P41A-1345
Paleodischarge of Inverted Fluvial Features in the Aeolis/Zephyria Plana, Mars
Approximately 150 sinuous ridge (SRs) segments of various morphologies -- including 'thin', 'flat', or 'multilevel' - are found in the Aeolis/Zephyria Plana region of Mars, where most of them are being exhumed from within the western Medusae Fossae Formation. Thin SRs are interpreted to be inverted fluvial channels. Flat SRs are interpreted to be inverted floodplains of meandering channels. Multilevel SRs, comprised of thin SRs superposed on flat SRs, are interpreted to be inverted channels superposed on inverted floodplains (Burr et al. in revision). Here, we present paleodischarge values of these features using measurements made on THEMIS visible wavelength images (res. 18 meters per pixel) from the Mars Odyssey spacecraft. For those SRs interpreted to be inverted channels (including both individual thin SRs and the thin SRs superposed on flat SRs in multilayer morphologies), we measured SR length, width, sinuosity, and meander wavelength. For those SRs interpreted to be inverted floodplains of meandering rivers (including both individual flat SRs and the flat SRs subjacent to thin SRs in multilayer morphologies), we measured SR meander belt width. These measurements were then used in various empirical equations, based on Earth data, for determining discharge, with scaling for Martian gravity. The average discharge of the 18 SRs for which likely valid measurements could be obtained is 27 cubic meters per second. These data, in conjunction with the inferred age of the Medusae Fossae Formation, indicate that there was geomorphically effective surface flow of water at the equator during the late Hesperian to early Amazonian epoch. These results may provide data for future research involving climate modeling and give a more detailed account of the history of water on Mars.
P41A-1346
Exploring the Martian Subsurface of Athabasca Valles Using MARSIS Radar Data: Testing the Volcanic and Fluvial Hypotheses for the Origin of the Rafted-Plate Terrain
High Resolution Stereo Camera (HRSC) images of the vicinity of Athabasca Valles area (5°N, 150°E) reveal a surface with a broken, rafted-plate morphology. Two formation hypotheses (volcanic or fluvial) have been proposed to explain this appearance. In order to test these volcanic and fluvial hypotheses for the origin of the rafted-plate terrain, we investigated the subsurface radar echo from the Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS) 5 MHz-band data over this area. The backscattered signal losses were compared to Finite Difference Time Domain (FDTD) simulations of those arising from two hypothetical geoelectrical subsurface models, which differed in their assumed ice content. Within this region, MARSIS experienced loss rate of 0.09dB/m in the first 160 m beneath the surface. FDTD simulations show that, if the near-surface environment (up to 45 m depth) is ice-rich (80% by volume), it will result in a loss rate of 0.048dB/m, whereas the losses associated with an ice-poor model (20% of ice by volume) increase to 0.10dB/m. Comparing the observed MARSIS losses with the simulated ones suggests that propagation characteristics of Athabasca's subsurface seems most consistent with a volcanic rather than a fluvial origin for the broken, rafted-plate terrain in the vicinity of Athabasca.
P41A-1347
Groundwater Discharge and Gully Formation on Martian Slopes
Young gullies and gully deposits on walls of martian craters have been cited as evidence that liquid water flowed on the surface of Mars relatively recently. Effects of variable environmental conditions at the surface of Mars are modeled and applied to the case of groundwater emergence from shallow aquifers to investigate whether groundwater is a viable source to enable the erosion of these gullies. The model includes detailed treatment of ice growth in the aquifer. Model results indicate that groundwater discharge can be maintained under the current environmental conditions if the aquifer permeability is like that of terrestrial gravel or higher, the aquifer is 350 K or warmer, or if the aquifer is a brine with a freezing point depressed to 250 K or below. Warm (350 K), pure water, moderate hydraulic conductivity aquifers and cold (275 K), pure water, high hydraulic conductivity (gravel) aquifers do not exhibit a strong dependence of discharge volume on season, latitude or slope orientation in our modeling. Low hydraulic conductivity, cold, CaCl2 brine aquifers and moderate hydraulic conductivity, cold, pure water aquifers both show azimuthal, latitudinal and seasonal dependencies in the calculated discharge rates. Discharges for low hydraulic conductivity, cold, CaCl2 brine aquifers favor equator-facing slopes at high southern latitudes and pole-facing slopes at low southern latitudes.
P41A-1348
Central Peak Gully Formation and Morphologies on Mars
This study quantifies the morphology, flow characteristics, and dimensions of gullies that formed on central peaks of impact craters on Mars in an effort to examine possible formation mechanisms. Using HiRise stereo image pairs, MOLA data, and CRISM data, we have focused on two impact craters in the Northern Lowlands: Lyot crater, and a younger central peak crater approximately 300km to the Northeast. Both have well- defined gully systems originating from near the top of their respective central peaks. The smaller crater is roughly 18km in diameter with a fresh central peak crater morphology, and is located at 52.4N and 39.8E. There are at least two well-defined gully systems apparent on this crater's central peak: one extending ~2100m to the north, and the other ~1300m long and trending to the West. Lyot is an Amazonian aged, multi-ring impact basin that is 215km in diameter, located at 50.4N and 29.3E. Lyot also has at least two well-defined gully systems on the central peak: one trending north with a length of ~4200m and another trending westwards for ~2300m. All of these gully systems show features consistent with fluid flow, including erosion of bedrock, channels with slopes much less than 30 degrees, meandering channel beds, point bars, and anastomosing channels on the debris aprons. The gullies on Lyot have a sinuosity varying from 1.064 to 1.092, with slopes decreasing from 29.7 to 12.7 degrees from the gully alcoves to the main channels and varying between 6.3 and 12.1 degrees for the debris aprons. Flow velocities based on the sinuosity of the channels, using terrestrial analogues, are calculated to be between 3.5 m/s and 5.3 m/s [1]. The discharge rates are also calculated using the channel dimensions with gravity scaled to Mars, resulting in rates between 1.1 m/s and 2.8 m/s [2]. A range of total flow volumes will also be estimated using debris apron and gully volumes. These gullies are particularly unique because they originate from the small isolated areas of the central peaks. Using our measurements and observations we will address multiple working hypothesis for possible formation mechanisms of these fluvial systems [3][4]. [1] Ikeda, S. et al. (1981) J. Fluid Mech, 112, 363-77 [2] Kleinhans, M.G. (2005) JGR 110, E12003 [3] Parsons, R.A. et al. This meeting [4] Gulick, V.C. (2001) Geomorphology 37, 241-268
P41A-1349
Modeling the Formation of Bright Slope Deposits Associated With Gullies in Hale Crater, Mars: Implications for Recent Liquid Water
Our study aims to uncover the formation mechanism of the recent bright slope deposits associated with gullies (BGDs) observed on Mars in order to assess the viability of liquid water involvement. We use a high resolution topography model generated from a stereo pair of images acquired by the High Resolution Imaging Science Experiment (HiRISE) aboard the Mars Reconnaissance Orbiter as input into a kinematic model and a flow model to assess whether or not a dry granular flow could form the recent BGDs seen in Hale Crater. There are no high-resolution images of the gullies with BGDs before the BGDs formed, but the BGDs have no resolvable modification and, therefore, are presumably recent. For the kinematic modeling of dry granular flows, we examine a range of particle sizes, flow thicknesses, and upslope initiation points to examine how these parameters affect the run-out distances. The results show that multiple combinations of realistic parameters could produce dry flows that travel to within the observed deposits" boundaries. The kinematic modeling provides a yield stress and an event- averaged viscosity for input into the flow model, FLO-2D (FLO Engineering, Inc., 2006. FLO-2D User Manual, Version 2006.1. Nutrioso, Arizona), which solves the dynamic wave momentum equation. We run FLO-2D to model both wet and dry flows and find that either could have formed the recent BGDs. We vary the inflow volume, inflow location, discharge rate, water loss rate, Manning roughness, and simulation time and examine the resulting maximum flow depths and velocities. Given the difficulty of producing liquid water on the martian surface under present-day pressure and temperature conditions, our results suggest that the recent BGDs are not convincing evidence of recent liquid water on the surface of Mars.
P41A-1350
Fluvial Discharge Rates of Martian Gullies: Slope Measurements From Stereo HiRISE Images and Numerical Modeling of Sediment Transport
Using a stereo pair of HiRISE images of a crater slope incised by fourteen gullies at -37.86 N, 217.92 E we calculate relative elevation changes between pairs of hand-selected points. Using the method of Kreslavsky [1]. The background slope on which the gullies are located has a slope of 22 degrees. Out of the five gullies we analyzed, all show a steadily decreasing slope from an average of 30 ± 4 degrees at the alcove to 16 ± 2 degrees at the apron. These measurements are in agreement with previous gully slope measurements done at MOLA resolution in a different region [2]. The slope beyond the base of the gully aprons is 4±1 degrees. The depth of alcove incision in nine of the gullies is 17±8.5~m. We take advantage of this slope and incision data to determine the evolution of a one-dimensional gully profile over time with a 1D sediment transport model [3]. The shear stress applied to the channel bed by flowing water is τ = ρ g h sinθ where h is the channel depth, g is gravity, and h is the channel depth. The rate of transport is non-linearly related to τ/τrg where the reference stress for a gravel bed is τrg = 0.035 ( (s-1)ρ g Dg ) where s is the ratio of sediment to water density, ρ is 1000~m3, and Dg is the sediment grainsize. The two significant unknowns in applying the theory to Martian gullies are the sediment grainsize and channel depth. We ran simulations for various channel depths and grainsizes to get a range of water discharges and simulation times that result in alcoves 25~m deep. Erosion is rapid due to the high slopes; incision rates decrease with decreasing channel depth and increasing grainsize. For grains 20~cm in diameter and a conservatively low channel depth of 20~cm, alcove incision occurs over a 5~h period, discharging a volume of 8500 m3 of water. These discharges assume a 1~m wide channel and a constant, bank-full discharge over the duration of the simulation. Gullies are spaced about every 500~m along the slope. If liquid water is sourced from melting ice within a semi-circular disk with a radius (r) equal to half the gully spacing, the discharge volume V=0.5π r2 h φ, where h is the disk thickness, and φ is the ice volume fraction. If φ=0.1 and r=250~m, then the thickness of ice-rich material is 1~m (per meter width of channel). [1] Kreslavsky, M.A., Mars Gully Conf. #8034, 2007 [2] Dickson, J.L. et al., Icarus 188, 315323, 2007 [3] Wilcock, P.R. and Kenworthy, T., Water Resources Res., 38, 1194, 2002
P41A-1351
Origin of Theater-Headed Tributaries to Escalante and Glen Canyons, Utah: Analogs to Martian Valley Networks
Some tributaries to Glen and Escalante Canyons in southern Utah share similar characteristics to typical Martian fluvial valleys, motivating their frequent use as process analogs. In the spring of 2008, we investigated six tributary canyons formed in Navajo sandstone (two branches of Bowns, Explorer, Fence, and two branches of a tributary between the latter two) to test the hypothesis that seepage weathering and erosion are the dominant geomorphic processes. Measurements included spring discharge, pH, and hardness; compressive strength by Schmidt hammer of Navajo and underlying Kayenta beds; Selby bulk strength of Navajo sandstone; discharge estimates for flash floods; size of transported rocks; and vertical profiles of valley headwalls and alcoves. Plateau slickrock surfaces are commonly rounded on 10-100-m length scales and yield abundant runoff, as during rainfall observed on May 21-22. Incision into the Navajo surface by overland flow yields narrow, high-gradient valleys with V-shaped cross-sections; abrasion by sediment and weathering by standing water in closely spaced potholes facilitate downcutting. These small contributing valleys funnel waterfalls over the broad headscarps, forming small plunge pools. Headwalls are largely swept clear of debris relative to sidewalls. Canyon dimensions increase significantly below seeps, and wide alcoves are found only at these locations. We found no significant difference in rock strength at the top and bottom of the Navajo headwalls, suggesting that headscarp retreat requires basal weathering. Diverse weathering processes affect different sections of the headscarp relief. An intermittent waterfall may directly attack the base of an alcove, processes related to vegetation usually affect its lower slope (wetted by seepage from a discrete layer exposed in the deepest zone), and salt weathering often occurs on the roof. Scarps above an alcove are relatively unweathered and retreat primarily by sheet fracturing. The parabolic shape maximizes strength and is not a direct consequence of sapping. Infrequent flash floods of ~1–10 m3/s (woody debris and erosion indicated depth) exceed the magnitude of 1-2 L/s spring discharges by more than three orders of magnitude, and flooding is primarily responsible for sediment transport, particularly imbricated rocks up to tens of cm in size. The tributary canyons are growing headward along their contributing streams rather than up the structural dip, except where the contributing plateau surface is a dip slope (e.g., Fence and Explorer canyons). Few headwalls and contributing streams follow a large exposed tectonic joint; any structural control is primarily due to cumulative smaller fractures. These observations suggest a hybrid model for theater-headed valleys in massive rocks. Seepage weathering is an essential factor in forming steep headwalls and alcoves in Navajo sandstone, but headward retreat and erosion of debris depends on flash floods rather than seepage erosion. Plateau topography, contributing streams, and small joints rather than structural dip or large tectonic fractures control the valley planform.
P41A-1352
Martian Post-Impact Hydrothermal Systems: Effects of Permeability and Freezing on Surface Discharge and Water:Rock Ratios
A km-scale bolide delivers enough energy to heat subsurface water, and drive hydrothermal circulation (Abramov and Kring, 2005). This post-impact hydrothermal (PIH) circulation can lead to surface discharge of water, and chemical alteration – both are potentially detectable. We present the effects that permeability and freezing have on discharge and water:rock (W/R) ratios. We simulate the evolution of PIH systems using MAGHNUM (detailed in Travis et al., 2003). MAGHNUM solves the time-dependent transport of water and heat through a porous medium, incorporating phase transitions between ice (applicable to Mars), vapor and water. PIH evolution depends on heat sources and permeability (k); these, in turn, control discharge and chemical alteration which depends on both the peak temperatures and the W/R ratio (Schwenzer and Kring, 2008). Recently, CRISM detected phyllosilicate-rich material within ~45 km craters (Mustard et al., 2008) and the HiRISE camera imaged gullies, some emanating from central peaks, within many high latitude craters. We model a 45 km crater created by a 3.9 km dia., 7 km/s impactor. Simulations run for 100,000 yrs in a 2D axisymmetric domain with a heat flux of 32.5 mW m-2. Thus far we have tested systems with a range of surface k's (10-4 to 1 darcys) that decay exponentially with depth and are exposed to two surface temperatures (5°C and -53°C). In general W/R ratios increase with increased k. Focusing in on the upper 200 m at the center of the crater, fluid temperatures remain > 100°C for 9000 yrs and flow yields W/R ratios of 10 when exposed to a surface temperature of 5°C. Dropping the surface temperature below freezing to a Mars-like - 53°C maintains upper 200 m temperatures > 100°C for only 600 yrs and W/R ratios are reduced to 1. Higher permeabilities yield higher W/R ratios but reduced time exposure to high temperatures. When surface temperatures are below freezing total system discharge is roughly independent of k for modest permeabilities but the time until the surface freezes increases with lower k. Freezing reduces both W/R ratios and discharge because ice closes pores and restricts flow.
P41A-1353
HiRISE Observations of Glacial Morphologies in Argyre Planitia, Mars
The landscape of the Argyre Planitia and adjoining Charitum and Nereidum Montes in the Martian southern hemisphere has been heavily modified since its formation. Major landforms in Argyre have previously been interpreted as glacial in origin. We re-examine morphologies in this region using MOLA topographic datasets and new details revealed in HiRISE images and discuss the implications for glacial processes. Parallel linear grooves (that often crosscut large bedrock features), and streamlined hills (that often have wider uphill and thinner downhill ends), are oriented approximately downhill and are consistent with glacial erosion. Deep semi-circular embayments in mountains resemble terrestrial cirques. Broad U-shaped valleys have stepped longitudinal profiles and tributary valleys have hanging valley morphology similar to terrestrial glacial valleys. Large boulders blanketing a valley floor may be ground moraine deposited by ice ablation. Sinuous ridges in southern Argyre cross topographic divides, form map patterns similar to terrestrial eskers, often have layers and occur in troughs, and are thus consistent with glacio-fluvial processes. In particular, variations in sinuous ridge height appear to be related to the surrounding surface slope. This is a characteristic commonly observed in terrestrial eskers. Glacial interpretations are supported by the location of these collective landforms at >50°S, and their association with other landforms typical of glaciated landscapes such as lobate debris aprons interpreted as debris covered glaciers or rock glaciers. At least portions of Argyre appear to have been modified by ice accumulation, glacial flow, erosion, deposition, stagnation and ablation. The type and significant amount of bedrock erosion and the presence of possible eskers suggests that the ice was, at times, wet-based. The number of superimposed craters is consistent with geologically-recent glacial activity, but may also be due to subsequent modification from non-glacial processes.
P41A-1354
Dichotomy Boundary Glaciation Models: Implications for Timing and Glacial Processes
An integrated system with glacial features exists at 34E, 41N in the Deuteronilus-Protonilus Mensae region. This 30,000 km2 valley system is typical of dozens of fretted valleys in this region along the dichotomy boundary. We compare features described in current geological observations with results from the University of Maine Ice Sheet Model (UMISM) that we feel support the glacial interpretation of these features and also allow speculation as to the timing and processes responsible for the formation of these features. Geological observations identify evidence for a number of features that are felt to be indicative of glacial flow. These include: 1) localized alcoves from which emanate narrow, lobate concentric-ridged flows interpreted to be remnants of debris-covered glaciers; 2) alcove depressions perhaps indicating sublimation of material from relict ice-rich accumulation zones; 3) plateau-ridge remnants between alcoves typical of glacially eroded aretes; 4) horseshoe-shaped ridges upstream of topographic obstacles; 5) convergence and merging of LVF fabric in the down-valley direction; 6) deformation, distortion and folding of LVF in the vicinity of convergence; 7) LVF with pits and elongated troughs in distorted areas; 8) distinctive lobe-shaped termini with associated pitting where the LVF emerges into the northern lowlands. This evidence defines a coherent, unified flow regime extending from the upper valley reaches down to the northern lowlands. Additional support for the glacial hypothesis comes from a GCM for a dusty-atmosphere Mars with obliquity set to 35o and a water source in the Tharsis region. The GCM generates a pattern of ice accumulation in good agreement with these geological observations. This climate is what one might expect to follow a high- obliquity excursion of the sort that built ice sheets on the Tharsis volcanoes. UMISM as used here is an adaptation for the Martian environment of a thermo-mechanically coupled shallow- ice approximation terrestrial ice sheet model used for time-dependent reconstructions of Antarctic, Greenland, and paleo-icesheet evolution on Earth. Starting with no ice, the model is run for 2 million years. While this is longer that is expected for any steady climate to hold on Mars, it delivers a flow pattern that can be compared to the geological interpretations. We present ice thicknesses, surface elevations, and velocity maps at four times during the growth of this ice sheet. At 300Ka the flow from the sides has not yet merged in the centers of the valleys, a configuration that would not produce the turning flow observed. By 500Ka the beginning of a coherent downvalley flow is observed with ice from each side of the valley merging in the center. By 1000Ka there is a well-established valley glacier extending to the mouths of the valleys. Velocities are as high as 250 mm/year. By 1500Ka, the glacier extends out of the valleys onto the northern lowlands. Either 1000Ka or 1500Ka would produce the observed landforms. We also present comparisons of flow features in high-resolution THEMIS images with modeled flow at 1000Ka.
P41A-1355
Flow-front Morphology at Martian Ocean "Shoreline" Locations.
Many Martian outflow channels and valley networks terminate at "contacts" between the southern highlands and northern plains that were identified based on Viking Orbiter images over 20 years ago and interpreted by Parker and others as regressive shorelines of an ancient Martian ocean. Accumulating high-resolution image data and gridded MOLA topography is enabling a reassessment of the elevations and morphologies of proposed shoreline contacts. Many of the better-preserved, lowest-elevation of these contacts exhibit lobate flow-front morphology suggestive of low-viscosity lava or debris flows advancing up the highland margin from the northern plains. As progressively more degraded, higher-elevation contacts are examined, the lobate flow-front morphology is less apparent, and can be found in association with terrace-like morphologies that resemble terrestrial strandlines. In the most-degraded contacts at higher elevations, flow morphology is largely or completely unrecognizable, giving way to topographic terraces on the highland margin that are evident only in the MOLA topography, and that resemble uplifted and abandoned terrestrial coastlines where most of the coastal morphology has been destroyed. The best-defined of the proposed shoreline contacts, the "Deuteronilus Level" or Contact 2 in Parker et al. 1989, exhibits lobate flow fronts at the margins that suggest a flow direction from the plains interior and up the flanks of the highlands to about an elevation of -3900m locally. This contact is the most extensively traced of the proposed shoreline features. For the most part, it does appear to define a formerly level surface, though it ranges from about -3900m elevation in west Deuteronilus Mensae, to -4200m in west Cydonia Mensae. Changes in elevation along the contact are typically abrupt, and associated with recognizable structural features such a wrinkle ridges, so vertical offsets of once-level features is a plausible explanation for these variations in elevation. In Parker et al 1993, high albedo lobate flow fronts were described that appear to have advanced up-channel in the northern Chryse region, and a similar margin was identified in western Utopia. These were interpreted as backfill deposits associated with ponding within the northern plains associated with the circum-Chryse floods. However, post-Viking high-resolution images from MGS, Odyssey, and MRO have revealed this material to be similar in texture to that in west Deuteronilus Mensae, and MOLA topography reveals the margin in this area to have as much as 50m of relief associated with the flows. Plainward of this margin, the topography drops off, suggesting lowering of the surface of the plains interior by up to several hundred meters (even kilometers, in the basin center) via removal of material. It is suggested, then, that these lobate deposits are debris-rich flow fronts at the margin of a late transgressive phase of a debris and ice-covered ocean that has since receded.
P41A-1356
Fluvial Channel System in the Eastern Rim Region of Hellas Basin, Mars
This study focuses on one of the several fluvial systems located on the eastern rim region of the Hellas basin, Mars. We have analyzed the morphologic and morphometric characteristics of this extensive, over 650km long channel system. Beginning from Hesperia Planum, it intersects several basins and exhibits a multitude of flow morphologies. Eventually, it merges with Reull Vallis at 39.5°S, 98.1°E. The eroded volume of surface materials by this particular channel system is roughly 73 km3. The existence of numerous other, similar channels along the Hellas margin suggests that surface runoff may have played a significant role in the degradation of the basin rim.