estimated from tube or channel depth), so any errors in depth estimates are not readily forgiven. We are currently exploring different methodologies to get the best depth estimates. To date, we have width and depth estimates of 42 relatively narrow (typically $<$100m diameter) lava tubes and channels from MOC images near the summit and on the flanks of Olympus Mons. We will expand this dataset to include larger lava tubes and channels on Olympus Mons from THEMIS images, as well as tubes and channels of various sizes on the other major Tharsis volcanoes (Arsia, Ascraeus and Pavonis Mons) from both MOC and THEMIS images.
P11A-0951 0800h
The evolution of volcanic material on Mars: Preliminary results of sand-lavas relationships from the analogy with sandy lavas in Iceland
The surface of Mars is covered by volcanic rocks from few tens of millions years to 3.5 by old. The presence of water and atmosphere can strongly affect these rocks, by both chemical and mechanical erosion and transport. The interpretation of multispectral and hyperspectral data of Mars requires a better comprehension of these surface processes in order to understand if the spectral data still corresponds to the volcanic composition at the time of formation. Volcanic material in Iceland is a good analog for the studies of possible landforms resulting from the formation, transport and deposition of basaltic sand on Mars. Iceland is amongst the unique places on Earth with a cold environment, abundant basaltic rocks and sands, and the presence of palagonite, a possible typical constituent of the Martian soil. A first field campaign has been achieved in fall 2003, with the objectives of sites selection and chemical analysis of sands and lavas in order to establish the sources of sands, and the mineralogical and chemical evolution from lava to sands. The first site is close to Skjalbreidur volcano, south of Langjokull and is composed of weathered lava blocks, sands and gravels. The second sampling site is close to Eldborgir volcano, also south of Langjokull, weathered lava flows and sands are observed here. The third sampling site is around Hekla volcano. The results of the chemical analysis indicate different situations for the origin of sands. For the first two sites, major, minor and traces elements are correlated and indicate that the sands, which are basaltic in composition, are genetically related to the surrounding lava. The sands at Hekla volcano, andesitic in composition, indicate a contamination of material eroded from basaltic lava flow by a more silicic component erupted from Hekla. Sands coming from different sources, of possibly different chemical and mineralogical composition, and of different nature of eruption can easily mix each other which has implications for the interpretation of infra-red data of the surface of Mars. A second result concerns the evolution of the mineralogical composition of basaltic sand compared to the lava. We observed a higher concentration of MgO and Ni in Skjalbreidur and Eldborgir sands than in the surrounding lava taken as a reference. Together, these observations indicate a higher concentration of olivine in the sands which may be due to its higher strength (compared to feldspaths and pyroxene) and sorting by wind from different grain size. On the other hand, the contribution of weathering seems not have destructed these olivine grains. Indeed, magnetic results show that magnetic phases such as titanomagnetite are poorly weathered despite being at the surface since 9000 years. The weathering by the wet climate is likely slow down by the cold temperatures all the year long. The detection of olivine at the surface of Mars is thus not a simple tool to conclude that the weather did not involve liquid water.
P11A-0952 0800h
The peridotitic laterites of New Caledonia: a possible analogue for the Martian regolith ?
Peridotites are very common in the Solar System where they generally form the mantle of rocky planets. On Earth, large volumes of peridotites are brought to the surface as obduction ophiolites (Oman, Newfoundland, Papua, New Caledonia). Such peridotites interact with various aqueous fluids throughout their geologic history. High temperature serpentinization (up to $600\deg$C) begins at the ocean ridge with seawater, followed by low temperature serpentinization (below $400\deg$C) away from the ridge. The obduction process generally leads to further alteration by continental aqueous fluids along structural discontinuities. Finally, exposure to the surface results in supergene serpentinization, the formation of weathering rinds, and the development of soils on peridotites that, particularly under hot and humid climates, leads to the formation of peridotitic laterites. The peridotitic laterites of New Caledonia provide excellent opportunities for geoscience investigations because, due to nickel mining, borehole cores through alteration profiles are available. The peridotite nappe, emplaced in the late Eocene, has an estimated thickness up to 3 km and is exposed over an area of 8000 km$^{2}$. Common lithologies include harzbugite, dunite, wherlite and cumulate gabbro. Regardless of the bedrock lithology, the lateritic profiles display an increase in concentration of Fe oxides and hydroxides towards the top. The alteration paths of distinct bedrock types tend to converge toward the same mineralogy. Low-field and high-field magnetic techniques have been combined to refine the vertical zonation of the lateritic profiles into distinct horizons. Magnetite is the main component in the magnetic susceptibility budget while hematite controls the coercivity. Several distinguishing features are observed in peridotitic laterites: 1) bulbous hematite concretions, 2) reddish fine-grained hematite, 3) gray flaky specular hematite (up to 10 mm in length), 4) ferricretes and manganocretes, and 5) preservation of olivine grains in spite of intense alteration in zones of high relief. Similar characteristics have been deduced from the MGS TES data or observed directly by the Mars Rovers. These results highlight the possibility that some of Martian surface features might have been inherited from a wet past and that some of the regolith may have developed from altered peridotites.
P11A-0953 0800h
Theoretical Radiative Transfer Emissivity Modeling of Quartz Fines At Nadir Incidence: Sensitivity Studies With Applications to Planetary Regolith
When scattering grains are in near or close contact, as is the case for planetary fines measured by thermal IR spectrometers aboard orbiters and rovers, the grains no longer scatter light like single particles. This is problematic for radiative transfer (RT) analyses of the photometric properties of surfaces such as the Martian regolith. After 10 years of advances within the planetary community, RT models are still inadequately predicting nadir emissivity values measured for laboratory geologic samples, calling even the most fundamental of theoretical assumptions into question. Here we present a continuation of sensitivity studies performed to quantify the effects of single scattering albedo and asymmetry parameter on emissivity for packed and unpacked model alpha quartz fines. Using hybrid computational solutions (Mie theory $+$ discrete ordinates RT algorithm) to calculate theoretical nadir emissivity values in the thermal IR, we upgrade numerical implementation of previous works and explore assumptions at different stages in the transfer theory to show that assumptions about the scattering grain itself, rather than the machinery of the computational method used, result in the mismatch between model and lab data. To facilitate connections between laboratory geology and theoretical physics studies, we also present ``the theorist's checklist'' of information most useful to RT modelers to incorporate real world geology into model parameters, as well as specific methods to quantitatively determine particle size distribution that are beneficial to both camps. This work is supported through NASA MDAP (MJW, KMP).
P11A-0954 0800h
Removing the Shock from the Thermal Emission Spectra of Shocked Terrestrial and Martian Basalts
An abundance of impact craters on the martian surface and shock effects in the martian meteorites indicate that the surface of Mars has been shocked. The thermal infrared (TIR) spectra of plagioclase feldspars experimentally shocked to various pressures have enabled the amount of shock to be correlated with changes in the TIR spectra [Johnson et al., 2002, 2003]. With these, estimates on the amount of shocked plagioclase feldspar on the martian surface has been investigated [Johnson et al., in press]. However, in relation to the use of shocked plagioclase feldspars as end-members for the deconvolution of remote TIR data, similar work need to be performed on the laboratory spectra of shocked rocks and subsequent deconvolutions with these shocked mineral end-members. In this work, laboratory TIR spectra of a shocked terrestrial basalt and it unshocked counterpart are deconvolved and compared to assess the possibility of removing the spectral contribution of shocked plagioclase feldspar from a shocked martian basalt spectrum. It has been suggested that the modal abundances of martian meteorite Los Angeles [Rubin et al., 2000] are similar to mineral abundances derived from the deconvolution of Thermal Emission Spectrometer (TES) Surface Type 1 (ST1) [Bandfield, 2002]. TIR spectra of shocked basalt from Lonar Crater, India were collected and deconvolved with an end-member library containing experimentally shocked calcic plagioclase feldspars [Johnson et al., 2002]. The deconvolution-derived mineral abundances were compared to those from deconvolutions of unshocked basalt, and it was determined that the spectrum of the original, unshocked bulk rock could be replicated by removing from the spectrum of a shocked basalt the shocked plagioclase end-member spectrum (scaled by its abundance). The same methodology was performed on another shocked basalt, martian meteorite Los Angeles. The TIR spectra of Los Angeles was deconvolved with an end-member library containing shocked plagioclase feldspar (An75), revealing an abundance of maskelynite (~46%) that closely matches modal abundances from petrographic studies (~45%) [Rubin et al., 2000]. TIR spectra of shocked intermediate calcic plagioclase feldspars such as those found in Los Angeles (An56-38) do not exist and therefore are not contained within the spectral library used here. A mean percentage of 46% of the shocked plagioclase feldspar end-member was removed from the bulk rock spectra of Los Angeles and replaced with unshocked labradorite. Similar to the Lonar Crater basalts, this should recreate the spectra of the pre-impact, unshocked basalt. This end-member replacement results in a TIR spectrum of a basalt with a Christiansen feature moved to lower wavenumbers, agreeing with previous work that suggested the position of this feature varies with the amount of shock [Johnson et al., 2002]. The new "unshocked" Los Angeles TIR spectrum is not an exact match for ST1, but it is more similar to TES spectra than a shocked Los Angeles spectrum is. Whereas laboratory TIR spectra of shergottites provide poor matches to orbital TES data [Hamilton et al., 2003], using this unshocked basalt spectrum as an end-member might provide additional constraints on the source region of Los Angeles. Further, it is suggested here that TIR spectra be acquired for various experimentally shocked end members of plagioclase feldspar solid solution series for better deconvolutions of shocked rocks.
P11A-0955 0800h
Identifying varnished rocks on Mars using thermal infrared spectroscopy
Thermal infrared (TIR) spectroscopy is widely implemented in attempts to determine the composition of the Mars's surface. Discoveries include basaltic rocks, possible andesites, and hematite-rich terrains associated with areas of probable hydrothermal alteration [Bandfield et al., 2000; Christensen et al., 2001; Glotch et al., 2004]. Some of the basaltic rocks appear to be covered by either a weathering rind or a varnish. The presence of a varnish would be interesting because it is believed to form through multiple wetting and drying events [reference]. The presence of these coatings can potentially be identified through unique nonlinear effects where both the substrate and varnish have strong spectral features. For example, varnished terrestrial quartz-rich rocks have a low-emissivity ~8.4-micron reststrahlan band diagnostic of a silicate-rich substrate which remains present while the longer wavelength reststrahlen band is obscured by the clay-rich varnish. In general, this non-linearity will conform to the Beer-Lambert Law, with additional reflection and scattering terms, so that the light emitted from the varnished stone will be similar to I=I$_{o}$ e$^{-ax}$, where `I$_{o}$' is the light emitted from a bare substrate, `a' is the absorption constant for the varnish coating, and `x' is the thickness of the coating. If the effect were linear, as expected for dusty surfaces [Johnson et al., 2002] or discrete patches of rock and clay, the emissivity of the emitted light would, at all wavelengths, possess equal contributions from the varnish and substrate; thus the clay feature would not completely dominate the longwave reststrahlan band without also erasing the shortwave reststrahlan band. After having theoretically determined a nonlinear at some wavelengths is probable, we have focused on laboratory spectral analyses of terrestrial varnished rocks. We have collected over 100 varnished stones from various pavements and unvarnished stones from other surfaces and have acquired over 200 spectra from which we have developed a preliminary, empirical TIR spectral algorithm for the detection of varnished rocks. Developed for terrestrial rocks, the current algorithm relies on 18 bands, each ~ 50 to ~100nm wide, in the 8-12-micron region. We will modify this model for best detecting nonlinear mixing of light emitted from the substrate basaltic rocks covered by a varnish that is spectrally dominated by the ubiquitous Martian dust.
P11A-0956 0800h
Evidence for high-silica, sulfate-bearing sedimentary rocks in Acidalia Planitia from TES and THEMIS data
Thermal infrared spectra of some layered materials in Acidalia Planitia suggest a mineralogy dominated by high-silica-poorly crystalline materials, intermediate-composition feldspars, and sulfates. These spectra are similar in shape to the Acidalia Planitia-type global spectral component (ST2) previously identified by Bandfield et al. [2000], but are more sulfate- and silica-rich than ST2. The mineralogy in Acidalia Planitia is equivocal and could be explained through a variety of contexts, including igneous rock compositions, volcanicastic sedimentary rocks, or alteration products related to rock surfaces (rather than bulk-rock compositions). If the materials indicate unaltered, primary igneous mineralogy, it would suggest the presence of intermediate-felsic (dacitic) lavas or pyroclastics. While a volcanic origin cannot be ruled out, the observation of widespread layering at a vertical scale of 10s of meters, lack of observed volcanic landforms, and location in a basin that is down-slope from fluvial channels all point to a sedimentary origin for these units. The spectral signature of surfaces in Acidalia Planitia require a high-silica (Si/O $>$0.4) material to be present which, though spectrally similar to felsic volcanic glass, could also be explained as aluminous-ferric opaline silica or an intimate mixture of pure opal with allophane, clay minerals, zeolites, or possibly other aluminosilicates. The high-silica materials in Acidalia Planitia appear to be tied to relatively high thermal inertia surfaces, which are overlain by a lower thermal inertia, discontinuous, dissected, and possibly ice-rich surface layer. We propose that the observed surface compositions are related to either: 1) chemical weathering products, possibly related to recent periglacial environmental conditions, or 2) a bulk component (cement?) within a sedimentary layer that was deposited during episodic flooding. The poorly crystalline nature of high-silica materials and lack of evidence for abundant crystalline, indurated clay minerals are both consistent with formation in a short-lived aqueous sedimentary system or with formation of weathering products in low-water surface conditions. In either case, the explanation of these surface materials has major implications for the interpretation of global TES surface compositions. It is intriguing that the silicate and sulfate spectral components of some Acidalia Planitia surfaces are similar to those of the sulfate-rich rocks at the Mars Exploration Rover landing site in Meridiani Planum, which are interpreted as aqueous sediments. It is possible that Meridiani-type rocks (minus the hematite) are present in the northern plains of Mars and related to episodic flooding throughout Mars' history.
P11A-0957 0800h
A Global View of Martian Surface Compositions With MGS-TES Revisited: Regional-scale Spectral Deviations From Surface Types 1 and 2
Previous global studies of martian low-albedo regions with MGS-TES data have identified 2 global spectral endmembers: Surface Type 1, which has been interpreted as basalt to basaltic-andesite, and Surface Type 2, which has been interpreted as either basaltic-andesite to andesite or basalt with alteration minerals or coatings. Using the 2 global endmembers plus the Meridiani Planum hematite spectrum, all low-albedo surfaces can be modeled with spectral RMS errors $<$0.5% at a spatial resolution of 1 pixel per degree (ppd), indicating that there are no regions of extreme spectral differences, such as those that would be caused by a large exposure of carbonate. However, spectral shape deviations from either global endmember that might be caused by small but geologically-significant differences in the characteristic mineralogy of Surface Types 1 and 2, such as a 10% increase in pyroxene abundance, would still be modeled reasonably well (RMS error $<$0.5%) with only the global endmembers. Model fits of spectra from regions that were previously mapped as mixtures of Surface Types 1 and 2 are often greatly improved if a set of mineral endmembers is used instead of the 2 global endmembers, indicating that Surface Types 1 and 2 may not account for all low-albedo regions on Mars. We have re-examined TES spectra from martian low-albedo regions to identify subtle spectral deviations from Surface Types 1 and 2 and found that in some regions, deviations occur in widespread but coherent distributions. Three regions have been found to have deviations from these spectral shapes across multiple orbits; the characteristic spectral shapes of these regions cannot be well-modeled using only Surface Types 1 and 2. Note that these regions are different than the local-scale exposures of olivine- and quartz-rich surfaces that have been previously-reported. The differences in the spectral shapes of these regions indicate that they may differ in primary mineralogy or alteration history from areas that exhibit the typical spectral character of Surface Types 1 and 2. When the characteristic spectral shapes of these regions are added to the global endmember set and used to model surface emissivity at 1 ppd, the distributions of Surface Types 1 and 2 are refined. Geologic context is an important consideration when interpreting the high-silica component of surfaces that exhibit Surface Type 2 spectral character (e.g. primary glass vs. secondary amorphous silica or zeolites); this work will help to clarify the distributions of the previously-identified surface types, and will identify the presence and distribution of newly-distinguished mineralogic assemblages.
P11A-0958 0800h
Mineral Classification of the Martian Surface Using THEMIS Multi-Spectral Infrared Imagery
Recent advancements in multi-spectral imaging and image analysis techniques have greatly enhanced our ability to do planetary research. Much has been discovered about Mars through recent missions such as Mars Global Surveyor, 2001 Mars Odyssey, and the Mars Exploration Rovers. The Thermal Emission Spectrometer on board the Mars Global Surveyor has allowed the mapping of surface mineralogies on Mars at several kilometers scale through hyperspectral imaging [1]. Here, we use the high resolution multi-spectral imagery of THEMIS (THermal Emission Imaging System) on board the 2001 Mars Odyssey to identify different mineral classes at spatial scales of hundreds of meters. THEMIS contains two independent multi-spectral imaging systems: a 10-band thermal infrared imager (IR) with a resolution of 100m/pixel, and a 5-band visible imager with a resolution of 10m/pixel. Here we will use the IR data. The 9 IR bands are centered from 6.8 microns to 14 .9 microns [2]. Using Arizona State University's online spectral library[3], we have been investigating the extent to which we can differentiate between different mineral classes. By identifying certain mineral classes we can better understand the geologic processes which created them and detect areas of interest for further study. Linear mixing of minerals and dust is investigated to estimate ratios of minerals and their resulting spectra. We then compare these spectra to observations of several regions on Mars. We compare these results with TES data and previous mineralogical maps. [1] Christensen et al, (2001) JGR 106, E10; [2] Christensen et al, (2002) Space Science Reviews 110, 1; [3] Christensen et al, (2000) JGR 105, E4
P11A-0959 0800h
Detailed Investigation of a Globally Unique, Orthopyroxene-Rich Deposit in Eos Chasma, Mars
Globally unique, ALH 84001-like (orthopyroxene-bearing) materials in Eos Chasma first were identified by the author and her coworkers in a search for Martian meteorite spectral signatures in MGS TES infrared data [1]. This is the only spatially significant detection of concentrated orthopyroxene discovered on Mars to date. These materials have complete THEMIS infrared coverage (day and night) covering multiple seasons and nearly 100 percent coverage by the visible subsystem. This presentation will describe the detailed mineralogy and morphology of these materials as observed by TES and the Mars Odyssey THEMIS, as well as related data from MOC and MOLA. Because pyroxenes are relatively susceptible to aqueous weathering, the abundances of secondary minerals will be of particular interest. THEMIS visible and infrared images show that the orthopyroxene spectral features are associated with a geomorphic unit that likely is a debris flow or lobate ejecta deposit. In either case, these materials appear to superpose Eos Chasma floor units, likely were mobilized from a location within the Valles Marineris wall rock, and may represent an orthopyroxene-rich igneous unit in the lower portions of the wall rock stratigraphy. Because THEMIS images provide information about this unit in areas where there are gaps in TES data, new estimates of the area covered by this material are on the order of ~1250 km2, or at least twice the size estimated by [1]. The nighttime temperatures of these materials are consistent with coarse, sandy materials, but likely represent an average of coarser and finer particles. [1] Hamilton V. E., P. R. Christensen, H. Y. McSween Jr., and J. L. Bandfield, Meteor. Planet. Sci., 38, 871-886, 2003.
P11A-0960 0800h
Thermophysical Characterization of the Tharsis Plateau Low-Inertia Layer from TES and THEMIS Observations
The surface of Mars can be divided into a few major surface types based on thermal inertia, albedo, and rock abundance measurements. The surfaces on the planet with the lowest thermal inertia, lowest rock abundance, and relatively high albedo represent a very fine-grained, poorly indurated, high albedo material. In order to completely blanket the geologically varied underlying materials and display such uniformly low thermal inertia, the surface layer must be at least a few centimeters thick, and is likely to be much thicker. The Tharsis-Amazonis region contains some of the brightest non-ice surfaces on the planet, often with morphologies that are uncoupled from the albedo and thermal inertia trends, suggesting that a regional or global process is responsible for the development of the relatively uniform low inertia layer. The Thermal Emission Spectrometer (TES) has provided well-calibrated datasets for characterizing the thermophysical properties of this low-inertia materials. THEMIS thermal infrared and visible images have been used to study the link between surface morphology and small-scale thermophysical characteristics, providing a better understanding of the physical nature of the units differentiated with TES observations. There are a few distinct surfaces in the Tharsis region that have anomalously low albedo and high thermal inertia, including sites in Noctis Labyrinthus, the Arsia Mons/Oti Fossae thermal anomaly, the scarp of Olympus Mons, and a small number of isolated impact craters throughout the region. These anomalous surfaces typically have distinct boundaries and contain evidence that they have resulted from current or recent aeolian activity constrained by morphologic features. The sites investigated provide insight into the uniform, widespread, low inertia unit, but most are not actual windows into underlying material. Dunes and other distinct morphologic features can be seen on top of the low intertia layer. The relationships seen among all the high inertia features in Tharsis provide some evidence that the low inertia unit may be more competent than previously determined, and probably significantly more complicated.
P11A-0961 0800h
A Large-Scale Atmospheric Equalization Technique for THEMIS Data
Spaceborne thermal infrared observations of Martian surface emissivity using the Mars Odyssey THEMIS instrument can offer a revealing look at the mineralogical makeup of the planet at unprecedented spatial resolution. In areas of high topographic relief, however, varying atmospheric path length makes it difficult to qualitatively or quantitatively compare spectra at different elevations. We describe a technique that normalizes the atmospheric signature across the image, minimizing the effects of path length. As expected, the amplitude of the required normalization is shown to vary seasonally, suggesting that atmospheric dust plays an important role in the extinction. A lack of this effect at 12.57$\mu$m appears to exclude water vapor or ice crystals as a major component of the phenomenon.
P11A-0962 0800h
Variation of Non-Condensables in the Martian Atmosphere
The Martian atmosphere is dominated by CO{_2}, but a small fraction is made up of non-condensables, predominantly Ar and N{_2}. These non-condensables make up about 4.7% of the atmosphere, but this fraction is likely not constant over the seasonal cycle or spatially. The reason for this is that the major gas, CO{_2} condenses in the autumn pole and sublimes at the spring pole, concentrating and diluting, respectively, the non-condensable mass-mixing ratio. The purpose of this study has been to examine the seasonal cycle of non-condensables with a Martian General Circulation Model (GCM) to provide a basis of comparison with data becoming available from the Mars Odyssey spacecraft, and to provide insight into the evolving non-condensable distributions. The model includes a full seasonal cycle of CO{_2} and atmospheric dust.
P11A-0963 0800h
Mesospheric Clouds on Mars in Nadir-Pointed THEMIS-VIS Images
We present images of very-high-altitude clouds obtained at near-zero emission angles by the Mars Odyssey spacecraft's Thermal Emission Spectrometer visible subsystem (THEMIS-VIS). Although THEMIS-VIS was not designed or intended for stereo imaging, the parallax effect caused by its multiple exposure color-imaging sequence allows the height of clouds with sufficient spatial contrast to be determined with a precision of 5-10 km. To determine cloud height in a given image, we make an initial guess of the height, reproject the subframes that make up the image onto a surface of that elevation, and then examine the reprojected imagine to determine whether the cloud features are properly aligned in overlapping subframes. This process is repeated until the best fitting elevation is found. The precision of the height estimate obtained in this manner is obviously limited by the sharpness of the cloud features. Although clouds and hazes are common in the THEMIS-VIS data set, images in which a cloud height can be measured are extremely rare. To date, we have only two detections of equatorial high-altitude clouds, both of which occur (possibly coincidentally) over high surface elevations in the eastern Tharsis bulge, but in quite different seasons --- $\mathrm{L}_\mathrm{s}$ $26^{\circ}$ and $\mathrm{L}_\mathrm{s}$ $114^{\circ}$. In both cases, the detections are in late-afternoon images, near 4:30 PM local solar time, the cloud heights are near 75 km above the local surface, and the cloud structure is wispy and lineated. Our detections are consistent with limb observations by TES (solar band) and MOC (Wide Angle) of high altitude clouds (Clancy et al., 2004, DPS). Clancy et al.~find that these clouds appear to be present only within a very restricted range of longitudes and seasonal intervals. In addition to the equatorial cloud measurements, we have also in a few instances been able to measure heights of 60-70 km for mid-latitude clouds near the fringes of the polar hood in early northern winter. This measurement requires the serendipitous confluence of unusually distinct cloud features and favorable lighting conditions, and so, at this time, it is difficult to assess the prevalence or nature of high altitude clouds in this region and season.
P11A-0964 0800h
3D simulations of the early Mars climate with a General Circulation Model
The environmental conditions that existed on early Mars during the Noachian period are subject to debate in the community. While some authors suggest that the difference resulted from a stronger geothermalism during that period, and that a warm climate was not necessary to explain the valley network, others claim that a warm, wet early climate capable of supporting rainfall and surface runoff is the most plausible scenario for explaining the entire suite of geologic and mineralogical features in the Martian cratered highlands. The recent observations from the Mars Exploration Rovers, Mars Express and Mars Odyssey are more than ever raising the issue. To help understand this key issue in Mars science, we have developped a 3D general circulation model similar to the one used on current Earth or Mars to study the details of the climate today. Our first objective is to answer the following questions : how is the Martian climate modified if 1) the surface pressure is increased up to several bars (our baseline: 1 or 2 bars) and 2) if the sun luminosity is decreased by 25% like 3.8 Billion years ago. We did not take into account the heat possibly released by impacts during short periods, although it may have played a role. Preliminary results obtained assuming a 2 bars atmosphere suggest that, even without taking into account the radiative effect of CO$_2$ clouds, temperature near or above the freezing point of water may be obtained. In particular, the CO2 gas greenhouse effect may be stronger than usually expected because pressure induced absoprtion may have been underestimated in studies Moreover CO$_2$ ice clouds are found to form almost everywhere on the planet in the upper atmosphere above 40 km. Their radiative effect on the climate is very model dependent but, in our baseline simulation, correspond to a15 K warming of the surface. We found that above freezing temperatures are especially likely in the lower plains, due to the atmospheric adiabating warming. This would support a scenario were the lower northern plains would be filled by liquid water and the higher regions more or less covered by snow and glaciers. To better investigate the kind of water cycle one can expect in such conditions, we plan to apply parameterisations currently used in Earth models to simulate the water cycle on our planet.
P11A-0965 0800h
Latitudinal Distribution and Seasonal Behavior of Mars Ozone from Infrared Heterodyne Spectroscopy
Ozone is an important tracer of the photochemical processes responsible for the stability of Mars' CO$_2$ atmosphere. Infrared heterodyne spectroscopy with a resolving power of $1-5 x 10^6$ has been used to directly measure fully-resolved lineshapes of Mars atmospheric O$_3$ features from the ground. Observations at 9.6 microns have been made at Mars apparitions from 1988 through 2003 to directly measure O$_3$ as various latitudes and local times. Carbon dioxide features are also present in the bandpass, providing temperature constraints during the radiative transfer analysis of the O$_3$ lines. We present ozone abundance as a function of latitude for various Mars seasons (L$_S$ = 40, 73, 102, 115, 202, 204, 291). The latitudinal behavior of the ozone is compared to that of water taken close in time in order to investigate the anti-correlation predicted by photochemistry. This work was supported by the NASA Planetary Astronomy Program.
P11A-0966 0800h
Representing Aerosols in Mars Climate Models
The study of aerosols is of prime importance for Mars Climate. Martian aerosols, a generic term for the airborne dust and the water ice crystals, and their related microphysical processes have a strong influence on the radiative budget and on the CO2 and the water cycles. For instance, previous studies have already stressed the necessity to understand how clouds and dust interact with each other and how eventually those interactions may lead to interannual climatic varability. Because aerosols are intimately tied to climatic issues, their study takes place in the context of General Circulation Model (GCM) simulations. Various treatments of microphysics have been introduced in Mars GCMs so far. Yet, it is still not known how these various schemes compare with each other. We have performed a suite of experiments using the NASA Ames General Circulation Model and several microphysical packages. This study aims at determining the most relevant formulation of mycrophysics for use in Mars climate models. Two criterions are considered: (i) the accuracy of each method and (ii) its computational cost. We show that dust and water ice clouds can be represented sufficiently by only two moments: mass and number. This simple formulation allows for fast computations, while remaining accurate enough with respect to more sophisticated schemes which suffer from prohibitive computational costs. However, we also show that using only one moment for a specific aerosol type, a method that has been widely employed within the community, results in a significant degradation of the predictions.
P11A-0967 0800h
Atmosphere lost by oxidation of the surface in the geological history of Mars
In this work I made an evaluation of the quantity of atmosphere lost by Mars by oxidation of the surface in his geological history. In a previous work I calculate the quantity of igneous rocks generated in the diverse Martian epochs. In this work I use that previous calculus and calculate the quantity of oxygen that reacts with these new rocks. The maximum oxidation of the soil occurred in the middle Noachian and the present rate of oxidation is the minimum.
P11A-0968 0800h
Martian Shock and Magnetic Pile-up Boundary Positions and Shapes Determined from the Phobos 2 and Mars Global Surveyor Data sets
A great many Martian bow shock and magnetic pile-up boundary crossings have been identified in the Phobos 2 and Mars Global Surveyor, MGS, data. From these observations the positions and shapes of the bow shock and magnetic pile-up boundary, MPB, have been derived and modelled, using curve-fitting techniques. The models thus derived separately from the Phobos 2 and MGS data sets do not differ drastically, despite the different time and space data coverages. The purpose of the presentation is therefore to show the results obtained from the mixing of the Phobos 2 and MGS data bases and to compare the derived bow shock and MPB models with the ones obtained previously. The underlying objective was to see whether it was possible to determine improved bow shock and MPB models or not. The answer is definitely yes, and particularly for the MPB, thanks to the complementary nature of the Phobos 2 and MGS observations. The boundaries crossed close to the subsolar direction or mostly far downstream by Phobos 2 indeed allow a better coverage of the Martian space environment to be considered. Nevertheless, in order to reduce the domination of the overabundant MGS data set and/or the crossings that are close to Mars, weighting factors have been introduced.