MER Results and Mars Surface
Presiding: M P Golombek, Jet Propulsion Laboratory, California Institute of Technology; W Luo, Northern Illinois University
P32A-01 10:30h
Search for the Origin of Hematite at Meridiani Planum and Gusev Crater
The landscape in Meridiani Planum encountered by the Rover Opportunity is different to all previous Mars landing sites. While those locations look like typical unsorted alluvials, Meridiani Planum consists of sorted sands with aeolian features like small dunes and desert pavements in places. Chemical compositions of soils and rocks at Gusev crater and Meridiani Planum were measured by the Alpha Proton X-Ray Spectrometer (APXS) [1, 2]. At Meridiani Planum all soils and outcrops have a higher mean Fe/Si ratio of 0.75 compared to rocks and soils in Gusev crater with a mean Fe/Si of 0.57. The enrichment of Fe results from an admixture of hematite (Fe2O3) as determined in-situ by the Mössbauer spectrometer (MB) [3]. The formation of hematite is an indicator for aqueous activities under oxidizing conditions. The highest portion of this mineral was found in the spherical grains, also nicknamed `blueberries', which cover most place at the landing site. These spherules were also found in rock exposures in Eagle crater to about 2 % by volume and were interpreted as concretions that formed by precipitation from aqueous fluids inside sedimentary rocks [4]. At Gusev crater no hematite was observed until sol 90 except for layering on a rock. However, about three months later at the foot of the Columbia Hills the MB detected hematite in a rock, dubbed `Pot of Gold'. Our investigations of hematite bearing materials, measured by APXS, MB, and Microscopic Imager (MI) [5], provide an integrated view of different occurrences of hematite on the Martian surface. Ratios of Fe to Mn are compared with Fe concentrations for various soils and outcrops in Meridiani Planum and Gusev crater. Most samples cluster at a mean Fe/Mn ratio of about 50 and range in Fe from 12 to 17 wt. %. Exceptions are found for those Meridiani Planum soils that have very high Fe contents of about 26 wt. %, such as targets dubbed `JackRussell', `FredRipple', and `Berry Bowl full', all showing Fe/Mn ratios of about 110. Based on APXS measurements we cannot distinguish, whether spherules consist of pure hematite or carry a thin layer of hematite. All these high hematite bearing soils are top surface samples, while corresponding subsurface soil samples or soils disturbed by rover wheels have low hematite contents. The very high Fe/Mn ratios of three undisturbed samples together with very high hematite contents suggest the presence of a hematite-rich top layer irrespectively of shape and area coverage of spherules or fragments and could be interpreted as a surface coating similar to terrestrial surface coatings. In the hematite rich outcrops with the same Fe/Mn ratio as found for the soil samples the formation of the main portion of fine dispersed hematite must be an isochemical re-crystallization process under strongly oxidizing conditions. [1] Gellert, R. et al. (2004) Science, 305, 829-832. [2] Rieder, R. et al. (2004) Science, 306, 1746-1749. [3] Klingelhöfer, G. et al. (2004) Science, 306, 1740-1745. [4] Squyres, S, et al. (2004) Science, 306, 1698-1703. [5] Herkenhoff, K. E. et al. (2004) Science, 306, 1727-1730.
P32A-02 10:45h
Assessment of Mars Exploration Rover Landing Site Predictions
Comprehensive analyses of remote sensing data during the 3-year effort to select the Mars Exploration Rover landing sites at Gusev crater and Meridiani Planum correctly predicted the safe and trafficable surfaces explored by the two rovers. Gusev crater was predicted to be a relatively low relief surface that was comparably dusty, but less rocky than the Viking landing sites. Available data for Meridiani Planum indicated a very flat plain composed of basaltic sand to granules and hematite that would look completely unlike any of the existing landing sites with a dark, low albedo surface, little dust and very few rocks. Orbital thermal inertia measurements of 315 J m-2 s-0.5 K-1 at Gusev suggested surfaces dominated by duricrust to cemented soil-like materials or cohesionless sand or granules, which is consistent with observed soil characteristics and measured thermal inertias from the surface. THEMIS thermal inertias along the traverse at Gusev vary from 285 at the landing site to 330 around Bonneville rim and show systematic variations that can be related to the observed increase in rock abundance (5-30%). Meridiani has an orbital bulk inertia of ~200, similar to measured surface inertias that correspond to observed surfaces dominated by 0.2 mm sand size particles. Rock abundance derived from orbital thermal differencing techniques suggested that Meridiani Planum would have very low rock abundance, consistent with the rock free plain traversed by Opportunity. Spirit landed in an 8% orbital rock abundance pixel, consistent with the measured 7% of the surface covered by rocks >0.04 m diameter at the landing site, which is representative of the plains away from craters. The orbital albedo of the Spirit traverse varies from 0.19 to 0.30, consistent with surface measurements in and out of dust devil tracks. Opportunity is the first landing in a low albedo portion of Mars as seen from orbit, which is consistent with the dark, dust-free surface and measured albedos. The close correspondence between surface characteristics inferred from orbital remote sensing data and that found at the landing sites argues that future efforts to select safe landing sites will be successful. Linking the five landing sites to their remote sensing signatures suggests that they span most of the important, likely safe surfaces available for landing on Mars.
P32A-03 11:00h
Spirit's Traverse to the Columbia Hills: Systematic Variations in Clast Morphometry and Texture of Pebble to Cobble Sized Clasts, With Implications for Geological Processes and History
During the course of Spirit's traverse from the Columbia Memorial Station to the Columbia Hills a systematic set of PanCam observations called the clast survey were taken to look for evidence of fluvial activity affecting the morphology of pebble to cobble sized material. These PanCam observations employed a single frame, blue filter shot at 4 bits/pixel looking just above the deck in front of the rover at an angle centering the frame at -72 degrees. These images were taken at 42 sites during the course of the traverse from the landing site to the base of the Columbia Hills. This traverse encountered approximately 6 different geological units that were inferred from orbital data including thermal inertia estimates from Odyssey's THEMIS instrument, as well as geomorphic features observed by the Mars Global Surveyor's Mars Orbital Camera (MOC). Clast survey observations enabled quantification of changes in the size, roundness, sphericity, sorting, density (clasts/meter2), dispersion (nearest neighbor distances) and vesicularity of clasts over the course of Spirit's traverse across the plains to the base of the Columbia Hills. The overall goal was to look for trends in the above parameters that could allow an objective discrimination between basic erosional/depositional processes, including impact, fluvial, debris flow, glacial, and aeolian. To assist the interpretation of this data set, a variety of potential terrestrial analogs were investigated using the same clast survey parameters that were employed during Spirit's traverse. Each terrestrial analog was selected to represent an end member geologic process that could have shaped local clast distribution and morphology. These data sets were analyzed using SAS/STAT statistical software, employing Principle Component Analysis (PCA) to reduce the dimensionality of the data set, focus attention on the relationships between independent variables, and to identify factors that, taken together, could provide an objective basis for discriminating between geological processes. During the course of the traverse, significant changes were observed in clast size when moving from the continuous ejecta blanket of Bonneville crater (high thermal inertia) onto the intercrater plains (low thermal inertia). However, this trend was not apparent when crossing the continuous ejecta of two smaller craters, Lahontan and Missoula. In fact, clast sizes for these two craters compared more closely to the smooth intercrater plains unit previously mapped from orbit. Over the traverse, significant variations were observed in the distribution of vesicular clasts and in clast density. Changes in vesicularity are interpreted as relfecting local changes in the distribution and impact excavation depths of buried lava flow surfaces. Observed trends in clast size correlated well with thermal inertia values, as estimated from orbital (THEMIS) data. Over the course of the traverse, clast roundness and sorting remained remarkably consistent, with mean estimates falling between sub-angular to subrounded, and poorly sorted. These observations do not support previous suggestions of water-based depositional systems (fluvial, debris flow, or glacial processes) at the Spirit landing site, based on orbital data. Instead, observed trends are consistent with a heavily cratered, wind modified ejecta surface, developed above a flow-dominated basaltic volcanic sequence.
P32A-04 11:15h
From Ripples to Dunes: Transitions in Bedform Profile Across Aeolian Deposits, With Implications for Mars
The physics of particle movement is distinctly different for aeolian ripples and dunes. Ripples are the result of both saltation and reptation of individual sand grains, and their interaction with the curtain of blowing sand. Dunes, on the other hand, are influenced by the drag on the wind imposed by the presence of saltating sand along with the modified wind flow over or around the sand pile. Intermediate between ripples and dunes are mega-ripples comprised of a bimodal mixture of sand with a surface layer of either granules or pebbles, where the large particles are rolled across the surface through impact creep. We have obtained detailed topographic profiles across both ripples and dunes in an effort to quantify the shape of the landforms produced by these distinct sediment transport conditions. Our profiles cover more than four orders of magnitude of length scale (taken here to be the bedform wavelength ), collected from several aeolian sites throughout the southwestern United States. There is a regular progression in the bedform profile from the asymmetric sand ripples through both symmetric and asymmetric granule and pebble ripples to more symmetric profiles across transverse dunes. There is also a progressive increase in the width of the surface area exposed between the bedforms themselves, from virtually no inter-ripple regions for sand ripples to inter-ripple flats comparable to the bedform length for granule and pebble ripples to inter-dune flats that are much larger the bedform itself in transverse dunes. When scaled to wavelength, these bedform patterns represent a suite against which remotely derived bedform shapes can be compared. We are optimistic that such length-scaled bedform profiles will prove useful as a method for attempting to resolve the current ambiguity that exits regarding whether the ubiquitous transverse aeolian ridges visible in high resolution Mars Orbiter Camera images are the result of ripple or dune formation mechanisms.
P32A-05 11:30h
Morphometric Analysis of Martian Valley Network Basins
Whether the Martian valley networks were formed predominantly by groundwater sapping or surface fluvial runoff continues to be debated and has profoundly different implications for the climatic history of Mars and the possible evolution of life there. This paper attempts to evaluate the relative role of groundwater sapping vs. surface runoff in valley networks generation through quantitative morphometic analysis at watershed basin scale using the highest resolution MOLA DEM data. Treating the hypsometric curve of each basin as a cumulative probability distribution, its integral, skewness, kurtosis, density skewness, and density kurtosis (collectively called hypsometric attributes) can serve as quantitative measures of the basin's morphology. The hypsometric attributes are sensitive to variations in overall basin characteristics and thus are diagnostic of possible processes. These attributes of typical terrestrial sapping basins, typical terrestrial fluvial basins, and typical lunar impact cratering basins were used as endmembers to establish discriminant functions to classify Martian basins. The posteriori probability of each class membership for each Martian basin then serves as the measure of the relative role of sapping, fluvial or cratering process in forming that basin. Initial results in Margaritifer Sinus area show that groundwater sapping played a more dominant role than surface runoff, generally consistent with previous findings based on lower resolution data and supporting a precipitation-recharged groundwater sapping origin for valley networks. Using a circularity function (a ratio of area and perimeter as a function of elevation) as the measure for basin morphology in the discriminant analysis produced similar results. Other traditional morphometric parameters such as drainage density, width function, cumulative area distribution, and slope-area relationship will also be derived for Margaritifer and other areas. The overall spatial pattern of the distribution will be examined to shed more light on past climatic conditions for generating the Martian valley networks.
P32A-06 11:45h
An Alternative Model to Explain the Origin of Recent Volcanism on Mars An Alternative Model to Explain the Origin of the Recent Volcanism on Mars
The detection of recent volcanism (2 - 100 Ma) in the areas of Tharsis and Elysium raises questions about its origin. In the past, it was widely believed that the interior of a small planet like Mars, which is only about half the size of the Earth, cooled very fast associated with a rapid decrease and disappearance of the volcanic activity. This picture has been modified with more recent parameterized evolution models based on temperature-dependent viscosity laws, which show that the interior of a small planet cools more slowly as previously thought due to the development of a thick stagnant lid on top of the convecting mantle. However, even these models predict that volcanism has ceased at least one billion years ago unless long-lived, stable plumes exists in the Martian mantle whose existence is strongly debated. An alternative explanation for the recent volcanic activity might be the inefficient heat transport through the basaltic crust. Previous thermal evolution models have assumed the same thermal conductivity in the mantle and the crust. Due to its composition and structure, the basaltic crust, however, has a thermal conductivity which is at least a factor of about 2 lower than that of the mantle. In the present study, we examine the influence of such a low crustal thermal conductivity on the thermal evolution. A value of 2 W/mK typical for basaltic material and 4 W/mK for the mantle has been used. As a consequence of the thermal blanketing due to the low conductivity, a partial melt zone is located below the stagnant lid during most if not the entire evolution. Underneath regions with increased crustal thicknesses, e.g. underneath Tharsis, this melt zone has locally even an increased melt content and might be the source region for the observed recent volcanism.