A Hydrological Analysis of Terrestrial and Martian Gullies: Implications for Liquid Water on Mars

45th Lunar and Planetary Science Conference (2014) 1119.pdf

A HYDROLOGICAL ANALYSIS OF TERRESTRIAL AND MARTIAN GULLIES: IMPLICATIONS FOR LIQUID WATER ON MARS. S. W. Hobbs1, D. J. Paull1 and J. D. A. Clarke2, 1A School of Physical, Environ- mental and Mathematical Sciences, University of New South Wales Canberra, Australian Defence Force Academy, Northcott Drive, Canberra, Australian Capital Territory 2600, Australia; [email protected], 2MarsSociety Australia, P.O. Box 327, Clifton Hill, VIC 3068, Australia.

Introduction: Many fluvial and non-fluvial mod- pools and dry waterfalls. Thus we used the modified els have been proposed for the formation of Martian formula for f from [14] to account for this increased gullies (e.g. [1], [2]). However most recent research roughness, and obtained particle sizes based on direct favours top down melting [3] given the observations of field measurements. gullies located on isolated peaks or elevations not con- sistent with an aquifer layer [4], and failure of ground penetrating radar studies to identify any aquifer reser- voirs at gullies previously thought to have formed by aquifers [5] (Nunes et al., 2010). Understanding the role and amounts of liquid water involved in Martian gully formation is critical in studies of the Martian hydrosphere and climate. The amount of liquid water required to form gullies on Mars remains controversial and has been subject to a number of studies using fluvial modelling (e.g. [6], [7]). Estimations of fluid velocities and discharge rates will provide insight to the amount of liquid water required to form Martian gullies, and whether this correlates with similar activity observed terrestrially. In this work we performed hydrological analysis using Manning and Darcy- Weisbach equations to estimate gully velocities and discharge rates at Kaiser Crater and additional gullies in Noachis Terra, Mars (hereafter termed the primary site). We compared these with terrestrial results obtained at a two sites in Australia, a semi-arid site at Island Lagoon and a temperate site at Lake George. Method: Previous research have highlighted the high variation and terrestrial channel beds [8], and the uncertainty in determining channel bed composition for Mars, which directly effects the Manning roughness coefficient, n, and Darcy-Weisbach friction fac- tor, f, respectively [9]. To provide upper and lower limits of velocity estimations given the uncertainty in estimating roughness data for most of Mars we performed sensitivity analysis by varying the Manning roughness coefficient of 0.06 by 20%. We also calcu- lated f using a 20% variation around a median particle size of 0.18 m, These values are within the range of those suggested by Wilson et al. (2004) and Kleinhans (2005) and equate to particle sizes that would be on the threshold for detection by HiRISE imagery. We used the Mars adapted Manning’s equation from [10] and compared with results obtained using the Darcy- Figure 1. Representative cross profiles and channel Weisbach equation from [11] for inference of Martian parameter identification for hydrological analysis. gully velocity. Note an additional profile sampled near the gully al- Our previous research of the terrestrial gullies [12, cove has been omitted for clarity. 13] indicated rough channel beds consisting of plunge 45th Lunar and Planetary Science Conference (2014) 1119.pdf

through a series of smaller scale events, given the dif- Gully channel parameters for the studied Martian gul- ficulty in requiring large amounts of flowing liquid lies were obtained using cross section measurements of water to be stable under current martian conditions. In HiRISE DEMs for the martian gullies (Fig. 1) and addition post fluvial channel widening may have acted field survey measurements for the terrestrial gullies. on the Martian gullies, probably by dry mass wasting, leading to larger channels than were originally carved Parameter - Earth Lake George Island Lagoon by liquid water. Future surface missions to gullied sites Manning: may lead to better characterisations of gully channel Velocity (m/s) 1-3 1-2 regolith properties, and in turn greater understanding Q (m3/s) 1-7 1-4 of the role fluvial processes have in shaping Martian gullies. Darcy-Weisbach:

Velocity (m/s) 1-3 0.5-2 References: [1] Malin, M. and Edgett K. S. (2000) Q m3/s 2-5 1-3 Science, 288, 2330-2335. [2] Trinman A. H. (2003) Parameter - Mars Mars primary Mars Kaiser JGR, 108, doi:8010.1029/2002JE001900. [3] Dickson, Manning: Main/inset J. and Head J. W. (2009) Icarus, 204, 63-86. [4] Levy Velocity (m/s) 2-12/2-7 1-4 J. S. et al. (2011) Geol. Soc. London 354, 167-182. Q (m3/s) 17-911/5-25 4-42 [6] Nunes, D.C. et al. (2010) JGR, 115, Darcy-Weisbach: doi:10.1029/2009JE003509. [6] Parsons, R.A. and Velocity (m/s) 3-17/2-7 1-6 Nimmo F. (2010) JGR, 115, Q m3/s 21-1090/5-28 4-57 doi:10.1029/2009JE003517. [7] Jouannic G. et al. (2012) Planet. Space Sci., 71, 38-54. [8] Leopold L. B. Results: Velocities and discharge rates for our et al. (1964) Fluv. Proc. Geomorph. 522pp. studied gullies, as derived from Manning and Darcy- [9] McIntyre N. et al. (2002) LPS XXXIII, Abstract Weisbach equations are shown in Table 1. We ob- #1402. [10] Komar P. D. (1979) Icarus 37, 156–181. served similar velocity and discharge rates (1-3m/s; 1- [11] Kleinhans M. G. (2005) JGR, 110, 7 m3/s) occurred in our studied Lake George and Island doi:10.1029/2005JE002521. [12] Hobbs S. W. et al. Lagoon gullies. These were slightly lower than our (2013a) Planet. Space Sci., 81, inferred velocities of the Kaiser crater gullies and inset doi:10.1016//j.pss.2012.10.009. [13] Hobbs S. W. et channels of the primary site gullies (1-7m/s; 4-57 al. (2013a) Geomorphology, 204, 344-365. m3/s), though generally concurred with other re- [14] Wilson L. et al. (2004) JGR, 109, searched terrestrial events [15]. Velocities of the main doi:10.1029/2004JE002281. [15] Rickenmann D. channels of the primary site gullies tended to be sig- (1999 Nat. Hazards, 19, 47-77. nificantly higher than the terrestrial or other Martian gullies (2-17m/s; 17-1090 m3/s), equating to larger scale terrestrial events [15]. In addition to variation between gully channels we also observed variation of velocities and discharge rates within gully channels. This variation was present in all of our studied Martian and terrestrial gullies, though we also observed a general trend downwards in values from alcove to depositional fan. Discussion: The variations of gully velocities within each channel may be reflective of small scale changes in gully morphology, such as changes in cross sectional area, affecting the the velocity and discharge rate along the gullies. The correlation of the smaller Martian channels with terrestrial gullies sug- gest comparable scales of erosive events. Degradation of ice rich material at the Kaiser and primary Martian gullies, as inferred in our previous research [13, 14] could have supplied the water for gully erosion. The larger gully channels were more likely to have formed