45th Lunar and Planetary Science Conference (2014) 2917.pdf

EVIDENCE FOR MULTIPLE STAGES OF EXTENSIVE LOW OUTFLOW CHANNEL FLOOR RESURFACING IN SOUTHERN CIRCUM-CHRYSE, . J.A.P. Rodriguez1,2,V. Gulick1,3, V. Baker4, T. Platz2 and A.G. Fairén5, 1NASA Ames Research Center, Moffett Field, CA, 2Planetary Science Institute, Tucson, AZ ([email protected]), 3 SETI Institute, Mountain view, CA., 4Department of Hydrology, University of Arizona, Tucson, AZ, 5Department of Astronomy, Cornell University, NY.

Background. The history of outflow channel activ- located in eastern , which suggests a ity in southern circum-Chryse is of particular interest history of large-scale sediment and fluid discharge because flows from these channels may have contrib- from Valles Marineris contributing to the formation of uted to episodes of ocean or sea formation on the plan- the lower southern circum-Chryse [7]. et’s northern lowlands [1,2]. The relevant channel The inferred paths of discharge are consistent with the floors of these outflow channels mostly occur within observed distribution of the smooth mantles within two distinct dissectional levels [3] that formed mostly extensive zones that exhibit lengths mostly oriented during the Late [3]. The higher elevation towards the northern-plains (Fig. 1). channel floor extends from large chaotic terrains that flank extensive zones of subsidence, which led some investigators to hypothesize that catastrophic floods emanated from aquifers [4] or from vast cavernous systems [5-7]. Flood dissection into intra-cryospheric lenses of briny fluids might have led to the formation of systems of secondary chaotic terrains (zones of ouflow channel floor collapse) within these high-level channels [8]. The lower channels extend from Hydraotes chaos, a much broader and lower-lying chaotic terrain, which opens south into Valles Marineris [3]. Thus, in addition, to ground-water evac- uation invoked to form the upper level channels, dis- charges from vast paleo-lakes within central [9] and eastern [10] Valles Marineris might have also inundat- ed these lower outflow channel floors. The smooth surface topography and texture of these floor levels has been interpreted as indicative of a formative history that ended with the emplacement of extensive debris flow deposits [7,11]. Here, we investigate the thus far poorly document- ed history of secondary chaotic terrain formation with- in these lower channel floors. Morphologic and morphometric analyses and interpretative synthesis. Using MRO CTX image data in combination with Mars Odyssey Thermal Emission Imaging System (THEMIS, ~512 pixels per degree) day and night Infrared (IR) mosaics and Mars Fig. 1 Map of southern circum-Chryse, which includes low Orbiter Laser Altimeter (MOLA, ~460 m/pixel hori- channel chaos (brown) and smooth mantles (dark gray) (MOLA DEM 512 pixels per degree). zontal and ~1 m vertical resolution) surface topogra- phy, we have produced a zonal distribution map show- ing secondary chaotic terrains and smooth mantles Most of the investigated chaotic terrains are elevat- within the southern circum-Chryse lower outflow ed relative to nearby smooth mantles (Fig. 2a and and channels (Fig. 1). We find that although these terrains elevation profile a-a'). Individual knobs and mesas are occur pervasively throughout these channel floors, they mostly embayed by these mantles [12] and their distri- are particularly extensive within two major zones bution retains a fabric of NE trending lineations (black (Chryse Chaos and Aurorae Chaos in Fig. 1). arrows in Fig. 2a). Locally, fracturing of low-lying The observed mantling materials on these terrains channel floor outcops marked by bedforms may have extend ~2500 km from the northern reaches of contributed to regional chaos formation [12]. Simud/Tiu Valles to the Ganges and Eos Chasmata, 45th Lunar and Planetary Science Conference (2014) 2917.pdf

These observations point to an early stage of out- Valles. However, there is evidence for at least two flow channel activity in the region that produced chan- stages of extensive collapse of these floors. Although nel floor materials, which was associated with exten- the formation of the Chryse Chaos might have been sive collapse. This stage was then followed by em- associated with a regional geothermal anomaly, subse- placement of the widespread mantle, which likely quent global climate change during the consisted of debris flow deposits [7]. What are now [15-17], perhaps resulting from obliquity variations extensively mantled outflow channel surfaces comprise [18] might have triggered channel floor collapse. a major regional unconformity, which likely corre- Significance and implications. Investigating the sponded to a time when the history of outflow channel geomorphology of the lower elevation outflow activity was not dominated by catastrophic floods, but channels is particularly important for understanding by the devolatilation of sedimentary deposits. the nature of the discharges that extended from Valles Marineris into the northern plains. The early flat topog- raphy of these channel floors at approximately the same elevation (-4000 m) likely resulted in extensive sedimentation during outflow activity. Broad collapse, as described in this abstract, might be linked to a histo- ry of volume loss in volatile-rich sediments, supporting the hypothesis that the Valles Marineris discharges transported significant water volumes (e.g., floods, debris flows, glaciers). These might have contributed to the formation of oceans or seas in the planet’s northern lowlands. A key and novel implication of this study is that an enormous volume of volatiles would have been re- leased during outflow channel floor collapse. Chryse Chaos has as an approximate area of ~75,000 km2 and a collapse depth that averages to ~150 meters. Thus, ~10,000 km3 of volatiles, which is about half of the total volume of the Great Lakes of North America. The Fig. 2 Views of Aurorae Chaos (a) and Chryse Chaos (b). S stands for smooth plains and C for chaotic terrains (MOLA released volatiles could have triggered microclimatic DEM 512 pixels per degree). See text for details. variations in the region that would explain post- flooding periglacial resurfacing [19]. In addition, we have also found evidence indicating that the formation of the Chryse Chaos mostly post- References: [1] Clifford S.M. and Parker T.J. (2001) dated the regional mantle emplacement in northern Icarus 154, 40-79. [2] Fairen A.G. et al. (2003) Icarus, Simud Valles, which based impact-crater statistics 165, 53-67. [3] Rotto S. and Tanaka K. L., (1995) U.S. occurred during the Middle Amazonian or later ( Fig. Geol. Surv. Misc. Inv. Ser. Map, I-2441-A 1). Chryse Chaos lies at a lower elevation than smooth (1:5000,000).[4] Carr M.H. (1979) J. Geophys. Res. 84, channel floor materials to the north ( Fig. 2b) and it 2995-3007. [5] Rodriguez J.A.P. et al. (2005) J. Geophys. exhibits lobated fronts [12] on its surface consistent Res. 110. [6] Rodriguez J.A.P. et al. (2005) Icarus, 175, with a history of dewatering. Palaeo-surface retention 36-57. [7] Rodriguez J.A.P. et al. (2006) Geophys. Res. within the chaotic terrain appears to have also contrib- Lett. 33. [8] Rodriguez J.A.P. et al. (2011) Icarus, 213, uted to the generation of knobs and mesas [12]. We 150-194. [9] Lucchitta B.K. et al. (1994) J. Geophys. have also identified exhumed ridges, which might Res., 99, 3783-3798. [10] Warner et al., (2013) Geology, represent the remmants of dikes [12] possibly associat- 675-678. [11] Tanaka K.L. (1999) J. Geophys. Res. 104, ed with previously proposed regional volcanic activity 8637-8652. [12] Rodriguez J.A.P. et al. (in prep) [5, 13-14]. The intrusion of such dikes into the mantle Geophys. Res. Lets. [13] Rodriguez J.A.P. et al. (2003) materials would have post-dated the latest stage of Geophys. Res. Lets. 30, 1304. [14] Messere S. et al., regional outflow channel activity, and would link the (2008) Icarus, 194, 487-500. [15] Kadish S. J. and Head development of the Chryse Chaos to the dissipation of III J. W. (2013) Planetary and Space Science, accepted. geothermal heat. This hypothetical scenario can also [16] Skinner et al., (2012) Geology, 10.1130/G33513.1. explain why the “recent” chaotic terrain formation is [17] Hobley D.E.J and Howard A.D. (2013) J. Geophys. spatially restricted to a portion of northern Simud Res. Accepted. [18] Head J.W. et al. (2005) Nature, 434, 346-350. [19] Pacifici et al. (2009) Icarus, 202, 60-77.