Embed Size (px)
Citation preview
1
Growth and form of the mound in Gale Crater, Mars: Slope-wind 1
enhanced erosion and transport 2
Edwin S. Kite1, Kevin W. Lewis2, Michael P. Lamb1, Claire E. Newman,3 and Mark I. 3
Richardson3 4
5
1Geological and Planetary Sciences, California Institute of Technology, MC 150-21, Pasadena CA 6
91125, USA. 7
2Department of Geosciences, Princeton University, Guyot Hall, Princeton NJ 08544, USA. 8
3Ashima Research, Pasadena CA 91125, USA. 9
10
ABSTRACT 11
Ancient sediments provide archives of climate and habitability on Mars. Gale Crater, the landing 12
site for the Mars Science Laboratory (MSL), hosts a 5 km high sedimentary mound (Mt. Sharp / 13
Aeolis Mons). Hypotheses for mound formation include evaporitic, lacustrine, fluviodeltaic, and 14
aeolian processes, but the origin and original extent of Gale’s mound is unknown. Here we show 15
new measurements of sedimentary strata within the mound that indicate ~3° outward dips oriented 16
radially away from the mound center, inconsistent with the first three hypotheses. Moreover, 17
although mounds are widely considered to be erosional remnants of a once crater-filling unit, we 18
find that the Gale mound’s current form is close to its maximal extent. Instead we propose that the 19
mound’s structure, stratigraphy, and current shape can be explained by growth in place near the 20
center of the crater mediated by wind-topography feedbacks. Our model shows how sediment can 21
initially accrete near the crater center far from crater-wall katabatic winds, until the increasing 22
relief of the resulting mound generates mound-flank slope-winds strong enough to erode the 23
2
mound. The slope-wind enhanced erosion and transport (SWEET) hypothesis indicates mound 24
formation dominantly by aeolian deposition with limited organic carbon preservation potential, and 25
a relatively limited role for lacustrine and fluvial activity. Morphodynamic feedbacks between 26
wind and topography are widely applicable to a range of sedimentary and ice mounds across the 27
Martian surface, and possibly other planets. 28
29
INTRODUCTION 30
Most of Mars’ known sedimentary rocks are in the form of intra-crater or canyon mounded 31
deposits like the mound in Gale crater (Hynek et al., 2003), but identifying the physical 32
mechanism(s) that explain mound growth and form has proved challenging, in part because these 33
deposits have no clear analog on Earth. The current prevailing view on the formation of intra-crater 34
mounds is that sedimentary layers (i.e., beds) completely filled each crater at least to the summit of 35
the present-day mound (Malin and Edgett, 2000). Subsequent aeolian erosion, decoupled from the 36
deposition of the layers, is invoked to explain the present-day topography (Andrews-Hanna et al., 37
2010; Murchie et al., 2009). Evaporitic, lacustrine, fluviodeltaic, and aeolian processes have each 38
been invoked to form the layers (e.g., Anderson and Bell, 2010; Andrews-Hanna et al., 2010; ; 39
Irwin et al., 2005; Niles and Michalski, 2009; Pelkey et al., 2004; Thomson et al, 2011). If the 40
sedimentary rocks formed as subhorizontal layers in an evaporitic playa-like setting, then >>106 41
km3 must have been removed to produce the modern moats and mounds (Zabrusky et al., 2012). 42
These scenarios predict near-horizontal or slightly radially-inward dipping layers controlled by 43
surface or ground water levels. 44
45
GALE MOUND LAYER ORIENTATIONS 46
3
To test this, we obtained bed-orientation measurements from six one-meter-scale stereo elevation 47
models using planar fits to extracted bedding profiles via the technique of Lewis and Aharonson 48
(2006). Each elevation model is constructed from a High Resolution Imaging Science Experiment 49
(HiRISE) stereopair using the method of Kirk et al. (2008). Individual measurements were rejected 50
where the angular regression error was greater than 2°, and the 81 remaining measurements were 51
averaged for each site to reduce uncertainty further, with the results shown in Figure 1. We find 52
that layers have shallow but significant dips away from the mound center, implying 3-4 km of pre-53
erosional stratigraphic relief if these dips are extrapolated to the rim. Measurements of the marker 54
bed of Milliken et al. (2010) show that its elevation varies by >1km, confirming that beds are not 55
planar. Postdepositional radially-outward tilting is unlikely. Differential compaction of porous 56
sediments, flexural response to the mound load, or flexural response to excavation of material from 57
the moat would tilt layers inward, not outward. Layers targeted by MSL near the base of Gale’s 58
mound show no evidence for halotectonics or karstic depressions at kilometer scale, and 59
deformation by mantle rebound would require the Gale mound to accumulate extremely quickly 60
(Figure 2).Therefore, these measurements permit only a minor role for deposition mechanisms that 61
preferentially fill topographic lows (e.g., playa, fluviodeltaic or lacustrine sedimentation), but are 62
consistent with aeolian processes (Figure 2). This suggests the mound grew with its modern shape, 63
and that the processes sculpting the modern mound may have molded the growing mound. 64
65
SLOPE WIND EROSION ON MARS 66
Mars is a windy place; saltating sand-sized particles are in active motion on Mars, at rates that 67
predict aeolian erosion of bedrock at 10-50 µm/yr (Bridges et al., 2012). Aeolian erosion of rock 68
has occurred within the last ~1-10 Ka (Golombek et al., 2010) and is probably ongoing. Because of 69
Mars’ thin atmosphere, slope winds are expected to dominate the circulation in craters and canyons 70
4
(Spiga and Forget, 2009). We have performed mesoscale (~4km horizontal resolution) simulations 71
of Gale Crater using the MarsWRF general circulation model (Richardson et al., 2007; Toigo et al., 72
2012) with embedded high-resolution nests, and these provide further evidence that winds in Gale 73
are expected to peak on the steep crater wall and mound slopes. Downslope-oriented yardangs, 74
crater statistics, exposed layers, and lag deposits suggest that sedimentary mounds in Valles 75
Marineris (e.g. Murchie et al., 2009) and Gale are being actively eroded by slope winds. Slope-76
enhanced winds appear to define both the large-scale and small-scale topography and stratigraphy 77
of the polar layered deposits (e.g. Holt et al., 2010; Smith and Holt, 2010), and radar sounding of 78
intracrater ice mounds near the north polar ice sheet proves that these grew from a central core, 79
suggesting a role for slope winds (Conway et al., 2012). Most of the ancient stratigraphy explored 80
by the Opportunity rover is aeolian (Metz et al., 2009), and aeolian deposits likely represent a 81
volumetrically significant component of the sedimentary rock record, including within the strata of 82
the Gale mound (Anderson and Bell, 2010). Evidence for fluvial reworking within sedimentary 83
mounds is comparatively limited and/or localized (e.g., Thomson et al., 2011; Irwin et al., 2005). 84
Quasi-periodic bedding at many locations including the upper portion of Gale’s mound implies 85
slow (~30 µm/yr) orbitally-paced accumulation (Lewis et al., 2008). These rates are comparable to 86
the modern gross atmospherically-transported sediment deposition rate (101-2 µm/yr ; Drube et al., 87
2010), suggesting that aeolian processes may be responsible for the layers. Sedimentary strata 88
within Valles Marineris are meters-to-decameters thick, laterally continuous, have horizontal-to-89
draping layer orientations, and display very few angular unconformities (Fueten et al., 2010). 90
These data suggest that sedimentary deposits formed by the accretion of atmospherically-91
transported sediment (ash, dust, impact ejecta, ice nuclei, or rapidly-saltating sand) formed readily 92
on early Mars as well as in the more recent past (Grotzinger et al., 2010; Cadieux et al., 2011). 93
5
Slope-wind erosion of indurated or lithified aeolian deposits cannot explain the outward 94
dips observed at Gale unless the topographic depression surrounding the mound (i.e., the moat) 95
seen in Figure 1 was present throughout mound growth. This implies a coupling between mound 96
primary layer orientations, slope winds, and mound relief. 97
98
MODEL 99
To explore this feedback, we aimed to develop the simplest possible model that can account for the 100
structure and stratigraphy of Mars' equatorial sedimentary rock mounds. In one horizontal 101
dimension (x), topographic change dz/dt is given by 102
103
dz/dt = D – E (1) 104
105
where D is an atmospheric source term and E(x,t) is erosion or sediment entrainment rate. Initial 106
model topography (Figure 3) is a basalt (nonerodible) crater/canyon with a flat floor of half-width 107
R and 20° slopes. Although dipping beds in the mound suggest a dominant role for aeolian 108
processes in mound growth, our model does not preclude the possibility of intermittent 109
fluvial/lacustrine deposition, which may have been partly later reworked by aeolian processes. To 110
highlight the role of slope winds in building mounds through erosion and deposition, we initially 111
assume D is constant and uniform (e.g., Niles and Michalski, 2009; Fergason and Christensen, 112
2008; Holt et al., 2010) and focus on E as the driver of wind entrainment and erosion. E typically 113
has a power-law dependence on maximum shear velocity magnitude at the air-sediment interface, 114
U: 115
E = k U α (2) 116
6
where k is an erodibility factor that depends on substrate grainsize and induration/cementation, and 117
α ~ 3-4 for sand transport, soil erosion, and rock abrasion (Kok et al., 2012). We assume that 118
sediments have some cohesive strength, most likely due to processes requiring liquid water (e.g., 119
damp or cemented sediment, bedrock, crust formation). Shallow diagenetic cementation 120
(McLennan and Grotzinger, 2008), if it occurred, could be driven by snowmelt, rainfall, or fog. 121
Eroded material does not pile up in the moat but is instead removed from the crater, for example 122
through breakdown to easily-mobilized dust-sized particles (Sullivan et al., 2008). We model shear 123
velocity magnitude as 124
(3) 125
which is the sum of a background bed shear velocity Uo and the component of shear velocity due to 126
slope winds. The max|±()| operator returns the maximum of downslope (nighttime) or upslope 127
(daytime) winds, z' is local topography, x and x' are distances from the crater center, and L is a 128
slope-wind correlation length scale that represents the effects of inertia. The slope winds are 129
affected by topography throughout the model domain, but are most sensitive to slopes within L of 130
x. 131
132
RESULTS 133
Model output characteristically produces Gale-like mound structure and stratigraphy (Figure 3 134
shows output for α = 3, D' = 0.4, L = 19km for a Gale-sized crater, the Data Repository shows the 135
results from sensitivity tests). Katabatic winds flowing down the crater walls inhibit sediment layer 136
accumulation both on the crater walls and for an inertial run-out length on the floor that scales with 137
L. Layer accumulation in the quiet crater interior is not inhibited, so layers can be deposited there. 138
Greater wind speeds close to the walls increase sediment erosion and entrainment. The gradient in 139
7
slope-wind shear velocity causes a corresponding gradient in sediment accumulation, which over 140
time defines a moat and a growing mound. Mound aggradation rate does not change significantly 141
upsection, consistent with observations that show no systematic decrease in layer thickness with 142
height (Cadieux et al., 2011; Lewis et al., 2008). Growth does not continue indefinitely; when the 143
relief of the mound becomes comparable to that of the crater walls, slope winds induced by the 144
mound itself become strong enough to erode earlier deposits at the toe of the mound. This erosional 145
front steepens the topography and further strengthens winds, so erosion propagates inward from the 146
edge of the mound, leading to a late-stage net erosional state. 147
This evolution does not require any change in external forcing with time; however, 148
simulating discrete, alternating erosional and depositional events with a constant, short 149
characteristic timescale produces the same model output. Exposure of layering at all elevations on 150
the Gale mound show it has entered the late, erosional stage. The mean dip of all sedimentary 151
layers formed in the radial cut shown in Figure 3a is 4.7°, and erosion progressively destroys the 152
steepest-dipping layers. Exhumed layers are buried to kilometer depths, but relatively briefly, 153
consistent with evidence that clay diagenesis at Gale was minimal (Milliken, 2010). During early 154
mound growth, dz/dt is not much slower than D. If D corresponds to vertical dust settling at rates 155
similar to today, then the lower Gale mound accumulated in 107-8 yr, consistent with the orbital-156
forcing interpretation of cyclic bedding (Lewis et al., 2010) that suggests that the time represented 157
by the lower Gale mound is a small fraction of Mars’ history. 158
Values of L and D on Early Mars are not known, but Gale-like shapes and stratigraphy 159
arise for a wide range of reasonable parameters (Figure DR2). Consistent with observations across 160
Mars, moats are infilled for small R/L, and for the largest R/L multiple mounds can develop within 161
a single crater. 162
8
D could vary on timescales much shorter than the mound growth timescale, for example if 163
orbital cycles pace the availability of liquid water for cementation. To illustrate this, we set D(t) = 164
D(t=0) + D(t=0)cos(nt) where n-1<<mound growth timescale, and find low-angle unconformities 165
can be preserved, with the likelihood of unconformities increasing with distance from the mound 166
center. In addition, a late-stage drape crosscuts layers within the mound core at a high angle, and is 167
itself broken up by further erosion (Fig. 2d). Thin mesa units mapped at Gale and more widely on 168
Mars have these characteristics (Malin and Edgett, 2000; Anderson and Bell, 2010). Deposition at 169
a constant long-term-average rate is unrealistic for the entire mound history because the rate of 170
sedimentary rock formation on Mars is close to zero in the modern epoch (Knoll et al., 2008), most 171
likely because atmospheric loss has restricted surface liquid water availability (Andrews-Hanna 172
and Lewis, 2011; Kite et al., 2012). To explore this, we decreased D' over time; this allows winds 173
flowing down the crater rim to expose layers and form a moat even when layers are originally near-174
horizontal. 175
176
IMPLICATIONS FOR THE “CURIOSITY” ROVER MISSION 177
Slope-wind enhanced erosion and transport is incompatible with a deep-groundwater source for 178
early diagenetic cementation of sedimentary rocks at Gale (e.g., Milliken et al., 2010), because 179
deep-groundwater-limited evaporite deposition would infill moats and produce near-horizontal 180
strata. A water source at or near the mound surface (ice weathering, snowmelt, or rainfall) is 181
predicted instead to explain those observations (e.g., Niles and Michalski, 2009; Kite et al., 2012). 182
Because perennial surface liquid water prevents aeolian erosion, we predict long dry windy periods 183
interspersed by brief wet periods at Gale, similar to observations along the Opportunity rover 184
traverse (Metz et al., 2009). Upon arriving at the mound, MSL can immediately begin to collect 185
observations that will test our model. MSL can confirm a dominantly aeolian origin using 186
9
sedimentology measurements, and constrain present-day winds using its meteorology package, past 187
winds by imaging fossilized bedforms, post-depositional tilting by measuring stream-paleoflow 188
directions, and subsurface dissolution using geochemical measurements. Unconformities, if 189
present, should be oriented away from the center of the present mound. Gale Crater’s geology is 190
diverse, and records many environments including alluvial fans, channels, and possibly lacustrine 191
sediments at the very bottom of the mound. We argue that these deposits are likely reworked by 192
aeolian processes and interbedded with aeolian deposits, necessary conditions for our model to 193
explain the dipping strata and morphology of the mound. If the bulk of the mound did form by 194
slow, perhaps orbitally-paced, aeolian sedimentation, then the preservation potential of organic 195
carbon would be low (e.g. Summons et al., 2011). 196
197
ACKNOWLEDGEMENTS 198
We thank W.E. Dietrich, W.W. Fischer, M. Mischna, A. Spiga, D.J. Stevenson, O. Aharonson, J. 199
Holt, T.C. Brothers, S. Christian, and especially K.E. Stack, for their intellectual contributions. 200
201
REFERENCES CITED 202
Anderson, R.B., and Bell, J.F., 2010, Geologic mapping and characterization of Gale Crater and 203
implications for its potential as a Mars Science Laboratory landing site. Mars 5, p. 76-128. 204
Andrews-Hanna, J.C. and Lewis, K.W., 2011, Early Mars hydrology: 2. Hydrological evolution in 205
the Noachian and Hesperian epochs, J. Geophys. Res. 116, E02007. 206
Andrews-Hanna, J.C., M.T. Zuber, R.E. Arvidson, and S.M. Wiseman, 2010, Early Mars 207
hydrology: Meridiani playa deposits and the sedimentary record of Arabia Terra, J. Geophys. 208
Res. 115, E06002, doi:10.1029/2009JE003485. 209
Bridges, N.T., et al., 2012, Earth-like sand fluxes on Mars. Nature 485, p. 339-342. 210
10
Cadieux, S.B., Constraining Martian sedimentation via analysis of stratal packaging, intracrater 211
layered deposits, Arabia Terra, Mars, http://trace.tennessee.edu/utk_gradthes/860. 212
Conway, S.J., et al., 2012, Climate-driven deposition of water ice and the formation of mounds in 213
craters in Mars’ North Polar Region. Icarus 220, p. 174-193. 214
Drube, L., et al., 2010, Magnetic and optical properties of airborne dust and settling rates of dust at 215
the Phoenix landing site. J. Geophys. Res. 115, E00E23, doi:10.1029/2009JE003419. 216
Fergason, R. L., and Christensen, P. R., 2008, Formation and erosion of layered materials: 217
Geologic and dust cycle history of eastern Arabia Terra, J. Geophys. Res. 113, E12001. 218
Fueten, F., et al., 2010, Structural analysis of interior layered deposits in Northern Coprates 219
Chasma, Mars. Earth Planet. Sci. Lett. 294, 343-356, doi:10.1016/j.epsl.2009.11.004. 220
Golombek, M., et al., 2010, Constraints on ripple migration at Meridiani Planum from Opportunity 221
and HiRISE observations of fresh craters. J. Geophys. Res. 115, E00F08. 222
Grotzinger, J., et al., 2010, Mars sedimentary geology: Key concepts and outstanding questions, 223
Astrobiology 11(1), p. 77-87. 224
Holt, J.W., et al., 2010, The construction of Chasma Boreale. Nature 465, p. 446-449. 225
Hynek, B.M., Phillips, R.J., and Arvidson, R.E., Explosive volcanism in the Tharsis region: Global 226
evidence in the Martian geologic record. J. Geophys. Res. 108(E9), 5111. 227
Irwin, R.P. III, Howard, A.D., Craddock, R.A., and Moore, J.M., 2005, An intense terminal epoch 228
of widespread fluvial activity on early Mars: 2. Increased runoff and paleolake development. J. 229
Geophys. Res. 110, E12S15, doi:10.1029/2005JE002460. 230
Kirk, R.L., et al., 2008, Ultrahigh resolution topographic mapping of Mars with MRO HiRISE 231
stereo images: Meter-scale slopes of candidate Phoenix landing sites. J. Geophys. Res. 232
113(E12), E00A24. 233
11
Kite, E.S., Halevy, I., Kahre, M.A., Wolff, M.J., and Manga, M., 2012, Seasonal melting and the 234
formation of sedimentary rocks on Mars, with predictions for the Gale Crater mound, astro-235
ph:EP, http://arxiv.org/abs/1205.6226 (submitted to Icarus) 236
Kok, J.F., Parteli, E.J.R., Michaels, T.I., and Karam, D. B., 2012, The physics of wind-blown sand 237
and dust, Rep. Prog. Phys. 75, 106901. 238
Knoll, A.H., et al., 2008, Veneers, rinds, and fracture fills: Relatively late alteration of sedimentary 239
rocks at Meridiani Planum, Mars. J. Geophys. Res., 113, E06S16. 240
Lewis, K.W., & Aharonson, O., 2006, Stratigraphic analysis of the distributary fan in Eberswalde 241
crater using stereo imagery, J. Geophys. Res. 111, E06001. 242
Lewis, K.W., et al., 2008, Quasi-periodic bedding in the sedimentary rock record of Mars. Science 243
322, p. 1532, doi:10.1126/science.1161870. 244
Malin, M.C. and Edgett, K.S., 2000, Sedimentary Rocks of Early Mars, Science 290, p. 1927-1937. 245
McLennan, S.M., and Grotzinger, J.P., 2008, The sedimentary rock cycle of Mars, in Bell, J., III, 246
editor, The Martian Surface - Composition, Mineralogy, and Physical Properties: volume 9 of 247
Cambridge Planetary Science Series, Cambridge University Press, 2008, p. 541. 248
Metz, J.M., et al., 2009, Sulfate-rich eolian and wet interdune deposits, Erebus Crater, Meridiani 249
Planum, Mars, J. Sedimentary Res. 79, 247-264, doi:10.2110/jsr.2009.033. 250
Milliken, R., 2010, The mineralogy of the four MSL landing sites, 4th MSL Landing Site 251
Workshop, http://marsoweb.nas.nasa.gov/landingsites/msl/workshops/4th_workshop/ 252
talks/4_Milliken_Mineralogy.pdf 253
Milliken, R.E., Grotzinger, J.P., and Thomson, B.J., 2010, Paleoclimate of Mars as captured by the 254
stratigraphic record in Gale Crater. Geophys. Res. Lett. 370, L04201. 255
Murchie, S., et al, 2009, Evidence for the origin of layered deposits in Candor Chasma, Mars, from 256
mineral composition and hydrologic modeling, J. Geophys. Res. 114, E00D05. 257
12
Niles, P.B., and Michalski, J.R., 2009, Meridiani Planum sediments on Mars formed through 258
weathering in massive ice deposits, Nature Geosci. 2, p. 215-220. 259
Pelkey, S.M., Jakosky, B.M., and Christiansen, P.R., 2004, Surficial properties in Gale Crater, 260
Mars, from Mars Odyssey THEMIS data, Icarus 167, p. 244-270. 261
Richardson, M.I., Toigo, A.D., and Newman, C.E., 2007, PlanetWRF: A general purpose, local to 262
global numerical model for planetary atmospheric and climate dynamics, J. Geophys. Res. 112, 263
E09001. 264
Smith, I.B., and Holt, J.W., 2010, Onset and migration of spiral troughs on Mars revealed by 265
orbital radar. Nature 465, p. 450-453. 266
Spiga, A. and Forget, F., 2009, A new model to simulate the Martian mesoscale and microscale 267
atmospheric circulation. J. Geophys. Res. 114(E13), E02009. 268
Sullivan, R., et al., 2008, Wind-driven particle mobility on Mars: Insights from Mars Exploration 269
Rover observations at “El Dorado” and surroundings at Gusev Crater. J. Geophys. Res., 270
113(E6), E06S07. 271
Summons, R.E., et al., 2011, Preservation of martian organic and environmental records, 272
Astrobiology 11, p.157-181. 273
Thomson, B.J., et al., 2011, Constraints on the origin and evolution of the layered mound in Gale 274
Crater, Mars using Mars Reconnaissance Orbiter data. Icarus 214, p. 413 – 432. 275
Toigo, A.D., Lee, C., Newman, C.E., and Richardson, M.E., 2012, The impact of resolution on the 276
dynamics of the martian global atmosphere: Varying resolution studies with the MarsWRF 277
GCM, Icarus 221, p.276-288. 278
Zabrusky, K., Andrews-Hanna, J.C., and Wiseman, S.M., 2012, Reconstructing the distribution and 279
depositional history of the sedimentary deposits of Arabia Terra, Mars. Icarus 220, p.311-330. 280
281
13
FIGURE CAPTIONS 282
283
Figure 1. Bedding orientation measurements from six locations around the margin of the Gale 284
crater mound. Individual measurements from HiRISE DTMs are marked in red, with the average 285
at each site indicated by the dip symbol. Table DR1 provides a full listing of results. At each 286
location, beds consistently dip away from the center of the mound, consistent with the proposed 287
model. Background elevation data is from the High-Resolution Stereo Camera (HRSC) 288
(http://europlanet.dlr.de/node/index.php?id=380), with superimposed geologic units from Thomson 289
et al. (2011). The MSL landing ellipse is shown in white; landing occurred within ~2 km of the 290
ellipse center. 291
14
292
Figure 2. Comparison of mound growth hypotheses to measurements, for an idealized cross-293
section of a mound-bearing crater. 294
Playa sourced byregional-to-global aquifer
Lacustrine
Fluviodeltaic
Ice-sheet/niveoaeolian processes
Growth processes Resulting stratigraphy Resulting stratigraphyModi!cation processes
Spring mound
Di"erential compaction
Landsliding/halotectonics
Lower-crustal #ow
Tectonic doming
Flexure under mound load
Aeolian processes(airfall+slope winds)
Preferential dissolution
Observations:
15
(a) 295
(b) 296
16
(c) (d) 297
298
Figure 3. Simulated sedimentary mound growth and form. Colored lines in (A) correspond to 299
snapshots of the mound surface equally spaced in time (blue being early and red being late), for a 300
radial cut from the crater wall to the crater center. The black line corresponds to the initial 301
topography. D’ is defined as the deposition rate divided by the mean erosion rate on the 302
crater/canyon floor at simulation start. (B) shows mound geometry, where the time step between 303
the dots is half as long as the time step between the lines in (A). I, II, and III highlight stages in the 304
evolution of the mound. (A) and (B) are for steady uniform deposition. Results for time-varying 305
uniform deposition appear very similar at this scale. The maximum model mound radius exceeds 306
its current radius, consistent with observations of a possible mound outlier (Fig. 24 in Anderson 307
and Bell, 2010). (C) shows stratigraphy formed for steady uniform deposition. Note moatward 308
dips. Flank erosion to form the modern deflation surface (gray) tends to remove any 309
unconformities formed near the edge of the mound, while exposing the stratigraphic record of 310
earlier phases for rover inspection. (D) shows stratigraphy resulting from sinusoidally time-varying 311
17
deposition. For visibility, only a small number of oscillations are shown. Color of strata 312
corresponds to deposition rate: blue is high D, which might correspond to wet climates, and red is 313
low D, which might correspond to dry climates (Kite et al., 2012; Andrews-Hanna and Lewis, 314
2011). Note low-angle unconformities, and late-stage flanking unit intersecting the mound core at a 315
high angle. 316
18
Data Repository materials. 317
318
1. Methods. 319
320
a. Determination of layer orientations. 1m-resolution stereo terrain models were produced from 321
High-Resolution Imaging Science Experiment (HiRISE) images, using the method of Kirk et al. 322
(2008), and best-fitting planar layer orientations were calculated via linear regression of points 323
along bedding contacts (procedure of Lewis et al., 2006). To confirm that our procedure is 324
measuring layers within the mound, and is not biased by surficial weathering textures nor by the 325
present-day slope, we made measurements around a small reentrant canyon incised into the SW 326
corner of the Gale mound (a DTM illustrating this may be obtained from the authors). Within this 327
canyon, present-day slope dip direction varies through 360°, but as expected the measured layer 328
orientations dip consistently (to the W). 329
330
b. MarsWRF simulations of Gale Crater. MarsWRF (Toigo et al., 2012) is the Mars version of 331
planetWRF (Richardson et al., 2007), an extension of the widely-used Weather Research and 332
Forecasting model. To produce the wind analysis shown in Figure DR1, MarsWRF was run as a 333
global model at 2° resolution, with three increasingly high-resolution domains “nested” over Gale 334
Crater to increase the resolution there to ~4 km. Each nested domain is both driven by its parent 335
domain, and feeds information back to the parent domain, while also responding to surface 336
variations (e.g. topography, albedo) at the higher resolution of the nest. 337
338
c. Assessment of alternative mechanisms for producing outward dips. Few geologic processes 339
can produce primary outward dips of (3±2)° (Figures 1, 2). Spring mounds lack laterally 340
19
continuous marker beds of the >10 km extent observed (Anderson & Bell, 2010). Preferential 341
dissolution, landsliding/halotectonics, post-impact mantle rebound, and lower-crustal flow can lead 342
to postdepositional outward tilting. On Early Mars, isostatic compensation timescales are <<106 yr. 343
In order for postdepositional mantle rebound to produce outward tilts, the mound must have 344
accumulated at implausibly fast rates. Mars’ crust is constrained to be ≲90 km thick at Gale’s 345
location (Nimmo & Stevenson, 2001), so lower-crustal flow beneath 155km-diameter Gale would 346
have a geometry that would relax Gale Crater from the outside in, incompatible with simple 347
outward tilting. Additionally, Gale is incompletely compensated (Konopliv et al., 2011) and 348
postdates dichotomy-boundary faulting, so Gale postdates the era when Mars’ lithosphere was 349
warm enough for crustal flow to relax the dichotomy boundary and cause major deformation (Irwin 350
& Watters, 2010). Any tectonic mechanism for the outward dips would correspond to ~3-4 km of 351
floor uplift of originally horizontal layers. This is comparable to the depth of a fresh crater of this 352
size and inconsistent with the current depth of the southern (mound-free) half of the crater if we 353
make the reasonable approximation that wind cannot quickly erode basalt. Tectonic doming would 354
put the mound's upper surface into extension and produce extensional faults (e.g., p.156 in Melosh, 355
2011), but these are not observed. Preferential dissolution leaves karstic depressions (Hovorka, 356
2000), which are not observed at Gale. Landsliding/halotectonics can produce deformed beds in 357
layered sediments on Earth and Mars (e.g. Metz et al., 2010, Hudec & Jackson 2011). These sites 358
show order-unity strain and contorted bedding, but the layers near the base of the mound show no 359
evidence for large strains at kilometer scale, except for a possible late-stage landslide on mound’s 360
north flank (Anderson & Bell, 2010). 361
362
d. Scaling sediment transport. Conservation of sediment (Anderson, 2008) in the atmospheric 363
boundary-layer can be written as: 364
20
dz/dt = D – E = CWs -E 365
Here C is volumetric sediment concentration, Ws is settling velocity, and E is the rate of sediment 366
pick-up from the bed. In aeolian transport of dry sand and alluvial-river transport, induration 367
processes are weak or absent and so the bed has negligible intergrain cohesion. C tends to E/Ws 368
over a saturation length scale that is inversely proportional to Ws (for dz/dt > 0) or E (for dz/dt <0). 369
This scale is typically short, e.g. ~1-20m, for the case of a saltating sand on Earth (Kok et al., 370
2012). Our simplifying assumption that D ( )f x≠ and therefore C ( )f x≠ implies that this saturation 371
length scale is large compared to the morphodynamic feedback of interest. For the case of net 372
deposition (dz/dt > 0) this could correspond to settling-out of sediment stirred up by dust storms 373
(e.g. Vaughan et al., 2010). These events have characteristic length scales >102 km (Szwast et al., 374
2006), larger than the scale of Gale’s mound and justifying the approximation of uniform D. For 375
the case of net erosion (dz/dt < 0), small E implies a detachment-limited system where sediment 376
has some cohesion. The necessary degree of induration is not large: for example, 6-10 mg/g 377
chloride salt increases the threshold wind stress for saltation by a factor of e (Nickling, 1984). Fluid 378
pressure alone cannot abrade the bed, and the gain in entrained-particle mass from particle impact 379
equals the abrasion susceptibility, ~2 ×10-6 for basalt under modern Mars conditions (Bridges et al., 380
2012) and generally <<1 for cohesive materials, preventing runaway adjustment of C to E/Ws. 381
Detachment-limited erosion is clearly appropriate for slope-wind erosion on modern Mars (because 382
sediment mounds form yardangs and shed boulders, indicating that they are cohesive/indurated), 383
and is probably a better approximation to ancient erosion processes than is transport-limited 384
erosion (given the evidence for ancient near-surface liquid water, shallow diagenesis, and soil 385
crusts; e.g., McLennan & Grotzinger, 2008). 386
387
21
e. Reference parameter choices. Coriolis forces are neglected because almost all sedimentary 388
rock mounds on Mars are equatorial (Kite et al., 2012). Additional numerical diffusivity at the 10-3 389
level is used to stabilize the solution. Analytic and experimental results show that in slope-wind 390
dominated landscapes, the strongest winds occur close to the steepest slopes (Manins & Sarford, 391
1987). L will vary across Mars because of 3D topographic effects, and will vary in time because of 392
changing atmospheric density. Ye et al. (1990) find L ~ 20km for Mars slopes with negligible 393
geostrophic effects, and Equation 49 in Magalhaes & Gierasch (1982) gives L ~ 25 km for Gale-394
relevant slopes. Simulations of gentle Mars slope winds strongly affected by planetary rotation 395
suggest L ~ 50-100 km (e.g., Savijärvi & Siili, 1993). Entrainment acts as a drag coefficient, ~0.02-396
0.05 for Gale-relevant slopes (Horst & Doran, 1986, and references therein), suggesting L = 20-50 397
km for a 1km-thick cold boundary layer. Therefore we take L ~ 101-2 km to be reasonable, but with 398
the expectation of significant L/R variability, explored in the next section. 399
400
2. Sensitivity tests: controls on mound growth and form. To confirm that our results do not 401
depend on idiosyncratic parameter choices, we carried out a parameter sweep in α, D’, and R/L 402
(Figure DR2). Weak slope dependence (α = 0.05) is sufficient to produce strata that dip toward the 403
foot of the crater/canyon slope (like a sombrero hat). Similarly weak negative slope dependence (α 404
= -0.05) is sufficient to produce concave-up fill. At low R/L (i.e., small craters) or at low α, D' 405
controls overall mound shape and slope winds are unimportant. When D' is high, layers fill the 406
crater; when D’ is low, layers do not accumulate. When either α or R/L or both are ≳1, slope-wind 407
enhanced erosion and transport dominates the behavior. Thin layered crater floor deposits form at 408
low D', and large mounds at high D'. If L is approximated as being constant across the planet, then 409
R/L is proportional to crater/canyon size. There is net aggradation everywhere for small R/L, 410
although a small moat can form as a result of relatively low net aggradation near the crater wall. 411
22
For larger R/L, moats form, and for the largest craters/canyons, multiple mounds can form 412
eventually because slope winds break up the deposits. This is consistent with data, which suggest a 413
maximum length scale for mounds (Figure DR3). Small exhumed craters in Meridiani show 414
concentric layering consistent with concave-up dips. Larger craters across Meridiani, together with 415
the north polar ice mounds, show a simple single mound. Gale and Nicholson Craters, together 416
with the smaller Valles Marineris chasmata, show a single mound with an undulating top. The 417
largest canyon system on Mars (Ophir-Candor-Melas) shows multiple mounds per canyon. Gale-418
like mounds (with erosion both at the toe and the summit) are most likely for high R/L, high α, and 419
intermediate D' (high enough for some accumulation, but not so high as to fill the crater) (Figure 420
DR2). 421
Uo is set to zero in Figure 3. Sensitivity tests show that for a given D', varying Uo has little 422
effect on the pattern of erosion because spatial variations are still controlled by slope winds. 423
Equation (3) implies the approximation E ~ max(U)α ~ ∑Uα, which is true as α à ∞. To check 424
that this approximation does not affect conclusions for α = 3-4 (Kok et al., 2012), we ran a 425
parameter sweep with E ~ (U+α + U-
α). For nominal parameters (Figure 3), this leads to only minor 426
changes in mound structure and stratigraphy (e.g., 6% reduction in mound height and <1% in 427
mound width at late time). For the parameter sweep as a whole, the change leads to a slight 428
widening of the regions where the mound does not nucleate or overspills the crater (changing the 429
outcome of 7 out of the 117 cases shown in Figure DR2). The approximation would be further 430
supported if (as is likely) there is a threshold U below which erosion does not occur. If MSL shows 431
that persistent snow or ice was needed as a water source for layer cementation (Niles & Michalski, 432
2009; Kite et al., 2012), then additional terms will be required to track humidity and the drying 433
effect of föhn winds (e.g. Madeleine et al., 2012). 434
23
These sensitivity tests suggest that mounds are a generic outcome of steady uniform 435
deposition modified by slope-wind enhanced erosion and transport for reasonable Early Mars 436
parameter values. 437
438
Data Repository References 439
Anderson, R.S., 2008, The Little Book of Geomorphology: Exercising the Principle of 440
Conservation, http://instaar.colorado.edu/~andersrs/The_little_book_010708_web.pdf 441
Horst, T. W., & Doran, J. C., 1986, Nocturnal drainage flow on simple slopes. Boundary-Layer 442
Meteorol. 34, p. 263-286. 443
Hovorka, S. D., 2000, Understanding the processes of salt dissolution and subsidence in sinkholes 444
and unusual subsidence over solution mined caverns and salt and potash mines, Technical 445
Session: Solution Mining Research Institute Fall Meeting, p. 12–23, downloaded from 446
http://www.beg.utexas.edu/environqlty/pdfs/hovorka-salt.pdf 447
Hudec, M.R., & Jackson, M.P.A., 2011, The salt mine : a digital atlas of salt tectonics. Austin, Tex: 448
Jackson School of Geosciences, University of Texas at Austin. 449
Irwin, R. P., III, and T. R. Watters, 2010, Geology of the Martian crustal dichotomy boundary, J. 450
Geophys. Res. 115, E11006, doi:10.1029/2010JE003658. 451
Konopliv, A.S. et al., 2011, Mars high resolution gravity fields from MRO, Mars seasonal gravity, 452
and other dynamical parameters, Icarus 211, p. 401-428. 453
Madeleine, J.-B., Head, J. W., Spiga, A., Dickson, J. L., & Forget, F., 2012, A study of ice 454
accumulation and stability in Martian craters under past orbital conditions using the LMD 455
mesoscale model, Lunar and Planet. Sci. Conf. 43, abstract no. 1664. 456
Magalhaes, J., & Gierasch, P., 1982, A model of Martian slope winds: Implications for eolian 457
transport, J. Geophys. Res. 87, p. 9975-9984. 458
24
Manins, P. C., & Sawford, B. L.,1987, A model of katabatic winds, J. Atmos. Sci. 36, 619-630. 459
Metz, J., Grotzinger, J., Okubo, C., & Milliken, R., 2010, Thin-skinned deformation of sedimentary 460
rocks in Valles Marineris, Mars, J. Geophys. Res. 115, E11004. 461
Melosh, H.J., 2011, Planetary Surface Processes, Cambridge University Press. 462
Nickling, W.G., 1984, The stabilizing role of bonding agents on the entrainment of sediment by 463
wind. Sedimentology 31, 111-117. doi: 10.1111/j.1365-3091.1984.tb00726.x. 464
Nimmo, F., and Stevenson, D.J. 2001, Estimates of Martian crustal thickness from viscous 465
relaxation of topography, J. Geophys. Res. 106, 5085-5098, doi:10.1029/2000JE001331. 466
Richardson, M.I., Toigo, A.D., and Newman, C.E., 2007, PlanetWRF: A general purpose, local to 467
global numerical model for planetary atmospheric and climate dynamics, J. Geophys. Res. 112, 468
E09001. 469
Szwast, M., Richardson, M. and Vasavada, A., 2006, Surface dust redistribution on Mars as 470
observed by the Mars Global Surveyor and Viking orbiters. J. Geophys. Res. 111, E11008. 471
Savijärvi, H., and Siili, T., 1993, The Martian slope winds and the nocturnal PBL jet. J. Atmos. Sci. 472
50, p. 77-88. 473
Vaughan, A.F., et al., 2010. Pancam and Microscopic Imager observations of dust on the Spirit 474
Rover: Cleaning events, spectral properties, and aggregates, Mars 5, p. 129-145. 475
Ye, Z.J., Segal, M., & Pielke, R.A., 1990, A comparative study of daytime thermally induced 476
upslope flow on Mars and Earth. J. Atmos. Sci. 47, p. 612-628. 477
25
Data Repository Table 1: Layer orientation measurements 478
Lat Lon Z (m) Dip (°) Dip Az (°) HiRISE Image ID -‐5.022347 138.394900 -‐3263.1 3.53 30.68 PSP_008437_1750 -‐5.023877 138.392660 -‐3201.9 2.52 62.01 PSP_008437_1750 -‐5.015876 138.386310 -‐3216.9 2.1 94.68 PSP_008437_1750 -‐5.015508 138.387020 -‐3201.4 7.31 41.72 PSP_008437_1750 -‐4.998358 138.391680 -‐3554.2 2.06 54.81 PSP_008437_1750 -‐5.003035 138.387800 -‐3429.9 0.43 -‐21.29 PSP_008437_1750 -‐5.004517 138.379910 -‐3425.8 5.04 89.54 PSP_008437_1750 -‐5.004179 138.379580 -‐3434.5 3.79 70.24 PSP_008437_1750 -‐4.997492 138.392530 -‐3583.8 4.65 51.98 PSP_008437_1750 -‐5.012374 138.396710 -‐3421.5 4.14 47.28 PSP_008437_1750 -‐5.031424 138.395590 -‐3290.1 4.07 40.92 PSP_008437_1750 -‐5.030106 138.396180 -‐3308.2 2.55 76.18 PSP_008437_1750 -‐5.03171 138.393740 -‐3260.1 3.24 43.31 PSP_008437_1750 -‐5.035189 138.392050 -‐3176.3 2.07 75.25 PSP_008437_1750 -‐5.035062 138.391700 -‐3173.3 2.21 86.77 PSP_008437_1750 -‐4.685812 137.494850 -‐4098.5 3.34 148.1 ESP_023957_1755 -‐4.684778 137.491970 -‐4103.8 3.98 174.47 ESP_023957_1755 -‐4.689331 137.480520 -‐4101.3 6.06 132.82 ESP_023957_1755 -‐4.659387 137.533150 -‐4120.3 6.48 114.03 ESP_023957_1755 -‐4.65886 137.537400 -‐4108.2 0.5 13.56 ESP_023957_1755 -‐4.662119 137.535520 -‐4073.1 3.9 -‐149.05 ESP_023957_1755 -‐4.664102 137.525670 -‐4127.1 7.16 130.19 ESP_023957_1755 -‐4.663656 137.524870 -‐4138.3 4.83 152.05 ESP_023957_1755 -‐4.665528 137.526530 -‐4101.2 4.59 -‐153.08 ESP_023957_1755 -‐4.676802 137.506150 -‐4083.2 2.32 83.97 ESP_023957_1755 -‐4.672137 137.510280 -‐4128 3.24 91.62 ESP_023957_1755 -‐4.871501 137.270980 -‐3849.6 1.09 10.19 PSP_001488_1750 -‐4.871837 137.266710 -‐3857.7 5.19 85.97 PSP_001488_1750 -‐4.872691 137.270730 -‐3833.2 2.61 -‐51.44 PSP_001488_1750 -‐4.917952 137.284340 -‐3513.9 6.78 136.55 PSP_001488_1750 -‐4.831027 137.330630 -‐3768.9 2.16 140.09 PSP_001488_1750 -‐4.828565 137.330360 -‐3792.5 2.65 143.84 PSP_001488_1750 -‐4.846642 137.303380 -‐3802.8 4.38 124.94 PSP_001488_1750 -‐4.845724 137.303070 -‐3815.7 4.51 111.27 PSP_001488_1750 -‐4.845501 137.304420 -‐3799.2 2.34 79.11 PSP_001488_1750 -‐4.863885 137.332420 -‐3507.1 2.04 132.56 PSP_001488_1750 -‐4.93696 137.311640 -‐3278.4 3.74 129.01 PSP_001488_1750 -‐4.938936 137.309840 -‐3287.6 4.46 119.79 PSP_001488_1750 -‐4.920126 137.322160 -‐3290.9 1.79 134.69 PSP_001488_1750 -‐4.922859 137.317020 -‐3306.6 4.2 160.7 PSP_001488_1750 -‐4.901912 137.338220 -‐3265.6 4.87 -‐174.83 PSP_001488_1750
26
-‐4.892401 137.332800 -‐3389.9 6.71 117.93 PSP_001488_1750 -‐4.863151 137.342100 -‐3464.4 5.66 105.35 PSP_001488_1750 -‐4.779928 137.409690 -‐3656.9 3.69 167.09 PSP_009149_1750 -‐4.777029 137.405330 -‐3736.8 2.13 168.14 PSP_009149_1750 -‐4.75303 137.438670 -‐3810.7 6.07 128.67 PSP_009149_1750 -‐4.752131 137.438100 -‐3823.7 5.92 123.62 PSP_009149_1750 -‐5.348423 137.227120 -‐2745.9 2.68 154.47 ESP_012907_1745 -‐5.348098 137.227470 -‐2757.1 2.32 144.88 ESP_012907_1745 -‐5.341016 137.209820 -‐2891.2 2.58 135.34 ESP_012907_1745 -‐5.337968 137.210930 -‐2796.8 4.54 151.03 ESP_012907_1745 -‐5.377012 137.207730 -‐2879.1 1.63 165.41 ESP_012907_1745 -‐5.384663 137.190610 -‐3005.3 4.73 156.7 ESP_012907_1745 -‐5.413138 137.185680 -‐2848.5 4.57 133.72 ESP_012907_1745 -‐5.40772 137.197850 -‐2779.3 4.05 158.01 ESP_012907_1745 -‐5.392686 137.208330 -‐2756.1 10.83 163.21 ESP_012907_1745 -‐5.371398 137.174520 -‐3359.7 6.02 159.19 ESP_012907_1745 -‐5.342561 137.176890 -‐3130.9 6.76 160.11 ESP_012907_1745 -‐5.458482 137.183800 -‐2752.8 2.49 142.76 ESP_012907_1745 -‐5.459221 137.188840 -‐2674.5 3.58 162.7 ESP_012907_1745 -‐5.460307 137.188440 -‐2696.5 3.97 166.94 ESP_012907_1745 -‐5.392535 137.196770 -‐2848.2 2.48 170.27 ESP_012907_1745 -‐5.390379 137.195930 -‐2898.2 3.04 156.57 ESP_012907_1745 -‐5.390219 137.191070 -‐2988.8 2.6 144.9 ESP_012907_1745 -‐5.411481 137.194250 -‐2811.9 2.09 146.02 ESP_012907_1745 -‐5.623904 138.328080 -‐2971.2 1.74 -‐43.56 ESP_014186_1745 -‐5.627848 138.337550 -‐3000.2 2.17 -‐71.76 ESP_014186_1745 -‐5.598681 138.339300 -‐2856.2 1.26 116.18 ESP_014186_1745 -‐5.584488 138.338620 -‐2705.9 2.25 -‐69.68 ESP_014186_1745 -‐5.584654 138.354940 -‐2710 2.45 -‐82.66 ESP_014186_1745 -‐5.605716 138.360870 -‐2907.7 0.3 -‐175.97 ESP_014186_1745 -‐5.570108 138.366460 -‐2674.6 5.23 -‐61.61 ESP_014186_1745 -‐5.572424 138.360000 -‐2589.6 2.17 -‐86.77 ESP_014186_1745 -‐5.588566 138.326250 -‐2685.2 2.6 -‐27.67 ESP_014186_1745 -‐5.580666 138.333270 -‐2681.5 1.76 -‐81.46 ESP_014186_1745 -‐5.534321 138.325470 -‐2424.8 4.35 -‐117.38 ESP_014186_1745 -‐5.542813 138.327860 -‐2454.8 3.37 -‐104.06 ESP_014186_1745 -‐5.573589 138.289520 -‐2658.6 2.2 -‐115.72 ESP_014186_1745 -‐5.566658 138.288200 -‐2644.4 1.15 74.46 ESP_014186_1745 -‐5.581102 138.309950 -‐2667.3 1.06 -‐65.97 ESP_014186_1745
479
480
481
27
Data Repository Figure Captions 482
483
Figure DR1. Annual maximum wind speed (m/s) within Gale Crater from MarsWRF simulations, 484
showing that the strongest winds within the crater are associated with steep slopes. Black 485
topography contours are spaced at 500m intervals. The winds are extrapolated to 1.5m above the 486
surface using boundary layer similarity theory (the lowest model layer is at ~9m above the 487
surface). 488
3.2966 6.4362 9.5758 12.715 15.855 18.995
4.8664 8.0060 11.146 14.285 17.425
Maximum wind speed
136.5 137.0 137.5 138.0 138.5Longitude (deg E)
-6.5
-6.0
-5.5
-5.0
-4.5
-4.0
Latit
ude
(deg
N)
28
489
Figure DR2. Overall growth and form of sedimentary mounds – results from a model parameter 490
sweep varying R/L and D', with fixed α = 3. Black square corresponds to the results shown in more 491
detail in Figure 3. Symbols correspond to the overall results:– no net accumulation of sediment 492
anywhere (blue open circles); sediment overtops crater/canyon (red filled circles); mound forms 493
and remains within crater (green symbols). Green filled circles correspond to outcomes where 494
layers are exposed at both the toe and the summit of mound, similar to Gale. 495
29
496
Figure DR3. Width of largest mound does not keep pace with increasing crater/canyon width, 497
suggesting a length threshold beyond which slope winds break up mounds. Blue dots correspond to 498
nonpolar crater data, red squares correspond to canyon data, and green dots correspond to polar ice 499
mound data. Gray vertical lines show range of uncertainty in largest-mound width for Valles 500
Marineris canyons. Blue dot adjacent to “G” corresponds to Gale Crater. Craters smaller than 10km 501
were measured using Context Camera (CTX) or HiRISE images. All other craters, canyons and 502
mounds were measured using the Thermal Emission Imaging System (THEMIS) global day 503
infrared mosaic on a Mars Orbiter Laser Altimeter (MOLA) base. Width is defined as polygon area 504
divided by the longest straight-line length that can be contained within that polygon. 505
100 101 102 1030
0.2
0.4
0.6
0.8
1
G
Container width (km)
Mou
nd w
idth
/con
tain
er w
idth
Nonpolar crater moundCanyon moundPolar ice mound
