Testing the water hypothesis: Quantitative morphological analysis of terrestrial and martian mid-latitude gullies

    Testing the water hypothesis: Quantitative morphological analysis of terrestrial and martian mid-latitude gullies S.W. Hobbs, D.J. Paull, J.D.A. Clarke PII: DOI: Reference:

S0169-555X(17)30325-2 doi:10.1016/j.geomorph.2017.08.021 GEOMOR 6116

To appear in:

Geomorphology

Received date: Revised date: Accepted date:

24 April 2017 6 August 2017 7 August 2017

Please cite this article as: Hobbs, S.W., Paull, D.J., Clarke, J.D.A., Testing the water hypothesis: Quantitative morphological analysis of terrestrial and martian mid-latitude gullies, Geomorphology (2017), doi:10.1016/j.geomorph.2017.08.021

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Testing the water hypothesis: Quantitative morphological analysis of terrestrial and martian mid-latitude Gullies

School of Physical, Environmental and Mathematical Sciences, University of New South Wales

IP

1

T

S. W. Hobbs1*, D. J. Paull1 and J. D. A., Clarke2

SC R

Canberra, Australian Defence Force Academy, Northcott Drive, Canberra, Australian Capital Territory 2600, Australia

Mars Society Australia. P.O. Box 327, Clifton Hill, VIC 3068, Australia

NU

2

AC

CE P

TE

D

MA

*Corresponding author. +6126268 8455, [email protected]

ACCEPTED MANUSCRIPT

Abstract

T

Although Martian gullies resemble terrestrial counterparts, there two conflicting hypotheses for

IP

their formation still invoke fluvial processes on the one hand or lubricated CO2 flows on the other.

SC R

In this work we compared the quantitative morphology of terrestrial gullies, known to have formed by liquid water, and mid-latitude Martian gullies in the Martian southern hemisphere. We also

NU

compared these results with measurements of Martian dry ravines adjacent to the gullies. Our results show a similarity between Martian and terrestrial gully formation, supporting the hypothesis

MA

that liquid water was involved in their erosion. Our results show dry ravines differ morphologically from gullies, further suggesting fluidised flows as a likely origin of the latter. Variations in the

D

relationships across various terrestrial and Martian gullies indicate the significance of local

TE

environmental and geological conditions. Our work supports the idea that Martian gullies may not have been formed by just one single process but may evolve through a more complex interaction of

CE P

processes and environment.

AC

Keywords: Earth; Mars; gully; fluvial; surface; process; slope

ACCEPTED MANUSCRIPT 1. Introduction The discovery on Mars of small scale, youthful erosive features resembling terrestrial hillslope

T

gullies suggested the possibility that liquid water existed on the red planet in recent times (Malin

IP

and Edgett, 2000; Golombeck et al., 2006; Parkner, 2016). This discovery led to a growing body of

SC R

work in efforts to determine how, and by what process, gullies on Mars are formed and erode (e.g., Baker et al., 2015). Martian gullies superficially resemble terrestrial features in that they possess Vshaped channels, often with an associated alcove, and V-shaped delta deposits (Malin and Edgett,

NU

2000; Dickson and Head, 2009). Terrestrial hillside gullies possessing these features have been formed by concentrated surface runoff or snowmelt from topographic catchments (Selby, 1991);

MA

water seepage from a subsurface aquifer (Soms, 2006; Grasby et al., 2014); or debris flow processes

D

(Selby and Hodder, 2000; Hartmann et al., 2003; Reiss et al., 2009a,b).

TE

Early theories of Martian gully formation suggested origins through groundwater processes (Malin and Edgett, 2000). The inability to identify perched aquifers via subsurface radar analysis (Nunes et

CE P

al., 2010) and subsequent discovery of gullies in locations inconsistent with subsurface flow and recharge mechanisms, such as on isolated topography, indicated differing explanations (Costard et

AC

al., 2002; Schon and Head, 2011, 2012). Research, inspired by studies of the Antarctic Dry valleys and identification of likely water-ice deposits in some gullies, has suggested snowmelt (Mellon and Phillips, 2001; Christensen, 2003; Head et al., 2008; Dickson and Head, 2009; Levy, 2014) with pole-facing gullies being influenced by solar heating of accumulated water ice (Mellon and Phillips, 2001). Other research identified many gullies existing within latitude dependent mantle (LDM), where ice mixed with sediment is likely to be found (Kreslavsky and Head, 2002; Dickson and Head, 2009; Araki, 2012; Schon and Head, 2012; Conway and Balme; 2014). Melting of ice could provide the source of liquid water for gully channel erosion (Dickson and Head, 2009; Reiss et al., 2009a, b; Conway et al., 2011; Schon and Head, 2011, 2012).

ACCEPTED MANUSCRIPT Because of the difficulty of obtaining sufficient liquid water for gully erosion in the hyperarid Amazonian conditions of Mars, dry mass wasting processes have been suggested, either consisting of landslides of fine-grained material (Treinmann, 2003) or frosted granular flows acting in a

T

similar manner to snow avalanches (Hugenholtz, 2008). Nonfluvial proponents have also suggested

IP

CO2-based processes for gully erosion (Hoffman, 2000) based partly on observations of gully

SC R

alteration during temperature ranges incompatible with melting of water ice (Dundas et al., 2010, 2012, 2015; Raak et al., 2015; Nunez et al., 2016). A possible terrestrial analogue and subsequent

NU

morphology for this process was postulated to be pyroclastic flows (Hoffman, 2000). Numerical modelling by Pilorget and Forget (2015) has suggested that atmospheric CO2 is capable of freezing

MA

into gully alcoves during the Martian winter. Later CO2 sublimation in the Martian spring has been hypothesized to form gas-lubricated, fluidized debris flows capable of eroding gully channels over

D

time. This hypothesis is supported by the ability of frozen CO2 ice to condense, sublimate, and

TE

pressurize at latitudes and slope orientations where gullies are observed (Pilorget and Forget, 2016).

CE P

Given the difficulty in obtaining liquid water on Mars cited by CO2 proponents, flume experiments have been conducted in simulated Martian conditions to directly test whether Amazonian Martian

AC

fluvial erosion is viable (Conway et al., 2011b; Masse et al., 2016). Part of this research was inspired by the discovery of recurring slope lineae (RSL), thought to be evidence of real-time fluvial activity (e.g., McEwan et al., 2011). Laboratory testing of water behaviour under Martian conditions revealed that the metastability of water was able to cause slope destabilisation and, in some cases, produce a higher degree of erosion and runout when compared to similar amounts of water under terrestrial conditions (Conway et al., 2011b; Masse et al., 2016). Conway et al. (2011b) found that water produced a thin film of ice below the liquid-sediment contact that increased runout distances. Additionally, Masse et al. (2016) postulated that water seeping into erodible material would boil, loosening particle grains that would then flow downhill and cause erosion. Additional research has identified erosion activity and types that are not consistent with a CO2based hypothesis for gully erosion such as sinuous channels, small scale lobes, and transportation

ACCEPTED MANUSCRIPT of boulders downslope (Stock and Dietrich, 2006; Mangold et al., 2010; Dickson et al., 2015; Vincendon, 2015; Harrison, 2016; Johnsson et al., 2017). Recent work has also indicated complex drainage systems on some gullies (Corrigan et al., 2017; Gulick et al., 2017) and multiple

T

generations of gullies whose activity is consistent with melting of ice within pasted-on material

IP

(Harrison et al., 2017). This previous research has suggested that it is unlikely that CO2 processes

SC R

could produce such morphology in gullies.

NU

1.1. Previous morphometric research

Unlike terrestrial features, the study of Martian gullies is curtailed to remote sensing methods.

MA

No direct field observations are possible, and the period of observations have been limited to the time high-resolution cameras have been in Mars orbit compared with the extended timescales at

D

which most gully processes operate (Barlow, 2008). Limited observations may lead to a heightened

TE

risk of equifinality, the concept that differing processes may produce similar morphology, and may lead to incorrect assumptions on the origin and formation of features under study (Schumm, 1988).

CE P

Although a detailed understanding of the environment surrounding the feature under study, such as its climate and geology, would assist in inferring the history and morphology of gullies, this

AC

knowledge is limited, and researchers have had to rely on geomorphic indicators of fluidized or dry flow (e.g., Mangold et al., 2010). Table 1 summarizes morphological characteristics of fluidized flow vs. dry mass wasting found on small-scale features on Earth and inferred for Mars (Luccita, 1978; Patton, 1981; Heldmann et al., 2007; Conway et al., 2009; Levy et al. 2010; Mangold et al., 2010; Harrison et al., 2015). In the absence of direct observations of flowing liquid water, the presence and identification of these characteristics on Martian gullies would greatly assist in inferring the type of erosion (wet or dry) that has acted on the gullies. For example, gullies possessing concave up longitudinal profiles, sinuous channels, and triangular-shaped depositional fans may have been eroded by liquid water,

ACCEPTED MANUSCRIPT contrasting with linear talus-like dry flows such as those observed on the Moon (Kumar et al., 2013). Some researchers have suggested limitations in Martian gully morphometric analysis, given the

T

inability for such analysis to distinguish between fluidized or debris flow processes, or possible

IP

removal of diagnostic features in depositional aprons (Reiss et al., 2009a; de Haas et al., 2013).

SC R

Recent work, however, has focused on conducting quantitative analysis on gully morphometrics and comparing them with analogous features on Earth. This was conducted in order to identify trends in

NU

morphology and to infer whether Martian gullies erode in a similar way to terrestrial features. Yue et al. (2014) performed statistical analysis of gullies in different geological context such as within

MA

crater walls, terraces, sand dunes, and on Earth. These authors were able to identify similar morphological characteristics in gullies in different geological settings, finding that these

D

characteristics were quite similar between Mars and Earth and that gullies were probably formed by

TE

the same process (Yue et al., 2014). Comparative analysis of longitudinal profiles of 78 Martian

CE P

gullies with 24 fluvial and 22 debris flow terrestrial gullies by Conway et al. (2015) identified a clear morphological marker between terrestrial debris flow and fluvial erosive processes. Their work used the exponential curve-equilibrium state that has been found as a common feature of

AC

mature fluvial systems (Hack, 1957) compared with a reduction in profile concavity and steepening of the profile that are caused by debris flow systems (Brardinoni and Hassan, 2006; Stock and Dietrich, 2006; Mao et al., 2009). The longitudinal profiles of debris flow gullies studied by Conway et al. (2015) were found to be consistently steeper and less concave than fluvial gullies, while the Martian longitudinal profiles showed a slightly greater affinity for fluvial processes (Conway et al., 2015). Thus, gully processes were able to be inferred from geomorphology. Later work by Conway and Balme (2016) used statistical and hydrological analysis in order to identify similarities and differences between gullies on Earth, Mars, and the Moon. This analysis was able to discriminate between dry slopes on Earth and the Moon and those formed by debris flow or fluvial erosion.

ACCEPTED MANUSCRIPT In this current work we expand our previous gully research (e.g., Hobbs et al., 2013, 2016) and conduct geomorphological analysis of 68 mid-latitude southern hemisphere Martian gullies. We compare our findings with similar analysis conducted on 51 terrestrial gullies where liquid water is

T

known to be a dominant erosive process to determine how closely Martian gullies conform to

IP

terrestrial geomorphic features of fluvial origin.

SC R

Many craters hosting mid-latitude gullies also feature alcoves and mass wasted material though lack the incised, V-shaped channel observed in pole-facing gullies (Dickson and Head, 2009; Schon and

NU

Head, 2012; Harrison, 2016). These features have also been observed in latitudes closer to the equator where gullies are not observed (Treiman, 2003; Shinbrot et al., 2004). Previous research has

MA

used comparative analysis between these features and terrestrial dry debris flows to infer nonfluidized erosion in these areas (Hartmann et al., 2003; Schon and Head, 2012; Diniega et al.,

D

2013). Johnsson et al. (2014) compared talus cones in Puna Vacas, Argentina, with similar features

TE

on Mars and found that the equator-facing talus flows exhibited morphology that was consistent

CE P

with dry granular flow. Further analysis of the morphology of dry ravines, the terminology used in Hobbs et al. (2013) to describe such features, will allow additional characterization into the differences between them and pole-facing gullies in our study sites. Identifying morphology typical

AC

of dry ravines and comparing with those obtained from pole-facing gullies will assist in inferring the nature of erosive processes operating in these feature types. We thus analyse 24 dry ravine features on the equator-facing sides of two Martian gullied craters where dry mass wasting was assessed as the dominant process (Dickson and Head, 2009; Schon and Head, 2011; Hobbs et al., 2013). Overall, our study of gullies and dry ravines expands on previous research by including a wider range of Martian and terrestrial gullies as well as assessing the role that host escarpment morphology has played in affecting gully shape. This aims to resolve ambiguities in previous studies. We test whether relationships such as volume and length, width and depth, depth and slope, and normalized width and length are comparable between Martian and terrestrial gullies and are consistently different from those of Martian dry ravines.

ACCEPTED MANUSCRIPT 2. Material and methods 2.1. Definitions

T

Selected Martian gullies (Figs. 1A, 2A–J) and terrestrial gullies (Fig. 1B) were studied to provide

IP

comparisons between key relationships and to determine the similarity or differences between

SC R

Martian and terrestrial gullies. For the purpose of this work we defined a hillside gully as a feature formed on steep (≥10°) slopes such as an escarpment or crater wall, possessing a V-shaped channel

NU

and terminating in a depositional fan. Although some previous work noted that alcoves are often present on Martian gullies (e.g., Malin and Edgett, 2000; Dickson and Head, 2009), many gullies do

MA

not have alcoves; so alcoves are not part of our definition. Previous research has also suggested that Martian gully alcoves probably formed as a result of dry mass wasting processes where they are

D

found close to the equator where erosion by liquid water is unlikely (Dickson and Head, 2009). As

CE P

alcoves from our analysis.

TE

our work is focused on the role of liquid water in Martian gully erosion we have excluded gully

Figures 3A-D show the location of 24 dry ravines that we analysed in addition to our investigation of hillside gullies. We placed an additional constraint on the analysis by including only mid-latitude

AC

gullies in the analysis. Previous research of gullies in these regions suggested a greater influence of liquid water-based erosion because of obliquity changes and deposition and subsequent melting of ice-rich material on pole-facing slopes (Costard et al., 2002; Christenson, 2003; Dickson and Head, 2009; Dickson et al., 2015). Gullies have been identified at higher latitudes of Mars, but colder temperatures in these locations suggest that other processes may be at work in these areas (Dundas et al., 2012; Raak et al., 2015). We also further constrained the Martian gully analysis by investigating only mid-latitude gullies in the southern hemisphere. This region possesses some of the oldest and most heavily cratered terrain on Mars, with steep slopes on which gullies have formed. Restricting our investigation this way maximized the chances that gullies in this region

ACCEPTED MANUSCRIPT would have shared similar climatic conditions and would have been subjected to similar geology, which has been heavily influenced by cratering (Barlow, 2008).

T

2.2. Morphometric analysis

IP

All of our analyses and measurements of the terrestrial and Martian gullies were conducted using

SC R

ESRI ArcGIS software on digital elevation models (DEMS) for each site. Longitudinal profile analysis of the gully channels and adjacent host escarpments was conducted to assess how local geology, such as exposures of bedrock, may have influenced the slope of selected gullies.

NU

Longitudinal profiles were sampled using ArcGIS 3D Analyst by digitizing the gully channel thalweg. Digitization commenced from the highest elevation of the gully channel or, if the gully

MA

possessed an alcove, where the channel commenced below the alcove. Digitization terminated at the base of the gully channel. Portions of the channel that incised the gully depositional fan were also

TE

D

included in the longitudinal profile measurement. Gully longitudinal profile digitization facilitated measurements of gully concavity; the presence of a concave gully profile may indicate fluvial

CE P

activity as increasing fluid discharge levels downstream would equate to a decrease in slope gradient required to transport available sediment (Smith et al., 2000; Larue, 2008). Other factors

AC

such as inherited topography or exposed bedrock may also influence gully channel concavity (Phillips and Lutz, 2008). In order to be consistent with our previous research (Hobbs et al., 2013, 2014), the maximum concavity of the sampled gully profiles was estimated by subtracting the elevation of the slope at maximum difference between the profile curve and that of a straight line distance between the uppermost and lowermost extents of the channel in the manner conducted by Demoulin (1998). The potential influence of host topography to the gully profile shape was inferred by sampling longitudinal profiles approximately parallel to the gully channels down the adjacent host escarpment or crater wall slopes as near as possible to the gullies under study. This provided geomorphic context to the gullies’ locations, as well as samples of nongullied slopes whose profiles were compared with those of the gullies. The comparison was conducted by measuring the

ACCEPTED MANUSCRIPT difference between the gully and host escarpment, with larger differences suggesting greater erosion of the gully channel. Additionally, we measured the difference in height between the top and base of the gully channel

T

for our terrestrial and Martian study sites. We also measured the position along the length of the

IP

gully profile where the concavity was at its maximum. This was expressed as a ratio between the

SC R

horizontal position of the maximum concavity, and the length of the gully channel. Ratios of < 0.5 indicated that the point of maximum concavity occurred in the upper part of the channel; values >

NU

0.5 indicated that maximum concavity occurred in the lower reaches of the gully channel. We conducted these measurements to compare with findings by Conway et al. (2015) and to infer

MA

fluvial or debris flow processes. Debris flow morphologies tend to decrease the concavity of a channel profile, occur over limited elevation ranges, and possess regions of maximum concavity

D

farther downslope than fluvial systems (Mao et al., 2009; Conway et al., 2015).

TE

Profile samples also allowed for calculation of gully sinuosity, being the quotient of total channel length divided by the straight line distance between the channel top and bottom (Leopold

CE P

and Luna, 1964). Sinuosity has been used to infer erosion by liquid water, with values ≥1.05 considered to be sinuous and likely eroded by liquid water (Mangold et al., 2010). Sinuosity was

AC

measured for the studied terrestrial and Martian gully channels using the same start and end points as the measured gully channel long profiles. Along with longitudinal profile, gully cross-sectional profiles were sampled at the start, end and at equidistant contours along the length of each gully channel. This interval varied between each site to ensure at least six cross-sectional profiles per gully. Larger features, such as those in Palikir Crater were sampled with seven to nine cross profiles. Channel depth was calculated for each cross-sectional profile by measuring the distance between the lower channel top to the deepest point of the thalweg. Channel width was measured between the two highest points along each crosssectional profile. Cross-sectional areas were multiplied by gully length and summed to obtain an estimate of channel volume and to infer the erosive power of the medium involved in their

ACCEPTED MANUSCRIPT formation (Billi and Dramis, 2003; Nasri et al., 2008). The surface area of each cross-sectional profile was calculated using a trapezoid numerical integration method (Hart et al., 2010). The height between channel floor and crest and width of one DEM cell (0.9 m) was used as inputs to the

T

surface area calculation. Each surface area calculation was multiplied by the distance between the

IP

initial and next downstream cross-sectional profile to estimate gully volume between adjacent

overall gully channel volume with an error of ±5%.

Orthorectified

NU

2.3. Mars data sets

SC R

cross-sectional profiles. The resulting volumes were then combined to provide an estimate of the

MRO High Resolution Imaging Science Experiment (HiRISE) imagery and a digital

MA

elevation model (DEM) created from HiRISE stereo pairs were used to conduct detailed analysis of the morphology of the Martian sites using ESRI ArcGIS software (Table 2).

D

Slope measurements of the gully thalwegs were undertaken at the intersection between longitudinal

TE

and cross-sectional profiles of the gullies and were derived at equivalent elevations for the host

CE P

escarpments. Additional slope measurements were sampled in order to characterise the role, if any, liquid water has played in Martian gully erosion. Slope measurements of the gully thalwegs were undertaken at the intersection between longitudinal and cross-sectional profiles of the gullies and

AC

were derived at equivalent elevations for the host escarpments. Additional slope measurements were sampled in order to characterise the role, if any, liquid water has played in Martian gully erosion. Slope measurements have been used to infer processes in Martian gullies (e.g., Kolb et al., 2010; Conway et al., 2011a) though significant discussion has occurred as to what slope angles constitute the angle of repose for Martian regolith (26–32°, Pouliquen et al., 1999), dry mass wasting (>35°, Conway et al., 2011a), and the angle below which deposition can be considered subject to fluid activity (less than the angle of kinetic friction, <21°, Kolb et al., 2010). The use of the angle of kinetic friction has been particularly important as previous research has used this angle as the minimum at which dry flows will continue to move (Pouliquen et al., 1999; Taylor-Perron et al., 2003; Pelletier et al., 2008; Kolb et al., 2010; Fowler-Tutt, 2012). Previous research has

ACCEPTED MANUSCRIPT indicated that kinetic friction angles should be very similar between Mars and Earth (Kein and White, 1988). Analysis into debris flow slope angles occurring at ~20° has suggested the action of a volatile (water ice creep) in order to explain angles lower than that of kinetic friction (Taylor-Perron

T

et al., 2003). Additional analysis into bright deposits used the angle of kinetic friction to

IP

demonstrate that the deposits stopped flowing at the kinetic friction angle, suggesting dry mass

SC R

wasting processes (Pelletier et al., 2008). Comparing gully slopes with angles of kinetic friction and angles of repose has also been used to infer the presence of fluidized erosive processes (Kolb et al.,

NU

2010; Conway et al., 2011a, 2014, 2016). Thus, although uncertainties remain with the precise nature of the material gullies erode into, studies of slope angles remain a viable analysis tool for

MA

Martian gully researchers. Thus, we have used an angle of kinetic friction of 21° as a value to infer fluidized processes. Previous research has investigated the angle at which Martian gully

D

depositional fans have formed where the gully channel leaves the confined channel and is free to

TE

spread out and lose velocity (Bull, 1977; Kolb et al., 2010). As described above, the angle at which

CE P

dry mass wasting decelerates has been found to be ~21° (Pouliquen et al., 1999; Okura et al., 2000). The presence of depositional angles lower than ~21° has been used by Kolb et al. (2010) to infer the influence of fluidized flows. We investigated this relationship by measuring the thalweg slope angle

AC

~10 m upslope of the commencement of the depositional fan and defined this measurement as deposition. We also measured slopes approximately mid-way along the depositional fan of each gully, defined as fan. This was conducted in order to provide comparisons between Martian and terrestrial gully deposition angles. The vertical resolution of HiRISE DEMs is in the order of ~80–100 cm (McEwan et al., 2007), leading to errors in measurements obtained from them to be <1 m. The horizontal resolution of 1 m for the HiRISE DEMs and orthorectified images relies on the photogrammetric method used in their production (Broxton et al., 2011). In order to remove spurious high frequency noise in the HiRISE DTM, we applied a 3 x 3 low pass filter in a similar manner to Kolb et al. (2010), resulting in a model with ~1 m vertical resolution.

ACCEPTED MANUSCRIPT 2.3.1. Martian regolith particle size and pore pressures Solving the problem of identifying the precise nature of what Martian gullies actually erode into has

T

been constrained by lack of any surface missions to study Martian gullies (Barlow, 2008). No hard

IP

evidence is currently available in determining regolith particle size and cohesiveness as all research

SC R

in identifying such properties is restricted to inferences from imagery (e.g., Dickson and Head, 2009; Reiss et al., 2009b) or thermal inertia (Christenson 2006; Putzig and Mellon, 2007). Thermal inertia measures a material’s resistance to changes in temperature during the day/night heating cycle

NU

with larger-grained materials generally possessing higher thermal inertia values than finer-grained material (Jakosky, 1979; Mellon et al., 2000; Christensen, 2006). Martian thermal inertia data sets

MA

have been derived from Mars Global Surveyor Thermal Emission Spectrometer (MGS TES) based on night-time thermal inertia measurements obtained by Putzig and Mellon (2007) with a spatial

TE

D

resolution of 3 km/pixel and global coverage or the Thermal Emission Imaging System (THEMIS) with a resolution of 100 m/pixel, though with less coverage (Christenson, 2006). Thermal inertia

CE P

analysis of Martian gullies have been conducted in previous research, with analysis suggesting a strong correlation between fine-grained material (e.g., 60 μm-3 mm, Harrison et al., 2015) and

AC

gullies (Reiss et al., 2009b; Hobbs et al., 2013, 2015; Harrison et al., 2015). Although the precise cohesiveness of Martian regolith, pore pressure, or the proportion of mixed volatiles such as water ice will remain unknown until surface missions target Martian gullies, thermal inertia provides an inference of particle size. In this work we derived thermal inertia values from previously published results. As THEMIS thermal inertia data sets did not provide sufficient coverage of some of our study sites we used TES thermal inertia published by Putzig and Mellon (2007) where previously published results were unavailable. An overview of the Martian gully sites is provided below (Figs. 1A, 2A-J). 2.3.2 Kaiser Crater

ACCEPTED MANUSCRIPT The 12-km-diameter crater studied here is within the larger Kaiser Crater in Noachis Terra and possesses pole-facing gullies. Thermal inertia analysis indicates gully regolith particle sizes of finemedium sand size particles. Previous research at this site using apex slope and slope area analysis

T

revealed that the slope and morphology of the gullies was consistent with pore-pressure-triggered

IP

fluidised flow (Kolb et al., 2010; Lanza et al., 2010; Conway et al., 2011a). Additional analysis

SC R

suggested a more complex erosive mechanism involving a combination of fluvial and nonfluvial processes (Hobbs et al., 2014). We measured seven gullies along the pole-facing rim of this crater

NU

(Fig. 2A).

MA

2.3.3. Hale Crater

Hale Crater is in the uplands north of Argyre Basin. Previous investigation into the Hale Crater

D

gullies indicated that they are strongly influenced by local topographic conditions (Reiss et al.,

TE

2009b) and that they possess well developed drainage systems consistent with liquid water erosion (Gulick et al., 2017). Thermal inertia analysis for Hale Crater suggest the gullies are located within

(Fig. 2B).

AC

2.3.4. Primary site

CE P

fine-grained material (Reiss et al., 2009b). We investigated two gullies on the crater’s eastern rim

This unnamed, geologically young crater is at 41:21:24S 020:06:25E, in the Martian southern hemisphere (Figs. 2C, D). The 13-km-diameter crater is at an elevation of 2165 m above the Martian datum and has a floor depth of 780 m. Previous research indicated pole-facing gullies within this crater had eroded into ice-rich material but had not yet eroded into bedrock (Hobbs et al., 2013; Conway and Balme, 2014). Thermal inertia analysis indicates the gullies are embedded in fine-grained material while the dry ravines are located in coarser grain material (Hobbs et al., 2013). In the present study, we analysed 16 gullies on the pole-facing wall (Figs. 2C, D) and 13 ravines on the equator-facing wall (Figs. 3A, B) of this crater.

ACCEPTED MANUSCRIPT 2.3.5. Promethei Terra We studied one large and two smaller gullies within a 5-km-diameter crater within eastern

T

Promethei Terra (Fig. 2E). Previous research suggested multiple episodes of deposition occurring at

IP

this site (Schon et al., 2009; De Haas et al., 2013). The gullies were interpreted as being formed by

SC R

emplacement and subsequent melting of ice-rich mantling deposits linked to recent obliquity changes on Mars (Schon et al., 2013). Analysis of TES thermal inertia data indicates that these

NU

gullies are embedded within fine-grained material. 2.3.6. Newton and Palikir craters

MA

Palikir Crater, at 15 km in diameter, is within the larger Newton Crater in the Martian southern hemisphere and has an extensive gully system on its west-facing wall. Newton Crater also has

D

gullies within the southern-facing rim of a smaller 10-km-diameter crater (Fig. 2F). Head et al.

TE

(2008) suggested that gullies in this crater were formed by climate change melting debris-covered

CE P

glaciers that had originally formed from accumulation of ice-rich material on the pole-facing crater wall. The morphology of the gullies in Palikir Crater, such as concave-up profiles (Narlesky and Gulick, 2014) and complex drainage systems (Gulick et al., 2017) suggests erosion by fluvial

AC

processes. Newton and Palikir Crater gullies are located within fine-grained material. Eleven gullies in Palikir Crater and seven gullies in Newton Crater were measured for this analysis (Fig. 2G). 2.3.7. Gasa Crater The Gasa Crater site is at 35.72° S, 129.45° E in eastern Promethei Terra in the Noachis Terra highlands and consists of the younger 7-km-diameter Gasa Crater situated within a larger 18-kmdiameter crater. This site has been the subject of a number of studies (e.g., Kolb et al., 2010; Okubo et al., 2011; Dundas et al., 2012; Schon and Head, 2012). The outer crater has evidence for latitudedependent mantling and polygonal-patterned ground (Mustard et al., 2001; Head et al., 2003; Levy et al., 2009). Materials identified as channels and ponded debris flows were considered to be formed

ACCEPTED MANUSCRIPT by meltwater from fluidized ejecta created by the Gasa Crater impact striking an ice-rich glacial substrate (Schon and Head, 2012). Geomorphic investigations of the Gasa site gully slopes and association with ice-related features have suggested that their erosion is consistent with fluvial

T

activity (Okubo et al., 2011; Schon and Head, 2012). Thermal inertia values for the Gasa Site range

IP

from loosely consolidated, fine grained material near the gully depositional aprons to values

SC R

consistent with bedrock near the gully alcoves (Harrison et al., 2016). We investigated 11 gullies in the host crater and 8 gullies in Gasa Crater (Figs. 2H, I, J). We also studied 11 ravines on the

NU

equator-facing wall of Gasa Crater (Figs. 3C, D).

MA

2.4. Terrestrial gullies

We conducted similar morphology measurements on 51 terrestrial gullies at five different sites.

D

These sites were chosen because they possess hillslopes that fit our definition and are

TE

predominantly eroded by liquid water in order to provide a terrestrial analogue baseline to our Martian studies. Data sets used for analysis included real time kinematic (RTG) GPS-derived field

CE P

survey data, LiDAR data derived from a NSW government survey and LiDAR data downloaded from Open Topography (www.opentopography.org). A summary of data sets and sources are given

AC

in Table 2. We were unable to obtain terrestrial data sets for features analogous to the Martian dry ravines. Thus, we were able to conduct analysis only on Martian dry ravines. An overview of the terrestrial sites is given below. 2.4.1. Meteor Crater Meteor, or Barrington Crater, is a 1-km-diameter crater formed by a meteorite impact in the Arizona desert ~50,000 years ago (Kring, 2007). The crater contains numerous gullies, influenced by impact fracturing in the crater walls and eroded by water during a wetter period in the region (Kumar et al., 2010). Meteor Crater was chosen because previous research on these gullies suggested this site was an ideal terrestrial analogue for Martian crater gullies (Kumar et al., 2010;

ACCEPTED MANUSCRIPT Yue et al., 2014). We sampled 22 gullies on the inner wall of this site (Fig. 4A) using Optech GEMINI Airborne Laser terrain mapping with an accuracy of 0.05–0.3 m (Palucis and McEnulty,

T

2010).

IP

2.4.2. Alaska

SC R

This site is located along the Gulf of Alaska. These gullies are similar in morphology to those investigated by Hartmann et al. (2003) and were chosen to represent gully activity within periglacial

NU

environments. They are caused by seasonal freeze-thaw processes leading to snowmelt and runoff during the spring. Material in the gully head becomes saturated with water through this process and

MA

flows downslope, carving V-or U-shaped channels (Anderson et al., 1969). Dry mass wasting and snow avalanches have also been noted to modify gullies in this region (Black and Thorsteinsson

D

2008). We identified 10 gullies in this area for analysis (Figs. 4B, C, D) that were similar in

TE

geomorphology to those investigated by Hartmann et al. (2003).

CE P

2.4.3. Lake George

The 930-km2 Lake George basin is located in highlands 40 km northeast of Canberra, Australia.

AC

Lake George is bounded by a north-south running scarp that rises 250 m above the lake bed on its western side (Abell and Truswell, 1985). The scarp runs along the Lake George Fault, creating a half-graben morphology for Lake George and indicating a tectonic origin for the system (De Dekker, 1982; Macphail et al., 2015). Gullies are present along the western scarp, four of which we analysed in previous research and concluded that their morphology, while shaped by water erosion, were heavily influenced by the presence of bedrock (Hobbs et al., 2013). In this current work, we have analysed a total of 11 gullies along the western escarpment (Fig. 4E, F, G) using ground return LiDAR data with a resolution of 1.0 m. 2.4.4. White Mountain

ACCEPTED MANUSCRIPT The White Mountain Fault Zone is in eastern California and comprises the White Mountains-Owens Valley system (Kirby et al., 2006). This site has experienced ongoing seismic activity, with recent events occurring in 1986 (Smith and Priestly, 2000). The flanks of the White Mountains site have

T

also been subject to extensive fluvial erosion, creating alluvial deposits formed by runoff generated

IP

by intense rainstorms creating debris flows (Kirby et al., 2006). We were able to identify three

SC R

gullies fed by watershed runoff for inclusion in this study (Fig. 4H).

NU

2.4.5. Cooma

The Cooma Creek gullies are on farmland in the New South Wales Southern Highlands in part of

MA

the Monaro Volcanic Province (MVP), an ~4480 km2 Cenozoic intraplate geologic region in the eastern margin of Australia (Roach, 1999). Gullies proliferate in the region and were surveyed using

D

Real Time Kinematic (RTK) GPS technology (e.g., Martínez-Casasnovas et al., 2002; Renschler

TE

and Flanagan, 2008). Detailed field surveys showed a complex regime of erosion dependent on multiple conditions and processes such as local geology, surface runoff, dry mass wasting, and

3. Results

AC

(Fig. 4I).

CE P

animal activity (Hobbs et al., 2016). We used measurements of five gullies for the current analysis

A summary of key morphometric findings for the studied Martian and terrestrial gullies can be found in Tables 3-6. Figure 5 shows examples of longitudinal (Figs. 5A-C) and cross (Figs. 5D-F). Other morphometric results are discussed below. 3.1. Quantile analysis Quantile box plots of the Martian and terrestrial gullies are shown in Figs. 6 and 7. Sinuosity values of the Martian gullies (median 1.02, Fig. 6A) were lower than the terrestrial gullies (median 1.05, Fig. 6A). The range of terrestrial gully sinuosity values was higher, with third-quartile values of 1.07, compared with third-quartile values of 1.04 for Mars. Martian gullies possessed the smallest

ACCEPTED MANUSCRIPT width vs. length ratio (median 0.076, Fig. 6B), with the Martian ravines having the highest (median 0.25, Fig. 6B). This indicates that the Martian gullies typically were longer for their width when compared to the terrestrial gullies. Conversely, the ravines were much wider for their length,

T

indicating the broad debris chutes that accompanied these features. Width vs. length can be used as

IP

a diagnostic of gully vs. ravine where the dry ravine ratio is consistently several times larger than

SC R

for fluvial carved features. The Martian gully volume vs. length ratios we derived are comparable (Martian gully median 430, dry ravine median 360; Fig. 6C), while terrestrial gully volume/lengths

NU

are much lower (median 45, Fig. 6C). Height differences between the terrestrial and Martian gully channels showed the Martian gully height ranges to be greater (median 348 m, Fig. 6D) than

MA

terrestrial gullies (median 94 m, Fig. 6D). The median position of basal concavity between studied terrestrial (median 0.45, Fig. 6D) and Martian gullies (median 0.44, Fig. 6D) was found to be

D

similar. The range of terrestrial gully basal concavities were greater (0.21-0.81, Fig. 6D) than that of

TE

studied Martian gullies (0.16-0.64, Fig. 6D).

CE P

Analysis of terrestrial gully slopes show comparable value ranges to those on Mars (8 –49°, Earth, Fig. 7A c.f. 16–45°, Mars, Fig. 7B), though are lower overall (median 24.5°, Earth, Fig. 7A c.f.

AC

median 27.5°, Mars, Fig. 7B). Martian gully slopes tend toward high values (third-quartile 35, Mars, Fig. 7B) than for terrestrial gully slopes (third-quartile 30, Earth, Fig. 7A). In contrast the terrestrial host escarpment slopes have a greater range of values (7–57°, Fig. 7A) and are higher (median 28°, Fig. 7A) than for Mars (14–52°, median 26°, Fig. 7B). Slopes at which the transition from erosion to deposition occurred are lower for terrestrial gullies (median 12°, Fig. 7A) than for Mars (median 18°, Fig. 7B). Approximately 75% of all deposition slopes range between 16.25° and 19°, with the Martian median value being lower than the angle of kinetic friction for dry material (19.5°, Lanza et al., 2010). Approximately 75% of the Martian gully range (16-19°, Fig. 7B) was higher than the equivalent 75% range of terrestrial gully deposition slopes, which were lower at 10 – 16° (Fig. 7A). Similarly, terrestrial fan slopes (median 8°, Fig. 7A) are consistently lower than those of Mars (median 16.5°, Fig. 7B). Approximately 75% of Martian and terrestrial fan slopes fall

ACCEPTED MANUSCRIPT within a 4° range. Median ravine deposition slopes (median 32°, Fig. 7C) are consistently higher than for the studied Martian and terrestrial gullies. This is also the case for ravine deposition

T

(median 33°, Fig. 7C) and fan slopes (median 32.5°, Fig. 7C).

IP

Analysis of Martian and terrestrial gully channel curvatures indicated a trend toward being concave

SC R

(Figs. 7D and E). Gullies also consistently exhibit more curvature than the host slopes (Figs. 7D and E). Terrestrial gully (median 5.95%, Fig. 7D) and host slope (median 3.92%, Fig. 7D) curvatures are higher than Martian channel (median 4.88%, Fig. 7E) and host slope (median 2.95%, Fig. 7E)

NU

curvatures. Curvature differences in terrestrial gully channels and host slopes are lower (median 0.87%, Fig. 7D) than for Martian features (median 1.26%, Fig. 7E). In contrast, ravines show little

MA

variation in the ravine (75th percentile 1.16%, Fig. 7F) and host slope (75th percentile 0.53%, Fig. 7F) positive curvature values (measuring the ravine concavity), with the curvatures themselves

TE

D

being lower than the Martian and terrestrial gullies (ravine median 0.85%, host slope median 0.25%, Fig. 7F). Positive curvature differences (median 0.51%, Fig. 7F) are similar to terrestrial and

CE P

Martian gully values. Ravines (median -0.36%, Fig. 7G) and host slopes (median-3.68%, Fig. 7G) also show convexity, with the host escarpment convexities showing higher variations than the

AC

concave values (75th percentile 5.4%, Fig. 7G c.f. 75th percentile 0.53%, Fig. 7F). Figures 8A–H show fitted relationships across all terrestrial (Figs. 8A, C, E and G) and Martian (Figs. 8B, D, F, H) gullies. Gully volumes were found to be up to an order of magnitude smaller for terrestrial gullies (maximum 269,800 m3, White Mountain, Fig. 4A) compared to Martian gullies (12 km3, Palikir gully, Fig. 8B). Martian gullies provided a better fit overall than terrestrial gullies, with Martian and terrestrial sites exhibiting a power law between volume and length (volume = 0.02Length2.3, terrestrial gullies, Fig. 8A; volume = 1.43Length1.9, Martian gullies, Fig. 8B). Our findings agreed with the power relationships found in terrestrial gully studies (Capra et al., 2005; Nourmohammadi and Haghizadeh, 2014; Qin et al., 2016) and showed that the relationship is important for gully evolution as previously reported by Nachtergaele et al. (2014). The

ACCEPTED MANUSCRIPT volume/length relationship for dry ravines (volume = 0.47Length2.2, Fig. 9A) indicates that volume generally increases with the square of the length for these features. Gullies exhibit a power trend between depth and slope (terrestrial average depth = 0.01slope1.6, Fig.

IP

T

8C; Martian average depth = 0.1slope1.32 Fig. 8D). The terrestrial gullies exhibit a much stronger

SC R

relationship (R2 = 0.3705, Fig. 8C) than the Martian gullies (R2 = 0.1927, Fig. 8D). The positive trend indicates that channel depth increases slowly with an increase in slope, before a more rapid increase with further increases in gully slope. In the case of terrestrial gullies, the threshold value

NU

prior to higher falloff of depth values is ~20–25o (Fig. 8C), which was also the value for Martian gullies. A very poor inverse power trend was found for the Martian dry ravines (depth = 38slope, R2 = 0.1258, Fig. 9B), contrasting with the positive relationships found for the terrestrial and

MA

0.05

D

Martian gullies.

TE

Gully depth vs. width revealed a linear trend for terrestrial (width = 3.29Depth + 6.33, Fig. 8E) and Martian sites (width = 6.62Depth + 23.83, Fig. 8F). Terrestrial gullies provided a better linear fit (R2

CE P

= 0.782, Fig. 8E) than the Martian sites (R2 = 0.5908, Fig. 8F), though both trends indicate that gully erosion increases in depth along with width and that wider gullies should be deeper as well.

AC

Results from the dry ravine width vs. depth (average width = 8.79Depth + 18.65) are shown in Fig. 9C with an R2 value of 0.76. The width/length vs. length relationships of terrestrial and Martian gullies are shown in Figs. 8G and H respectively. Terrestrial and Martian gullies show an inverse power relationship (terrestrial average width/length = 25Length-1.2, Fig. 8G; terrestrial average width/length = 145Length-1.3, Fig. 8H), suggesting that erosion falls off in power along the length of the gully. This finding is consistent with gullies eroding primarily from their head, as would occur from fluid flows originating from the top of the gully and then flowing downslope. The larger multiplicative number for the Martian gullies (145 c.f. 25) reflects the greater length of the Martian gullies compared to their terrestrial counterparts. Width/length vs. length analysis of the dry ravines (average

ACCEPTED MANUSCRIPT width/length = 68Length-0.97, Fig. 9D) indicates an inverse power law that is more linear than for the terrestrial and Martian gullies.

T

Results of fitted relationships for individual Martian and terrestrial gully sites are shown in Tables 4

IP

and 5. Although similar relationship types exist across the gully sites (linear relationship for

SC R

width/depth, power relationships for depth/slope, and width/length vs. width), prominent variations in the formulae and R2 values exist between sites. Martian gully site fitted relationships are varied, ranging from high (R2 = 0.8944, Lyot Crater gullies, Table 4) to very low (R2 = 0.0144, Newton

NU

Crater gullies, Table 4). The Lyot Crater width vs. depth fitted relationship is the highest (R2 = 0.8944, Table 4), while R2 values of the Gasa host crater and Promethie Terra 377 gully width vs.

MA

length are also high. We found that a high R2 in one fitted relationship did not necessarily equate to high R2 in others (e.g., Promethie Terra width vs. depth, R2 = 0.882, depth vs. slope R2 = 0.08,

TE

D

width/length vs. length R2 = 0.63), suggesting localised variations in gully morphology. In the terrestrial gully case, the Meteor Crater site showed the highest R2 values of all sites except

CE P

for depth vs. slope, where White Mountain exhibited the highest R2 (R2 = 0.6452). Width vs. depth fitted relationships varies considerably between terrestrial gully sites. The Lake George gullies

AC

show a very poor fit for this relationship (R2 = 0.0907, Table 5), compared with Meteor Crater (R2 = 0.8818, Table 5). Two of the Lake George gully channels were very wide and shallow, creating extreme outliers in the fitting relationship. As with the Martian gullies, fitted relationships are generally not consistent in R2 values. Table 6 shows results of slope analysis for Martian and terrestrial gullies. We found that median gully slopes are mostly closely related to host escarpment slopes across Martian and terrestrial sites. The Cooma site is the exception, where gully slopes are 6° higher than those of the host escarpment (Table 6). In the case of depositional slopes, we found these to be consistently below the angle of kinetic friction for all Martian gully sites (median 9–19°, Table 5), though often higher than terrestrial ranges (8–27.5°, Table 6). Fan slopes for Martian gully sites (7–17.5°) also tend to be

ACCEPTED MANUSCRIPT greater than those for Earth (6–23°, Table 6). We found the Palikir site to possess the lowest deposition and fan slopes (median 9° and 7°, Table 6), and the Promethei site possesses the steepest (median 19° and 18°, Table 6). In the case of the terrestrial gullies, we found that the Alaska site

IP

T

has depositional and fan angles much steeper than the other sites (median 27.5° and 23°, Table 6).

SC R

4. Discussion and interpretations 4.1. Morphology

NU

The height drop ranges of the studied terrestrial (median 93.5 m, Fig. 6D) and Martian gullies (median 348 m, Fig. 6D) were less than that researched in Conway et al. (2015) though the greater

MA

range of Martian height drops was found to agree with the previous research. Martian gullies in general possessed greater volumes than terrestrial features, suggesting the greater height drop is

D

because of their larger size. This is because of the greater availability of larger escarpments and

TE

crater walls to erode into than Earth (Evans, 2003; Barlow, 2008). Position of basal concavities of

CE P

our study were similar to those found in previous research, suggesting that Martian gully erosion is similar to terrestrial debris flow and fluvial processes (Conway et al., 2015). The width vs. depth ratio of the terrestrial gullies we studied (width range = 0.05 – 4.2 depth, Table 5, average depth

AC

~3.3, Fig. 8E) was found to be linear and to approximate the 1.75–3 ratio found in previous research of gullies in soil (USDA-SCS, 1966; Soufi and Isae, 2012). In contrast the Martian gully width vs. depth ratio, while also linear, is higher in width range = ~3.2–7.5 depth, Table 4, average width = ~6.2 depth (Fig. 8F). This indicates that Martian gullies tend to erode horizontally more than vertically when compared to terrestrial gullies. Although greater than the terrestrial gully ratio, it was less than that of dry ravines, suggesting lower quantities of water available for channel incision compared to terrestrial features. We considered the possibility of channel widening by sidewall retreat in the Martian gullies as an explanation for the increased gully width (Martinez-Casasnovas et al., 2004; Wu et al., 2008; Stot, 2011). This process occurs in many terrestrial gullies where channel width increases from oversteepening and collapse of gully channel sidewalls. Many of the

ACCEPTED MANUSCRIPT studied gullies affected by sidewall retreat have eroded into easily erodible material and are subjected to high amounts of rainfall (Martinez-Casasnovas et al., 2004; Wu et al., 2008). Gully slope angles and tension crack development was found by Martinez-Casasnovas et al. (2004) to be

T

the main factors affecting channel wall collapse. Previous research has indicated that lower

IP

quantities of water available to erode Martian gullies compared to terrestrial features (Dickson and

SC R

Head, 2009; Conway et al., 2011b; Schon and Head, 2011, 2012; Johnsson et al., 2014) and rainfall in the Amazonian period when Martian gully erosion was thought to have commenced is not

NU

thought likely (Barlow, 2008). Thus, sidewall retreat on Mars is more likely to have been influenced by high channel wall slopes. Analysis of slopes in previous research has suggested regions within

MA

gullies subject to mass wasting due to slopes at or higher than the angle of repose, such as gully alcoves (Dickson and Head, 2009; Conway et al., 2011b; Glines and Gulick 2014). Thus, the

D

presence of high slope angles in these regions would indicate areas of likely channel sidewall

TE

collapse. Previous research into Martian crater wall and terrace area gullies indicates that the

CE P

majority of studied gully channels possess slopes below the angle of repose (Xue et al., 2014). Our earlier research into channel angles of the primary site and Kaiser Crater gullies also revealed that most gully channel slope angles were low (Hobbs et al., 2014). Therefore Martian gully erosion by

AC

sidewall retreatwould be less prevalent in Martian gullies than the studied dry ravines where higher slope angles are more common. Results from the dry ravine width vs. depth (average width = ~8.8Depth) indicated that the ravines are consistently wider than their depth than Martian and terrestrial gullies. This suggested a tendency for ravines to erode more horizontally than vertically than either Martian or terrestrial gullies. As discussed above, sidewall retreat would be a more dominant process for dry ravine erosion, as these features possess higher slope angles than for the Martian and terrestrial gullies (Figs. 7A–C). Width/length vs. length analysis of terrestrial and Martian gullies suggested that erosion shows a falloff in power along the length of the gully. Xue at al. (2014) suggested that this relationship

ACCEPTED MANUSCRIPT indicated that fluid carving the gullies was mainly sourced from the gully head. Terrestrial gullies can erode through headwall retreat, where undercutting causes the gully head to retreat upslope (Howard, 1995; Lamb et al., 2006). In this case of terrestrial gully erosion the primary erosive

T

source and main source of fluid would be from the gully head, before flowing downslope. Such a

IP

process has not been observed on Mars, given the extended timeframes required for eroding gullies

SC R

at the Martian scale (Dickson and Head, 2009). Previous research has suggested accumulation of ice-rich deposits in alcoves or catchments above the gully head, the melting of which would cause

NU

gully erosion (Christenson et al., 2003; Dickson and Head, 2009; Johnson et al., 2014). Evaporation of volatiles, such as water would also decrease the erosive power as it moves downslope (e.g.,

MA

Masse et al., 2016). Analysis thus shows that gully erosion is mainly from the gully head. Similar analysis conducted on dry ravines, while revealing an inverse power law, also showed this to be

D

more linear. This would be expected where slope failure would occur near the ravine head and lose

TE

energy as it falls on slopes near the angle of repose and above the angle of kinetic friction

CE P

(Poulequin et al., 1999; Kolb et al., 2010). Our findings for Martian and terrestrial gully depth vs. slope were in general agreement with those by Zue et al. (2014), who found depth increased linearly or quadratically with increases in slope.

AC

When fitting linear or quadratic functions to our data, we found that applying a power law provided the best fit to our studied gullies. This is consistent with exponential relationships between soil loss and slope by Odemerho (1986). The depth vs. slope inverse power relationship for the dry ravines was found to be profoundly different from either of our analysed or previously published gully results (e.g., Xue et al., 2014) and showing a very poor fit. The relationship indicated that ravine slopes are much more random than terrestrial or Martian gullies and tend to decrease with depth for dry ravines. This analysis indicates that dry ravines have eroded via a different process than Martian and terrestrial gullies. These features possess spurs and gullies within their wide alcoves, with talus flows extending down the crater wall consistent with terrestrial dry mass wasting features (Johnsson et al., 2014).

ACCEPTED MANUSCRIPT 4.2. Slopes Previous research has indicated the relationship between host slope and gully curvature (Rowntree,

T

1991; Hobbs et al., 2013). Our comparisons with Martian dry ravines and geologically controlled

IP

sites such as Lake George and Cooma highlight the influence that local geology has on morphology.

SC R

Most of the Martian gully erosion/deposition slopes are below the angle of kinetic friction for dry material, supporting previous assessments that dry mass wasting is not enough to explain the

NU

erosion of Martian gullies (Conway et al., 2011; Conway and Balme, 2016). This is also highlighted by the consistently higher erosion/deposition slopes (median 33o, Fig. 7C) and depositional fan

MA

slopes (median 32.5o, Fig. 7C) found for our dry ravines. The escarpment slopes on which these features are located are also higher (median 32.5o, Fig. 7C). These higher slopes suggest that the

D

flows within the dry ravines are likely dry mass wasting deposits coming to rest at angles

TE

consistently higher than for the Martian and terrestrial gullies (Figs. 7A, B). Analogous features have been studied within volcanic flows in Argentina (Reiss et al., 2014) and on escarpments in

CE P

New Zealand (Hobbs et al., 2014a). Volatiles are thus required to enable erosion at these lower slope angles. In the case of terrestrial gullies that can be directly surveyed, evidence of liquid water

AC

erosion has emerged as a dominant process, even in the case of the Antarctic Dry Valley gullies where the presence of liquid water is rare (Levy et al., 2009; Levy, 2014). As Earth possesses no CO2 cycle, CO2 processes can be safely ruled out for terrestrial gullies. Unfortunately the lack of high resolution elevation data available for terrestrial dry ravines precluded our ability to analyse these features. In the case of Mars, CO2 has been observed eroding gullies in present times (Dundas et al., 2010, 2012, 2015). A body of work has indicated differing conditions between present day Mars and that of periods of high polar obliquities, where modelling suggests water-based processes would be more dominant (Harrison et al., 2015; Gulick et al., 2017). Additionally, some analysis has indicated multiple erosive stages for gullies where water-based erosion may have dominated gully erosion followed by CO2 and dry-based modification (Harrison et al., 2015).

ACCEPTED MANUSCRIPT The higher range of host escarpment slope angles for the terrestrial gullies (Fig. 7A) indicates greater geomorphic variability in terrestrial slopes and possibly underlying geology as compared to Martian gullies (Fig. 7B). This was also observed in talus slope measurements by Conway and

T

Balme (2016). Our Martian gullies are exclusively located within pole-facing crater walls, eroded

IP

into material with little or no evidence of bedrock. In contrast, our previous research of Lake

SC R

George and Cooma gullies revealed channel longitudinal profiles influenced by bedrock exposures. The more similar nature of the geologic settings (fine-grained regolith, little or no evidence of

NU

bedrock) of our studied Martian gullies may thus likely have played a role in constraining the range of host escarpment slopes. In future work we aim to expand on these findings to include a greater

MA

range of Martian gullies, including those in the northern hemisphere and those closer to the north and south poles. These features were beyond the scope of this work as we wished to study mid-

D

latitude gullies in order to characterise features where possible historic mechanisms involved

TE

erosion by liquid water (e.g., Dickson and Head, 2009; Kolb et al., 2010; Conway et al., 2011a) as

CE P

opposed to other processes that may have influenced higher latitude gullies (Dundas et al., 2010, 2012, 2015; Raak et al., 2015; Nunez et al., 2016).

AC

Our finding that Martian gully sinuosity is lower than terrestrial gullies is in agreement with previous research (Dickson and Head, 2009; Conway and Balme, 2016). This is likely because of the higher median slopes for the Martian gullies we measured (median 27.5°, Fig. 7C) compared to terrestrial gullies (median 24.5°, Fig. 7B). Relationships between sinuosity and slope have been studied by Mangold et al. (2010) and Zue et al. (2014), as well as the presence and abundance of volatiles that enhance erosion (Mangold et al., 2010). Our sinuosity findings along with other geomorphic measurements are consistent with the presence of volatiles capable of producing sinuous channels through erodible material. The lower abundance of these volatiles is also consistent with higher deposition and fan slopes of Martian gullies compared to those found on Earth (median deposition 18°, median fan 16.5°, Mars c.f. median deposition 10°, median fan 8°, Earth, Figs. 7B and C). Previous research has indicated the paucity of erosive volatiles compared

ACCEPTED MANUSCRIPT with terrestrial gullies, where many of the gully sites are subject to snowmelt, rainfall, and even animal erosion (Kumar et al., 2010; Hobbs et al., 2013). This greater variety of erosive processes on

T

Earth may account for lower gully slope angles.

IP

The very poor fit of width and length for Lake George was probably caused by projections of

SC R

bedrock close to the surface, as observed in our previous study (Hobbs et al., 2013), precluding channel erosion past a certain depth. The channel widens around the location of the bedrock protrusion, causing a departure from the width/depth trends observed in other gullies. This trend,

NU

where bedrock at the channel base inhibited further incision, was also observed in gullies in southwest Montana (Gabet and Bookter, 2008). Gullies studied in this survey showed variations in

MA

their fitted relationships as indicated by these trends not being consistent across all tested parameters for the gullies we studied. Examples included gullies exhibiting low R2 values for one

D

relationship (e.g., R2 = 0.0907, Lake George width/depth, Table 5) not necessarily equating to

TE

consistently low R2 values in other relationships (0.7611, Lake George width/length/length, Table

CE P

5). We noted that the quality of fitted relationships for each Martian gully site ranged from high (R2 = 0.8944, Lyot Crater gullies, Table 4) to very low (R2 = 0.0144, Newton Crater gullies, Table 4). Similar variability was also found within the terrestrial gully sites from high (R2 = 0.8818, Meteor

AC

Crater, Table 5) to very low (R2 = 0.0907, Lake George, Table 5). This was also highlighted by variability of R2 values between relationships for each gully site. These suggest that variability of gully morphology is present within each gully site; gully erosion responds to differences in geology, slope, and climate at the local level. Martian gullies in the mid-latitudes tend to be located on escarpments facing the south Martian pole. Temperature modelling and previous research has indicated that pole-facing escarpments allow accumulation of ice-rich deposits while such processes are limited on equator-facing slopes during high obliquity excursions (Christensen, 2003; Dickson and Head, 2009; Dickson et al., 2009; Morgan et al., 2010; Harrison et al., 2015). Melting of shallow ground-ice deposits, often combined with changes in regolith or threshold slopes, may have enabled channel erosion. The power of the erosion would be affected by localized characteristics of

ACCEPTED MANUSCRIPT individual gullies, such as the size of the alcove in which volatiles could accumulate or the local aspect angle at which the gully faces the pole, allowing shading to provide cold traps (Dickson and Head, 2009; Schon and Head, 2011). Localized changes in regolith have also been shown to

T

influence channel type (Harrison et al., 2013), and the precise nature into what Martian gullies

IP

erode remains uncertain because of a lack of surface missions to these areas. Thus, localized

SC R

conditions and gully context are important factors that influence how gullies have eroded. Within the constraints of localized influences, our study of mid-latitude southern Martian

NU

hemisphere gullies indicated morphological relationships (Fig. 7) more consistent with erosion by liquid water than mass wasting processes (Araki, 2012; Reiss et al., 2014; Yue et al., 2014; Conway

MA

et al., 2015; Conway and Balme 2016). These relationships were also observed in analogous terrestrial gullies where water is the dominant process (Fig. 4). Some research has indicated that

TE

D

CO2 processes are active in Martian gullies and could even explain all gully erosion in the absence of water. If this was the case, then such a process would have to behave in a manner similar to

present study.

AC

5. Conclusion

CE P

liquid water erosion, capable of generating the geomorphologic relationships observed in the

We analysed the channel and adjacent host escarpment morphology of mid-latitude Martian gullies in the Martian southern hemisphere and compared them with terrestrial gullies from various sites where liquid water erosion is the dominant process. We excluded gully alcoves from this analysis in order to concentrate on regions more likely eroded by liquid water versus dry mass wasting processes. These results were then compared with Martian dry ravine sites. We found the following similarities between Martian and terrestrial gullies:

•

Martian and terrestrial gullies are heavily influenced by local geology and environment, influencing geomorphometric relationships.

ACCEPTED MANUSCRIPT •

Martian and terrestrial gully channels were sinuous (median 1.05, Earth; median 1.02, Mars), whereas dry ravines were linear.

•

Martian and terrestrial gullies exhibited U or V-shaped channels whose width vs. depth ratios

IP

With local variations, power relationships were found between volume vs. length, depth vs.

SC R

•

T

were comparable and showed greater depths compared to widths than the dry ravines studied.

slope, and width/length vs. length for the Martian and terrestrial gullies. The depth vs. slope power relationship for the dry ravines was very poor and inverse to the Martian and terrestrial

•

NU

gullies.

Gully depositional slopes were consistently below the angle of repose (median 12o, Earth;

MA

median 18o, Mars), as were those of the depositional fans (median 8o, Earth; median 16.5o, Mars). In contrast, depositional and fan slopes of the dry ravines were higher (deposition

TE

•

D

median 33o, median fan 32.5o) with little difference between the two. Martian and terrestrial gullies were on concave host escarpment slopes and, with local

CE P

variations, had profiles that were more concave than the host escarpment profiles and more concave than dry ravines. In contrast, dry ravine profiles were more linear or convex.

AC

We found that morphological relationships between Martian and terrestrial gullies were similar, indicating water-based erosion as a likely process for erosion. These are distinctly different from results obtained from the dry ravine analysis. Variations in relationships across gully sites reflect the importance of considering the part that local environment and geology play on gully morphology. In addition, a wealth of terrestrial gully research has shown that gullies may erode through a complex interaction of a number of processes. Thus the likelihood of complex erosion factors should not be ruled out for Martian gullies. Acknowledgements We thank Trent Hare for his advice on acquiring and converting planetary data sets into GIS. We also thank the people responsible for managing the MRO mission, CTX and HiRISE instrument,

ACCEPTED MANUSCRIPT and MGS MOLA instruments, and for providing the data used in this work. In addition, we thank Dr. Chris Okubo for supplying the DTM and orthorectified imagery of the Kaiser Crater gullies. We

AC

CE P

TE

D

MA

NU

SC R

IP

T

would also like to thank two anonymous reviewers and the editor for enhancing this work.

ACCEPTED MANUSCRIPT References

T

Abell, R.S. & Truswell, E.M., 1985, Geology of the Lake George Basin, N.S.W. / by R.S. Abell; with

IP

an appendix by E.M. Truswell, Bureau of Mineral Resources, Geology and Geophysics,

SC R

Canberra, A.C.T., 57 pp.

Anderson, D.M., Reynolds, R.C., Brown, J., 1969. Betonite debris flows in Northern Alaska.

NU

Science 164, 173-174, doi: 10.1126/science.164.3876.173.

Araki, S., 2012. Morphology and formation of gully features on Mars using Mars Reconnaissance

MA

Orbiter context images. M.Sc. thesis, University of Illinois, Chicago.

D

Baker, V.R., Hamilton, C.W., Burr, D.M., Gulick, V.C., Komatsu, G., Luo, W., Rice Jr, J.W.,

TE

Rodriguez, J.A.P., 2015. Fluvial geomorphology on Earth-like planetary surfaces: A review.

CE P

Geomorphology 245, 149-182, doi:10.1016/g.geomorph.2015.05.002. Barlow, N., 2008. Mars: An Introduction to its Interior, Surface and Atmosphere. Cambridge

AC

University Press, New York.

Billi, P. & Dramis, F., 2003, Geomorphological investigation on gully erosion in the Rift Valley and the northern highlands of Ethiopia, Catena, 50(2-4), 353-368, doi: 10.1016/S03418162(02)00131-5. Black, B.A. & Thorsteinsson, T., 2008, Mars gully analogs in Iceland: Evidence for seasonal and annual variations. Abst 8026, Workshop on Martian Gullies: Theories and Tests, February 45,

2008,

Houston,

Texas.

Available

from:

http://www.lpi.usra.edu/meetings/gullies2008/pdf/8026.pdf. Bourke, M. C., 2013. A new dry hypothesis for the formation of Martian linear gullies. Icarus, 225, 526-537.

ACCEPTED MANUSCRIPT Brardioni, F., Hassan, M.A., 2006. Glacial erosion, evolution of long-river profiles, and the organization of process domain in mountain drainage basins of coastal British Columbia. J.

T

Geophys. Res. F. Earth Surf. 111, doi: 10/1029/2005/F000358.

IP

Broxton, M.J., Beyer, R.A., Moratto, Z., Lundy, M. & Husmann, K., 2011, The Ames Stereo

SC R

Pipeline: NASA’s Open Source Automated Stereogrammetry Software, A part of NASA NeoGeography Toolkit, Version 2.00, Intelligent Robotics Group, NASA Ames Research

NU

Centre, 88 pp.

Bureau of Meteorology, 2015. http://www.bom.gov.au/climate/averages/tables/cw 070217.shtml.

MA

Website accessed 28 June 2015.

Capra, A., Mazzara, L.M., Scicolone, B., 2005. Application of the EGEM model to predict

TE

D

ephemeral gully erosion in Sicily, Italy. Catena 59, 133-146. Christensen, P.R., 2003. Formation of recent Martian gullies through extensive melting of water-

CE P

rich deposits. Nature 422(6927), 45-48. Christensen, P.R., 2006, Mars Global Surveyor Thermal Emission Spectrometer Data Processing

AC

User’s Guide Version 1.6. Available from: http://tes.asu.edu/mgst/document/process.pdf [Last accessed 9 April 2012]. Coelho Netto, A.L., Fernandes, N.F., De Deus, C.E., 1988, Gullying in the south- eastern Brazilian Plateau, Bananal (SP). In: Bordas, M.P., Walling, D.E. (Eds.), Sediment Budgets (Proceedings of the Porto Alegre Symposium, December 1988), IAHS Publ., 174, International Association of Hydrological Sciences Wallingford, Oxfordshire, pp. 35-42. Conway, S.J., Balme, M.R., 2014. Decemeter thick remnant glacial ice deposits on Mars. Geophys. Res. Let. 41, doi:10.1002/2014GL060314.

ACCEPTED MANUSCRIPT Conway, S.J., Balme, M.R., 2016. A novel topographic parameterization scheme indicates that Martian gullies display the signature of liquid water. Earth and Planetary Science Letters 454,

T

doi:10.1016/j.epsl.2016.08.031.

IP

Conway, S.J., Balme, M.R., Murray, J.B. & Towner, M.C., 2009, Debris flow as a mechanism for

SC R

forming Martian gullies, Geophysical Research Abstracts, 11, EGU2009-5095. Conway, S.J.; Balme, M.R.; Murray, J.B.; Towner, M.C., Okubo, C.H., Grindod, P.M., 2011a. The

NU

indication of Martian gully formation processes by slope-area analysis. Geol. Soc. Spec. Publ. 356, 171-201.

MA

Conway, S.J., Lamb, M.P., Balme, M.R., Towner, M.C. & Murray, J.B., 2011b, Enhanced runout and erosion by overland flow at low pressure and sub-freezing conditions: Experiments and

TE

D

application to Mars, Icarus, 211(1), 443-457, doi: 10.1016/j.icarus.2010.08.026. Corrigan, S., Gulick, V., Glines, N., Freeman, P., 2017. Gully distribution and the analysis of two

CE P

gullies in Hale Crater using HIRISE DTMS. Proceedings of the 48th Annual Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute,

AC

Houston, TX, abst 2876.

Costard, F., Forget, F., Mangold, N., Peulvast, J.P., 2002. Formation of recent Martian debris flows by melting of near-surface ground ice at high obliquity. Science 295(5552), 110-113. De Dekker, P., 1982. Late Quaternary ostrocods from Lake George, New South Wales. Alcheringa 6, 305-318. Dickson, J.L., Head, J.W, 2009. The formation and evolution of youthful gullies on Mars: Gullies as the late-stage phase of Mars’ most recent ice age. Icarus 204(1), 63- 86. Dickson, J.L., Kerber, L., Fassett, C.I., Head, J.W., Forget, F., Madeleine, J-B., 2015. Formation of gullies on Mars by water at high obliquity: quantitative integration of global climate models

ACCEPTED MANUSCRIPT and gully distribution. Proceedings of the 46th Annual Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute, Houston, TX, abst 1035.

T

Diniega, S., Hansen, C. J., McElwaine, J. N., Hugenholtz, C. H., Dundas, C. M., McEwen, A. S.,

IP

de Haas, T., Hauber, E., Kleinhans, M. G., 2013. Local late Amazonian boulder breakdown and

SC R

denudation rate on Mars, Geophys. Res. Lett., 40, 3527–3531, doi:10.1002/grl.50726. Dundas, C.M., McEwan, A.S., Diniega, S., Byrne, S., Martinez-Alonso, S., 2010. New and recent

NU

gully activity on Mars as seen by HIRISE. J. Geophys. Res. Lett., 37, doi: 10.1029/2009GL041351.

changes

Martian

gullies.

Icarus

220,

124-143,

doi:

TE

10.1016/j.icarus.2012.04.005.

in

D

morphological

MA

Dundas, C.M., Diniega, S., Hansen, C.J., Byrne, S., McEwen, A.S., 2012. Seasonal activity and

Dundas, C.M., Diniega, McEwen, A.S., 2015. Long term monitoring of Martian gully formation

CE P

and evolution with MRO/HiRISE, Icarus 251, 244-263, doi: 10.1016/j.icarus.2014.05.013. Evans, L.S., 2003. Scale-specific landforms and aspects of the land surface: in Concepts and

AC

Modelling in Geomorphology: International Perspectives. TERRAPUB, Tokyo. Fowler-Tutt, S, 2012. The formation of Martian gullies and their “wetness”: a review. PH751 Research Review, University of Kent. Gabet, E.J., Bookter, A., 2008. A morphometric analysis of gullies scoured by post- fire progressively bulked debris flows in southwest Montana, USA. Geomorphology 96, 298-309. Glines, N.H., Gulick, V.C., 2014. Comparative study of gullies in Kaiser Crater on Mars. Proceedings of the 45th Annual Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute, Houston, TX, abst 2926.

ACCEPTED MANUSCRIPT Golombeck, M.P., and 10 colleagues, 2006. Erosion rates at the Mars Exploration Rover sites and long-term climate change on Mars. J. Geophys. Res. 111, E12S10.

T

Grasby, S.E., Proemse, B.C., Beauchamp, B., 2014. Deep groundwater circulation through the High

IP

Arctic cryosphere forms Mars-like gullies. Geology 42, 651-654.

SC R

Gulick, V.C., Glines, N.H., Freeman, P.M., Morkner, P., Narlesky, C., Corrigan, S., 2017. Geomorphic analysis of integrated gully systems on Mars. Proceedings of the 48th Annual

NU

Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute, Houston, TX, abst 1970.

MA

Hack, J.T., 1957. Studies of longitudinal stream profiles in Virginia and Maryland. US Geol. Surv.

D

Prof. Pap. 294-B, 45-97.

TE

Harrison, T.N., Osinski, G.R. & Tornabene, L.L., 2013, Relationship between host material and gully morphology on Mars. Abst 1420, 44th Lunar and Planetary Science Conference, Lunar

CE P

and Planetary Institute, Houston, Texas. Available from: http://www.lpi.usra.edu/meetings/lpsc2013/pdf/1420.pdf.

AC

Harrison, T.N., Osinski, G.R., Tornabene, L.L., Jones, E., 2015. Global documentation of gullies with the Mars Reconnaissance Orbiter Context Camera and implications for their formation. Icarus 252, 236-254. Harrison, T.N., Tornabene, Osinski, G.R., L.L., Conway, S., 2016. Thermal inertia variations in Gasa Crater, Mars, driven by gully and mass wasting activity. In: Harrison, T.N., Martian gully formation and evolution: studies from the local to global scale. PhD Thesis, The University of Western Ontario, http://ir.lib.uwco.ca/etd/3980. Harrison, T.N., Osinski, G.R., Tornabene, L.L., Stuurman, C.M., 2017. Multiple generations of gully activity in Western Utopia Planitia, Mars. Proceedings of the 48th Annual Lunar and

ACCEPTED MANUSCRIPT Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute, Houston, TX, abst 1497. Hartmann, W.K., Thorsteinsson, T., Sigurdsson, F., 2003. Martian hillslide gullies and Icelandic

IP

T

analogues. Icarus 162(2), 259-277.

SC R

Head, J.W., Marchant, D.R. Kreslavsky, M.A., 2008. Formation of gullies on Mars: 655 Link to recent climate history and insolation microenvironments implicate surface water

flow

NU

origin. Proc. Natl. Acad. Sci. 105, 13,258–213,263. doi:10.1073/ pnas.0803760105. Heldmann, J.L., Carlsson, E., Johansson, H., Mellon, M.T. & Toon, O.B., 2007, Observations of

MA

Martian gullies and constraints on potential formation mechanisms II. The northern hemisphere, Icarus, 188(2), 324-344, doi: 10.1016/j.icarus.2006.12.010.

D

Hobbs, S.W., Paull, D.J., Clarke, J.D.A., 2013. The influence of slope morphology on gullies:

TE

Terrestrial gullies in Lake George as analogues for Mars. Plan. Space Sci., 81,

CE P

http://dx.doi.org/10.1016/j.pss.2012.10.009. Hobbs, S.W., Paull, D.J., Clarke, J.D.A., 2014a. A comparison of semiarid and subhumid terrestrial

AC

gullies with gullies on Mars: implications for Martian gully erosion. Geomorphology 204, 344-365, http://dx.doi.org/10.1016/j.geomorph.2013.08.018. Hobbs, S.W., Paull, D.J., Clarke, J.D.A., 2014b. A hydrological analysis of terrestrial and Martian gullies: implications for liquid water on Mars. Geomorphology 226, 261-277, http://dx.doi.org/10.1016/j.geomorph.2014.07.034. Hobbs, S.W., Paull, D.J., Clarke, J.D.A., Roach, I.C., 2016. Multi-agent gully processes: Evidence from the Monaro Volcanic Province, Australia and Noachis Terra, Mars. Geomorphology Hoffman, N., 2000. White Mars: A new model for Mars’ surface and atmosphere based on CO 2. Icarus 146(2), 326-342.

ACCEPTED MANUSCRIPT Howard, A.D., 1995, Simulation modeling and statistical classification of escarpment planforms, Geomorphology, 12(3), 187-214, doi: 10.1016/0169-555x(95)00004-O.

T

Hugenholtz, C., 2008. Frosted granular flow: A new hypothesis for mass wasting in Martian gullies.

IP

Icarus 197(1), 65-72.

SC R

Jakosky, B.M., 1979, Effects of nonideal surfaces on the derived thermal-properties of Mars, Journal of Geophysical Research: Solid Earth, 84(B14), 8252-8262, doi:

NU

10.1029/Jb084ib14p08252.

Johnsson, A., Reiss, D., Conway, S.J., Hauber, E., Hiesinger, H. 2017. Small-scale lobes on Mars,

MA

solufluction, thaw and clues to gully formation. Proceedings of the 48th Annual Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute,

TE

D

Houston, TX, abst 1865.

Klein, S. P., White, B.R., 1988. Dynamic shear of granular material under variable gravity

CE P

conditions, AIAA Ann., 88– 0648.

Kirby, Eric, Burbank, D.W., Reheis, M., Phillips, F., 2006. Temporal variations in slip rate of the

AC

White Mountain Fault Zone, Eastern California. Earth Planet. Sci. Lett. 248, 168-185, doi: 10.1016/j.eps2006.05.026. Kolb, K.J., McEwen, A.S., Pelletier, J.D., 2010. Investigating gully flow emplacement mechanisms using apex slopes. Icarus 208(1), 132-142. Kreslavsky, M.A. & Head, J.W., 2002, Mars: Nature and evolution of young latitude-dependent water-ice-rich mantle, Geophysical Research Letters, 29(15), Artn 1719, doi: 10.1029/2002gl015392. Kring, D.A., 2007. Guidebook to the geology of Barringer meteor crater, Arizona (aka Meteor Crater). Lunar and Planetary Institute, Houston, Texas.

ACCEPTED MANUSCRIPT Kumar, P.S., Head, J.W., Kring, D.A., 2010. Erosional modification and gully formation at Meteor Crater, Arizona: Insights into crater degradation processes on Mars. Icarus 208(2), 608-620.

T

Kumar, P.S., Keerthi, V., Senthil Kumar, A., Mustard, J., Gopala Krishna, B., Amitabh, Ostrach,

IP

L.R., Kring, D.A., Kiran Kumar, A.S. & Goswami, J.N., 2013, Gullies and landslides on the

SC R

Moon: Evidence for dry-granular flows, Journal of Geophysical Research-Planets, 118(2), 206-223, doi: 10.1002/Jgre.20043.

NU

Lamb, M.P., Howard, A.D., Johnson, J., Whipple, K.X., Dietrich, W.E. & Perron, J.T., 2006, Can springs cut canyons into rock?, Journal of Geophysical Research-Planets, 111, Artn E07002,

MA

doi: 10.1029/2005je002663.

Lanza, N.L, Meyer, G.A., Okubo, C.H., Newsom, H.E., Wiens, R.C., 2010. Evidence for debris

TE

D

flow gully formation initiated by shallow subsurface water on Mars. Icarus 205(1), 103-112. Larue, J.P., 2008, Effects of tectonics and lithology on long profiles of 16 rivers of the southern

CE P

Central Massif border between the Aude and the Orb (France), Geomorphology, 93(3-4), 343367, doi: 10.1016/j.geomorph.2007.03.003.

AC

Leopold, L.B. & Luna, B.L., 1964, Fluvial Processes in Geomorphology, W.H. Freeman, San Francisco.

Levy, J.S., 2014. A hydrological continuum in permafrost environments: the morphological signatures of melt-driven hydrology on Earth and Mars. Geomorphology 240, 70-82, http://dx.doi.org/10.1016/j.geomorph.2014.02.033. Levy, J.S., Head, J.W., Marchant, D.R., Dickson, J.L., Morgan, G.A., 2009. Geologically recent gully-polygon relationships on Mars: Insights from the Antarctic Dry Valleys on the roles of permafrost, microclimates, and water sources for surface- flow. Icarus 201(1), 113-126.

ACCEPTED MANUSCRIPT Levy, J.S., Head, J.W., Dickson, J.L., Fassett, C.I., Morgan, G.A. & Schon, S.C., 2010, Identification of gully debris flow deposits in Protonilus Mensae, Mars: Characterization of a water-bearing, energetic gully-forming process, Earth and Planetary Science Letters, 294(3-

IP

T

4), 368-377, doi: 10.1016/j.epsl.2009.08.002.

SC R

Lucchitta, B.K., 1978, Morphology of chasma walls, Mars, Journal of Research of the US Geological Survey, 6(5), 651-662.

NU

Macphail, M., Fifield, L.K., Pillans, B., Davies, M. and Hope, G. 2015. Lake George revisited: New evidence for the origin and evolution of a large closed lake, Southern Tablelands, NSW

MA

Australia. Australian Journal of Earth Sciences 62, 853-871. Malin, M.C., Edgett, K.S., 2000. Evidence for recent groundwater seepage and surface runoff on

TE

D

Mars. Science 288(5475), 2330-2335.

Mangold, N., Mangeney, A., Migeon, V., Ansan, V., Lucas, A., Baratoux, D., Bouchut, F., 2010.

CE P

Sinuous gullies on Mars: Frequency, distribution, and implications for flow properties. J. Geophys. Res. [Planets] 115(E11001), doi: 10.0129/2009JE003540.

AC

Mao, L., et al., 2009. Sediment transfer process in two Alpine catchments of contrasting geomorphological settings. J. Hydrol. 364, 88-98. Martinez-Casasnovas, J.A., Ramos, M.C., Ribes-Dasi, M., 2002. Soil erosion caused by extreme rainfall events: Mapping and quantification in agricultural plots from very detailed digital elevation models. Geoderma 105(1-2), 125-140. Martinez-Casasnovas, J.A., Ramos, M.C., Poesen, J., 2004. Assessment of sidewall erosion in large gullies using multi-temporal DEMs and logistic regression analysis. Geomorphology 58, 305321.

ACCEPTED MANUSCRIPT Masse, M., Conway, S.J., Gargani, J., Patel, M.R., Pasquon, K., McEwan, A., Carpy, S., Chevrier, V., Balme, M.R., Ojha, L., Vincendon, M., Poulet, F., Costard, F., Jouannic, G., 2016. Transport processes induced by metastable boiling water under Martian surface conditions.

IP

T

Nature Geoscience 9, doi: 10.1038/NGEO2706.

SC R

McEwen, A.S., Ojha, L., Dundas, C.M., Mattson, S.S., Byrne, S., Wray, J.J., Cull, S.C., Murchie, S.L., Thomas, N. & Gulick, V.C., 2011, Seasonal flows on warm Martian slopes, Science,

NU

333(6043), 740-743, doi: 10.1126/science.1204816.

Mellon, M.T. & Phillips, R.J., 2001, Recent gullies on Mars and the source of liquid water, Journal

MA

of Geophysical Research-Planets, 106(E10), 23165-23179, doi: 10.1029/2000je001424. Mellon, M.T., Jakosky, B.M., Kieffer, H.H. & Christensen, P.R., 2000, High-resolution thermal

D

inertia mapping from the Mars Global Surveyor Thermal Emission Spectrometer, Icarus,

TE

148(2), 437-455, doi: 10.1006/icar.2000.6503.

CE P

Nachtergale, J., Poesen, L., Vandekerckhove, L., Oostwoud-Wijdenes, D., Roxo, M., 2001. Testing the ephemeral gully erosion model (EGM) in Mediterranean environments. 10th Int’l Soil

AC

Conservation Organization Meeting, Purdue University, USA. Narlesky, C.A., Gulick, V.C., 2014. Geomorphic and flow analysis for gullies in Palikir Crater. Proceedings of the 45th Annual Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute, Houston, TX, abst 2870. Nasri, M., Feiznia, S., Jafari, M. & Ahmadi, H., 2008, Using field indices of rill and gully in order to estimate erosion (case study: Menderjan Watershed in Isfahan Province, Iran), International Journal of Environmental, Ecological, Geological and Mining Engineering, 2(7), 18-24. Nourmohammadi, F., Haghizadeh, A., 2014. Factors controlling the morphology and volume-length relations of epheremeral gullies in the Western arid regions of Iran. Ecopersia 2, 613-628.

ACCEPTED MANUSCRIPT Okubo, C.H., Tornabene, L.L., Lanza, N.L., 2011. Constraints on mechanisms for the growth of gully alcoves in Gasa crater, Mars, from two-dimensional stability assessments of rock slopes.

T

Icarus 211, 207-221.

IP

Nunes, D.C., Smrekar, S.E., Safaeinili, A., Holt, J., Phillips, R.J., Seu, R. & Campbell, B., 2010,

SC R

Examination of gully sites on Mars with the shallow radar, Journal of Geophysical ResearchPlanets, 115, Artn E10004, doi: 10.1029/2009je003509.

NU

Nunez, J.I., Barnouin, O.S., Murchie, S.L., Seelos, F.P., McGovern, J.A., Seelos, K.D., Buczkowski, D.L., 2016. New insights into gully formation on Mars: constraints from

MA

composition as seen by MRO/CRISM. American Geophysical Union, in press, doi: 10.1002/2016GL068956.

D

Odemerho, F.O., 1986. Variation in erosion-slope relationship on cut-slopes along a tropical

CE P

9493.1986.tb00175.x

TE

highway. Singapore Journal of Tropical Geography, 7: 98–107. doi:10.1111/j.1467-

Okubo, C.H., Tornabene, L.L. & Lanza, N.L., 2011, Constraints on mechanisms for the growth of

AC

gully alcoves in Gasa crater, Mars, from two-dimensional stability assessments of rock slopes, Icarus, 211(1), 207-221, doi: 10.1016/j.icarus.2010.09.025. Okura, Y., Kitahara, H., Sammori, T., 2000. Fluidization in dry landslides. Eng. Geol. 56, 360–437. doi:10.1016/S0013-7592(99)00118-0. Palucis, M., McEnulty, T., 2010. Meteor Crater, Az: a terrestrial analog to study gully formation on Mars, Mapping Project Report, http://www.lpl.usra.edu/publications/books/barringer crater guidebook/LiDAR/NCA LM mapping project.pdf. Parkner, T., 2016. Gullies on Mars: a review. Journal of Geography 125, 155-161. Doi: 10.5026jgeography.125.155.

ACCEPTED MANUSCRIPT Patton, P.C., 1981, Evolution of the spur and gully topography on the Vallis Marineris wall scarps, Reports of the Planetary Geology Program 1981. National Aeronautics and Space

T

Administration Technical Memorandum 84211, 324-325.

IP

Pilorget, C., Forget, F., 2015. Formation of gullies on mars by debris flows triggered by CO 2

SC R

sublimation. Nature geoscience 9, doi: 10.1038/NGEO2619.

Pouliquen, O., 1999, Scaling laws in granular flows down rough inclined planes, Physics of Fluids,

NU

11(3), 542-548, doi: 10.1063/1.869928.

Putzig, N.E. & Mellon, M.T., 2007, Apparent thermal inertia and the surface heterogeneity of Mars,

MA

Icarus, 191(1), 68-94, doi: 10.1016/j.icarus.2007.05.013.

D

Qin, C., Zheng, F., Xu, X. Wu, H., 2016. Active stage gully morphological characteristics in the

TE

Loess Hilly-Gully region based on 3D laser scanning technique. Purdue University, Indiana.

CE P

Raack, J., Reiss, D., Appere, T., Vincendon, M., Ruesch, O., Hiesinger, H., 2015. Present-day seasonal gully activity in a south polar pit (Sisyphi Cavi) on Mars. Icarus 251, 226-243, doi:

AC

10.1016/j.icarus.2014.03.040. Reiss, D., Heisinger, H., Haulber, E., Zanetti1, M., Preusker, F., Trauthan, F., Reimann, G.M., Raack, J, Carlsson, A.E., Johnsson, A., Olvmo, M., Jaumann, R., Johansson, H.A.B., Johansson, L., McDaniel, S., 2009a. Morphologic and morphometric comparison of gullies on Svalbard and Mars. Proceedings of the 40th Annual Lunar and Planetary Science Conference, The Woodlands, Texas. Lunar and Planetary Institute, Houston, TX, abst 2870. Reiss, D., Heisinger, H., Haulber, E., Gwinner, K., 2009b. Regional differences in gully occurrence on Mars: A comparison between the Hale and Bond craters. Plan. Space Sci. 57(8), 958-974.

ACCEPTED MANUSCRIPT Renschler, C.S., Flanagan, D.C., 2008. Site specific decision making based on RTK GPS survey and six alternative elevation data sources: Soil erosion predictions. Trans. ASABE 51(2), 413-

T

424.

IP

Roach, I.C., 1999. The setting, structural control, geochemistry and mantle source of the Monaro

SC R

Volcanic Province, southeastern New South Wales. PhD Thesis University of Canberra, Australia.

NU

Rowntree, K.M., 1991. Morphological characteristics of gully networks and their relationship to host materials, Baringo District, Kenya. GeoJournal 23, 19-27, doi:10.1007/BF00204405.

MA

Schon, S.C., Head, J.W., 2011. Keys to gully formation processes on Mars: Relation to climate

D

cycles and sources of melt-water. Icarus 213(1), 428-432.

TE

Schon, S.C., Head, J.W., 2012. Gasa impact crater, Mars: Very young gullies formed from impact into latitude-dependant mantle and debris-covered glacier deposits? Icarus 218(1), 459-477.

CE P

doi: 10.1016/j.icarus.2012.01.002.

Schon, S. C., Head, J. W., Fassett, C. I., 2009. Unique chronostratigraphic marker in depositional

AC

fan stratigraphy on Mars: Evidence for ca. 1.25 Ma gully activity and surficial meltwater origin, Geology, 37, 207–210. Schumm, S.A., 1988, Variability of the fluvial system in space and time, in: Scales and Global Change, T. Rosswall, R.G. Woodmansee & P.G. Risser (eds), Wiley, New York, pp. 225-250. Selby, M.J., 1991, Earth's Changing Surface: An Introduction to Geomorphology, Clarendon Press; Oxford University Press, Oxford; New York. Selby, M.J., Hodder, A.P.W., 2000. Hillslope Materials and Processes. Oxford University Press, New York.

ACCEPTED MANUSCRIPT Smith, K.D., Priestly, K.F., 2000. Faulting in the 1986 Chalfant, California, sequence: Local tectonics and earthquake source parameters, Bulletin of the Seismological Society of

T

America, 90, 813-831.

IP

Smith, T.R., Merchant, G.E. & Birnir, B., 2000, Transient attractors: Towards a theory of the

SC R

graded stream for alluvial and bedrock channels, Computers & Geosciences, 26(5), 541-580, doi: 10.1016/S0098-3004(99)00128-4.

eastern Latvia. Baltica 19(2), 72-79.

NU

Soms, J., 2006. Regularities of gully erosion network development and spatial distribution in south-

the

Northeast

of

MA

Soufi M, Isaie H. 2012. The relationship between gully characteristics and sediment production in Iran,

Golestan

province.

Accessed

from

D

http://www.tucson.ars.ag.gov/isco/isco15/pdf/Soufi%20M_The%20relationship%

TE

20between.pdf, on 10/03/2012: 1-3.

CE P

Stock, J.D., Dietrich, W.E., 2006. Erosion of steepland valleys by debris flows. Bull. Geol. Soc. Am. 118, 1125-1148.

AC

Stot, T.A., 2011. Fluvial geomorphology progress report 2008-2009. Progress in Physical Geography, doi: 10.1177/0309133311415785. Taylor-Perron, J., Dietrich, W.E., Howard, A.D., McKean, J.A., 2003. Ice-driven creep on Martian debris slopes. Geophys Res Lett 30, doi:10.1029/2003GL017603. Treiman, A.H., 2003. Geologic settings of Martian gullies: Implications for their origins. J. Geophys. Res. [Planets] 108(E4), 8031, doi: 8010.1029/2002je001900. USDA-SCS, 1966, Technical release No. 32(Geology), US Department of Agriculture. Washington. DC, 125-142.

ACCEPTED MANUSCRIPT Wu, Y., Zheng, Q., Zhang, Y., Xiu, B., Cheng, H., Wang, Y., 2008. Development of gullies and sediment production in the black soil region of northeastern China. Geomorphology 101, 683-691.

T

Yue, Z., Hu, W., Liu, B., Liu, Y., Sun, X., Zhao, Q., Di, K., 2014. Quantitative analysis of the

AC

CE P

TE

D

MA

NU

SC R

IP

morphology of Martian gullies and insights into their formation. Icarus 243, 208-2014.

ACCEPTED MANUSCRIPT Figure Captions

T

Fig. 1. Overview and location of gullies studied and described in this paper. (A) Martian mid-

IP

latitude gullies; (B) terrestrial gullies.

SC R

Fig. 2. Martian gullies showing the location of longitudinal and cross-sectional profiles of the gullies (shown in white) and longitudinal profiles of the host slope (shown in black). (A) Kaiser

NU

Crater. (B) Hale Crater. (C, D) Primary site. (E) Newton Crater. (F) Promethei gully site. (G) Palakir Crater. (H) Gasa gully site showing insets for parts (I) and (J). (I) Gasa host crater gullies.

MA

(J) Gasa Crater gullies.

Fig. 3. Martian ravines showing the location of longitudinal and cross-sectional profiles of the

D

ravines (shown in white) and longitudinal profiles of the host slope (shown in black). (A) Primary

TE

site showing inset for Part B. (B) Primary site. (C) Gasa Crater showing inset for part (D). (D) Gasa

CE P

Crater.

Fig. 4. Overview and location of studied terrestrial gullies showing the location of longitudinal and

AC

cross-sectional profiles of the gullies and longitudinal profiles of the host slope. (A) Meteor Crater. (B) Alaska gully site showing insets for parts (C) and (D). (C, D) Alaska gullies. (E) Lake George gully site showing insets for parts (F) and (G). (F, G) Lake George gullies. (H) White Mountain. (I) Cooma gullies. Fig. 5. Examples of longitudinal and cross-profiles of studied terrestrial and Martian gullies. (A) Meteor Crater gully long profile example. (B) Palikir Crater gully long profile example. (C) Dry ravine long profile example. (D) Alaska gully cross-profile example. (E) Primary site gully crossprofile example. (F) Dry ravine cross-profile example.

ACCEPTED MANUSCRIPT Fig. 6. Quartile plots of studied terrestrial and Martian gullies and ravines. (A) Sinuosity. (B) Width vs. length. (C) Volume vs. length. (D) Elevation difference between the terrestrial and Martian gully

T

channel head and base and position of maximum concavity (percentage of total channel length).

IP

Fig. 7. Quartile plots of studied terrestrial and Martian gullies and ravines. (A) Terrestrial gully

SC R

slope. (B) Martian gully slope. (C) Martian ravine slope. (D) Terrestrial gully curvature. (E) Martian gully curvature. (F) Martian ravine positive curvature. (G) Martian ravine negative

NU

curvature.

Fig. 8. Statistical plots analysing relationships for studied terrestrial and Martian gullies. Fitted

MA

relationship formula and R2 value shown for each graph. (A) Terrestrial gully volume vs. length. (B) Martian gully volume vs. length. (C) Terrestrial gully depth vs. slope. (D) Martian gully depth vs.

D

slope. (E) Terrestrial gully depth vs. width. (F) Martian gully depth vs. width. (G) Terrestrial gully

TE

width/length vs. length. (H) Martian gully width/length vs. length.

CE P

Fig. 9. Statistical plots analysing relationships for studied Martian ravines. Fitted relationship formula and R2 value shown for each graph. (A) Ravine volume vs. length. (B) Ravine depth vs.

AC

slope. (C) Ravine depth vs. length. (D) Ravine width/length vs. length.

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 1

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 2

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 3

Figure 4

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 5

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

Figure 6

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 7

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 8

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

NU

Figure 9

ACCEPTED MANUSCRIPT Table 1

Wet flow

Dry flow

Sinuosity

Generation of sinuous channels

Little sinuosity, straight flows or following local topography

Longitudinal profile

Concave (influenced by water)

Depositional apron

Broadened, triangular shaped with noticeable relief

IP

T

Parameter

SC R

Comparison of wet vs. dry flow morphology characteristics

Linear—slightly concave

AC

CE P

TE

D

MA

NU

Narrow, linear streaks gradually tapering downslope. Little topographic relief

ACCEPTED MANUSCRIPT Table 2 Summary of datasets used in this study

Stereo pair

Resolution (m/pixel)

Gasa

ESP_014081_1440; ESP_014147_1440

Primary

ESP_011817_1395; ESP_011672_1395

1.0

Kaiser

PSP_003418_1335; PSP_003708_1335

1.0

Palikir

PSP_005943_1380; PSP_011428_1380

Newton

PSP_002620_1410; PSP_002686_1410

1.0

Hale

ESP_012241_1440; ESP_012663_1440

1.0

SC R

NU

MA

D

TE

ESP_012459_1450; ESP_012314_1450

CE P

Promethei Terra

1.0

1.0

1.0

LIDAR

0.5

NSW LIDAR

1.0

Field Survey

0.2

Alaska

NCALM LIDAR

1.0

White Mountain

LIDAR

1.0

Lake George

Cooma

AC

Meteor

IP

T

Site

ACCEPTED MANUSCRIPT Table 3 Median key morphometric values of studied Martian and terrestrial gullies

Width (m)

Depth (m)

Range (Earth)

0.8–144

<1–31

Median (Earth)

14

2

Range (Mars)

4–627

<1–55

Median (Mars)

74

8

Length (m)

Volume (m3)

NU

MA D TE

CE P AC

SC R

IP

T

Gully

96–1010

421–269810

252

421

159–3295

6260–12183650

840

418190

ACCEPTED MANUSCRIPT Table 4 Fitted relationships and R² values for geomorphic parameters of individual Martian gully sites

Width/depth

Depth/slope

Width-length/length

Gasahost large

y = 3.6855x + 25.179

y = 17.687x

y = 138.46x

R² = 0.8282

R² = 0.3226

R² = 0.8821

y = 3.1836x + 21.963

R² = 0.3293

R² = 0.8125

0.1927

-0.697

y = 77.541x

R² = 0.2615

R² = 0.7683

y = 19.165x0.1136

y = 131.42x-0.502

R² = 0.1666

R² = 0.6632

y = 4.8495x + 9.6765

y = 24.343x0.0607

y = 81.295x-0.606

R² = 0.882

R² = 0.0788

R² = 0.6288

y = 4.3602x + 70.301

y = 21.362x0.0524

y = 175.17x-0.614

R² = 0.2411

R² = 0.0144

R² = 0.7313

y = 7.468x + 41.397

y = 9.7319x0.2748

y = 369.36x-0.449

R² = 0.629

R² = 0.374

R² = 0.7754

y = 6.0826x + 14.268

y = 22.25x0.0317

y = 60.193x-0.665

R² = 0.8944

R² = 0.0127

R² = 0.7486

TE

R² = 0.6023

y = 6.9888x + 28.289

CE P

Primary

-0.58

y = 102.97x

y = 16.009x

D

Gasa

0.2036

-0.591

y = 18.536x

MA

R² = 0.4072

0.1987

NU

y = 5.0076x + 30.676

Gasahost

SC R

IP

T

Gully

R² = 0.6774

AC

Promethei Terra

Newton

Palikir

Lyot

ACCEPTED MANUSCRIPT y = 10.188x + 12.812

y = 17.116x0.1932

y = 168.53x-0.36

R² = 0.8806

R² = 0.694

R² = 0.8151

AC

CE P

TE

D

MA

NU

SC R

IP

T

Kaiser

ACCEPTED MANUSCRIPT

Table 5

Depth/slope

Meteor Crater

y = 4.1623x + 4.6271

y = 17.905x0.2643

y = 23.151x0.347

y = 21.79x0.4385

R² = 0.2635

R² = 0.7611

y = 5.1627x0.7589

y = 42.863x0.2353

R² = 0.6452

R² = 0.4833

y = 26.766x0.1343

y = 58.127x0.4747

R² = 0.3122

R² = 0.6385

y = 0.054x + 0.4526

y = 18.855x0.2522

y = 16.5x0.4217

R² = 0.3484

R² = 0.2087

R² = 0.5794

MA

D

y = 2.9053x + 15.738

TE

CE P

R² = 0.3051

Alaska

37.049x0.5571

R² = 0.792

R² = 0.0907

White Mountain

Width-length/length

R² = 0.4737

y = 1.4831x + 5.2629

Lake George

NU

R² = 0.8818

IP

Width/depth

SC R

Gully

T

Fitted relationships and R² values for geomorphic parameters of individual terrestrial gully sites

y = 2.184x + 11.994

Cooma

AC

R² = 0.7204

ACCEPTED MANUSCRIPT Table 6

Median gully, host escarpment (HE), fan and deposition (Dep) slopes in degrees for the Martian and terrestrial gully

Gully /HE

Fan

Gasahost large

30/30.5

17.5

Gasahost

30/28

17

18

Gasa

24/23

17

18

Primary

25/24.5

15

17

Promethei Terra

28/30

18

19

Newton

26/23

10

15

Palikir

20/19

7

9

19/19

13

14

22/24

14.5

18

Gully /HE

Fan

Dep

Meteor

24/26.5

8

12

Lake George

24/23

8

10.5

White Mountain

14.5/14.5

5

8

Alaska

35/37

23

27.5

Cooma

19/13.5

6

10

Terrestrial gully

SC R

MA D

TE

CE P

Lyot

AC

Kaiser

IP

Mars gully

NU

T

study sites

Dep

19

ACCEPTED MANUSCRIPT Gully /HE

Fan

Dep

Ravines

32/35

32.5

33

AC

CE P

TE

D

MA

NU

SC R

IP

T

Dry ravine

ACCEPTED MANUSCRIPT Testing the Water Hypothesis: Quantitative Morphological Analysis of Terrestrial and Martian Mid-latitude Gullies. S. W. Hobbs1*, D. J. Paull1 and J. D. A., Clarke2 1

T

School of Physical, Environmental and Mathematical Sciences, University of New South Wales Canberra, Australian Defence Force Academy, Northcott Drive, Canberra, Australian Capital Territory 2600, Australia. 2

CE P

TE

D

MA

NU

We compared terrestrial gullies with Martian mid-latitude gullies and dry ravines; Our analysis indicated Martian and terrestrial gullies shared similar geomorphology; The morphology of our studied Martian gullies is consistent with liquid water erosion; and Local geology and environmental conditions also heavily influenced morphology of all study sites.

AC

   

SC R

IP

Mars Society Australia. P.O. Box 327, Clifton Hill, VIC 3068, Australia. *Corresponding author. +6126268 8455, [email protected]