Geophysical Research Letters

Candidate ice-rich material within equatorial craters on Mars



[1] The floors and walls of many mid-latitude (∼30–60°) craters on Mars appear to be mantled by relatively young material(s) with distinct morphology and erosional properties. Collectively, this material (“fill”) is often interpreted as ice-rich, with emplacement and modification related to climatological processes. Here, I document material and associated landforms within 38 craters between 4–13°S in the Sinus Sabaeus region that appear morphologically similar to material and landforms within mid-latitude craters. These equatorial/mid-latitude materials may also share a common composition and emplacement mechanism. Near-surface ice is unstable at equatorial latitudes under present conditions, suggesting that emplacement could have occurred under different climate conditions in the past. High-obliquity (35–45°) general circulation model (GCM) simulations show surface ice accumulation in Sinus Sabaeus and Tharsis, where similar material and landforms have been documented. These observations are consistent with the hypothesis that past obliquity-driven climate change resulted in equatorward volatile migration on Mars.

1. Introduction

[2] The mid-latitude regions (∼30–60°) on Mars display many distinctive landforms and textures [Squyres and Carr, 1986], including aprons around mesas (“lobate debris aprons” [e.g., Mangold, 2003]), lineated material within valleys (“lineated valley fill”) and craters (“concentric crater fill” [e.g., Levy et al., 2010]), “softened” terrain [Squyres, 1989], crater wall “viscous flow” features [Milliken et al., 2003], dissected mantle material [Mustard et al., 2001] and patterned ground [e.g., Levy et al., 2009]. Most interpretations of these mid-latitude landforms involve subsurface ice, with concentration varying from debris/dust with interstitial ice to relatively pure ice beneath a debris layer [e.g., Head et al., 2010, and references therein]. Other interpretations involve exposures of layered rock [Malin and Edgett, 2001] and eolian deposits [Zimbelman et al., 1989]. Hypotheses involving ice are consistent with morphological evidence suggestive of past deformation [e.g., Head et al., 2010] and material removal [e.g., Shean, 2010, and references therein]. This interpretation is also consistent with Mars Reconnaissance Orbiter (MRO) Shallow Radar (SHARAD) data that suggest mid-latitude aprons are composed of a material with dielectric properties similar to those of water ice containing <10% lossy contaminants (e.g., dust/rock) beneath a meters-thick debris cover [Holt et al., 2008; Plaut et al., 2009].

[3] Here, I describe previously-undocumented material within equatorial craters in the Sinus Sabaeus region (Figures 123) that appears morphologically similar to “fill” observed in mid-latitude craters, especially those between ∼25–35°N/S (Figure 3). Sinus Sabaeus is a notable low albedo feature near the martian equator, just south of the ∼460-km diameter Schiaparelli Crater (Figure 1). The region is believed to be underlain by ancient highland crust and is characterized by numerous impact craters of varied size and degradation state [e.g., Craddock et al., 1997].

Figure 1.

(a) Map of Sinus Sabaeus region and crater database sites over 128 px/deg MOLA topography. (b) Crater diameter plot with MRO Mars Color Imager (MARCI) 725 nm albedo over MOLA shaded relief. Labels indicate figure numbers. (c) Simulated surface water ice budget after 11 years at 40° obliquity (adapted from Levrard et al. [2004] with permission from Macmillan Publishers Ltd: Nature).

Figure 2.

(a) CTX image P17_007848_1729_XN_07S337W of ∼12.3 km diameter crater at 8.15°S, 336.88°W containing fill with large areal extent (∼30 km2). Note concentric fill ridges/furrows. (b) Detail of crater floor. Note extensional features (white arrows) on surfaces around fill margins, including a protruding knob. Black arrow shows apparent continuation of fill surface ridge beyond fill margins.

Figure 3.

Representative examples of equatorial (Figures 3a–3e) and mid-latitude (Figures 3f–3j) crater fill and associated landforms: (a) 8.60°S, 347.69°W (B19_017184_1711_XN_08S347W), (b) 8.66°S, 343.93°W (G01_018542_1713_XI_08S343W), (c) 9.13°S, 349.35°W (B20_017250_1694_XI_10S349W), (d) 9.76°S, 347.47°W (B19_017184_1711_XN_08S347W), (e) 6.48°S, 347.83°W (B19_017184_1711_XN_08S347W), (f) 33.21°S, 234.94°W (P17_007818_1457_XN_34S234W), (g) 24.72°N, 313.51°W (G04_019622_2034_XN_23N313W), (h) 33.91°S, 236.09°W (B17_016257_1461_XI_33S236W), (i) 35.34°S, 167.59°W (P17_007776_1429_XI_37S167W), (j) 28.01°S, 91.09°W (B18_016700_1498_XN_30S090W).

[4] A preliminary survey of available 6 m/px MRO Context Camera (CTX) data did not identify this material within any other equatorial craters. This prompted the creation of a database containing 2140 craters >2 km in diameter within a study area spanning 30°S–5°N and 330–360°W. All available high-resolution (<20 m/px) image data were examined for each location, with primary reliance on CTX data. At least 38 equatorial craters in the Sinus Sabaeus region contain distinctive floor fill material, with >30 additional craters displaying evidence suggestive of past fill presence.

2. Observations and Interpretations

2.1. Crater Fill Morphology

[5] The fill surface morphology in Sinus Sabaeus ranges from smooth to dissected with numerous pits, fractures, knobs and ridges (Figures 2 and 3). Generally, this material appears domical (convex-upward in cross-section), with thicker central regions and thinner margins. In some craters (e.g., Figure 3a3b), Mars Global Surveyor Mars Orbiter Laser Altimeter (MOLA) shot data indicate that this material is >100 m thick relative to adjacent crater floor surfaces, and up to 300–400 m thick for extrapolated cavity shapes (2nd-order polynomial and power law fits to crater walls, see auxiliary materials). Figure 3 shows examples of fill morphology, which could represent different fill “classes” or stages of fill evolution.

[6] Many fill surfaces display surface ridge-forms and/or collections of concentric ridges/furrows that appear arcuate in planform (typically convex relative to crater walls, Figures 2 and 3). Some ridge crests display longitudinal fractures and/or central depressions (Figure 2), while others appear boulder-strewn. These ridges/furrows appear to indicate that this material experienced deformation during emplacement and/or subsequent evolution.

2.2. Crater Floor/Wall Ridges

[7] Several craters in the Sinus Sabaeus region (Figure 1) display concentric, curvilinear ridges on their floors and/or interior walls. Mutliple collections of these ridges are present on the floor of a youthful ∼16 km diameter crater (Figure 4), with some ridges extending ∼5.5 km from the base of the southeast crater wall. Surfaces circumscribed by these ridges often display knobs, and fill material is present between some proximal ridges and crater walls (Figure 4). These ridges appear draped over irregular topography and several display “cross-cutting” superposition relationships (Figure 4), suggesting multiple episodes of ridge formation occurred with limited modification of existing landforms.

Figure 4.

(a) CTX image B19_017052_1685_XN_11S343W of ∼16.6 km diameter crater at 9.63°S, 343.74°W with crater floor ridges. Several of the ridges closest to crater walls bound fill material (white arrows). Black arrows show extent of ridges on east/southeast floor. Inset shows ridge cross-cutting relationships (X). (b) Detail of ridges.

[8] These floor/wall ridges could potentially be marginal depositional features related to lateral fill retreat. Similar ridges are observed elsewhere on Mars, including those beyond fill margins within mid-latitude craters [e.g., Shean, 2010; Levy et al., 2010], within at least two ∼70°N craters [e.g., Garvin et al., 2006], and on the western flanks of Arsia Mons [Shean et al., 2007] and Olympus Mons (e.g., CTX image P03_002329_1986_XN_18N139W, Milkovich et al. [2006]).

2.3. Evidence for Fill Removal

[9] Several morphological indicators of fill removal previously documented for mid-latitude craters [Shean, 2010; Levy et al., 2010] are also observed within craters in Sinus Sabaeus. In addition to the crater floor/wall ridges, these include: 1) fill surface ridges and/or collections of knobs that appear continuous across fill margins and adjacent surfaces (Figure 2), 2) circumferential fill-facing scarps, benches, and/or extensional features concentric to, and typically within ∼200–600 meters of, fill margins on adjacent surfaces (Figures 2 and 3), and 3) depressions, pits, and/or fractures on fill surfaces (Figures 2 and 3).

[10] The circumferential features appear to be relatively young, as they cut materials on crater walls (talus) and are superposed only by dunes. These features may delineate the former location(s) of fill margins before/during removal, implying several hundred meters of apparent lateral retreat (possibly the result of fill surface lowering). Some of the extensional features could also have formed as a result of mass movements, potentially triggered by instabilities related to fill removal on lower slopes. At least 27 additional craters in the region display these circumferential features with no distinguishable floor fill material (Figure 1), suggesting that complete fill removal may have occurred at these locations.

[11] The processes responsible for this removal could involve subaerial erosion (e.g., eolian deflation) and/or subsurface material loss (e.g., sublimation of volatiles). Surface depressions and pits appear to be the result of collapse related to the latter.

2.4. Spatial Distribution

[12] The low-latitude crater fill in Sinus Sabaeus is concentrated between 4–13°S and 335–357°W, spanning an area of ∼7.2 × 105 km2 and a broad elevation range (−1.30 to 2.63 km, Figure 1). Craters containing the thickest fill are observed between 8–11°S and 344–350°W, within and adjacent to the southeast rim of a ∼500 km diameter basin that predates Schiaparelli (Figure 1b). Nearly all low-latitude craters containing fill material are 2.0–9.0 km in diameter (median 5.3 km) and they tend to appear relatively youthful with steep interior wall slopes of ∼15–30°.

[13] The observed spatial distribution suggests that this material was preferentially emplaced and/or preserved within these craters. Observed differences in present fill thickness and surface morphology could be related to relative age and/or local setting (e.g., crater morphometry, wall material properties), which could also affect subsequent removal processes. Alternatively, a continuous layer of this material could have covered the region earlier in Mars' history, implying that the observed distribution of fill within smaller, steep-walled craters could be primarily related to preservation, with complete removal from larger craters and inter-crater surfaces.

3. Discussion

[14] Several possible interpretations for the fill material were considered. It could be lithified/cemented material with a composition and origin similar to crater floor materials observed in Arabia Terra and Meridiani Planum, potentially representing past localized subaerial/subaqueous crater infilling or the remnants of a regional layer [e.g., Malin and Edgett, 2000b]. However, the morphology of these well-documented sedimentary outcrops (e.g., prominent layering, yardangs) differs considerably from the Sinus Sabaeus crater fill, arguing against a shared origin.

[15] The fill could represent an extrusive material, such as salt domes/diapirs sourced from a regional subsurface layer [e.g., Jackson and Talbot, 1986]. Similar processes have been invoked to explain domical features elsewhere on Mars [e.g., Adams et al., 2009], although these examples do not display the distinctive evidence for removal seen in Sinus Sabaeus. This explanation is also difficult to reconcile with the observed distribution, elevation range, and the lack of fill within adjacent, larger craters.

[16] The morphological similarities between the equatorial and mid-latitude fill material suggest that they potentially share a similar internal composition, are susceptible to similar erosional processes, and share a similar emplacement mechanism. Neither of the aforementioned interpretations can effectively account for the presence of fill in thousands of craters within constrained mid-latitude ranges in both hemispheres.

[17] A composition involving volatiles (most likely water ice) and debris can explain the observed fill morphology and evidence for removal in the mid-latitudes, suggesting that the same may be true for the equatorial fill. The distribution of these mid-latitude landforms is consistent with climatically controlled/induced processes for emplacement and subsequent modification [e.g., Squyres and Carr, 1986; Head et al., 2003, 2010]. Past interpretations suggest that favorable environmental conditions, coupled with local setting, could initiate direct deposition/accumulation of surface frost/snow, diffusion of atmospheric water vapor into the subsurface, and/or deformation of existing ice-rich material. Present-day seasonal H2O frost distribution is controlled by insolation and surface temperature variations, with deposition on steep, pole-facing slopes at latitudes as low as 13°S [Vincendon et al., 2010]. In principle, a similar phenomenon could preferentially deposit and/or trap volatiles within steep-walled craters in Sinus Sabaeus under favorable climate conditions.

[18] The present mean annual temperature in Sinus Sabaeus is ∼220 K and thermal stability and diffusion models suggest that near-surface water ice is unstable at equatorial latitudes under present conditions [e.g., Mellon et al., 1997]. However, Mars experiences large orbital variations [e.g., Laskar et al., 2004], and both thermal/diffusion models [e.g., Jakosky and Carr, 1985; Mellon and Jakosky, 1995] and GCM simulations [e.g., Forget et al., 2006] predict increased atmospheric water vapor levels and enhanced ice stability at lower latitudes for high obliquities (>35–45°). These conditions favor water ice precipitation [Forget et al., 2006] and/or diffusion into the subsurface [Mellon and Jakosky, 1995] at low latitudes.

[19] GCM simulations for an obliquity of 40° and the current north polar cap predict two equatorial locations for surface ice accumulation, likely due to orographic effects and a “westerly monsoon-like circulation” [Montmessin, 2006] - the western flanks of the large Tharsis volcanoes and southeast of Schiaparelli basin (Figure 1c) [Levrard et al., 2004]. More recent GCM simulations also predict accumulation in Sinus Sabaeus for alternate orbital/atmospheric parameters and volatile sources [Madeleine et al., 2009]. These results suggest that Sinus Sabaeus is a favorable location for surface ice accumulation during high-obliquity conditions, and they potentially provide an explanation for the observed crater fill distribution on and to the west-northwest of elevated topography (Figure 1).

[20] The western flanks of the Tharsis Montes [Head and Marchant, 2003; Shean et al., 2005, 2007] and Olympus Mons [Milkovich et al., 2006] display morphological evidence suggestive of recurrent glacial activity, with proposed formation under high obliquity conditions [e.g., Forget et al., 2006]. Materials within depressions between 4–11°S at Arsia Mons (e.g., CTX image P08_004122_1705_XI_09S125W, Shean et al. [2007]) share many morphological similarities with the fill material in Sinus Sabaeus. Both appear relatively young [Shean et al., 2006] and display similar evidence for deformation and subsequent removal.

[21] It is interesting to note that although some surfaces in the Sinus Sabaeus region appear mantled, no definitive evidence for the characteristic mid-latitude dissected mantle [Mustard et al., 2001] or gully activity [e.g., Malin and Edgett, 2000a] is observed. The documentation of this material in Sinus Sabaeus independently supports hypotheses suggesting that equatorward volatile migration occurred under different climate conditions on Mars in the relatively recent past. It is unclear from available data whether any relict ice is currently present at these locations, although estimates for fill thickness are noteworthy. The equatorial setting suggests that if present, this ice is likely buried by a thick, insulating debris layer or a near-surface layer of reduced permeability.


[22] I would like to thank the MSSS science group, B. Haberle, J. Dickson, and C. Fassett for helpful input and productive discussions. S. Byrne and J. Levy provided constructive reviews. This work was funded by the NASA Mars Reconnaissance Orbiter project through JPL contract 1275776.