Author's personal copy - Brown University · Author's personal copy Concentric crater ll: Glacial and periglacial change 465 Fig. 4. An elevated region (center) surfaced by closed-cell

Author's personal copy

Icarus 202 (2009) 462–476

Contents lists available at ScienceDirect

Icarus

www.elsevier.com/locate/icarus

Concentric crater fill in Utopia Planitia: History and interaction between glacial“brain terrain” and periglacial mantle processes

Joseph S. Levy a, James W. Head a,∗, David R. Marchant b

a Brown University, Department of Geological Sciences, 324 Brook Street, Box 1846, Providence, RI 02912, United Statesb Boston University, Department of Earth Sciences, 675 Commonwealth Ave., Boston, MA 02215, United States

a r t i c l e i n f o a b s t r a c t

Article history:Received 21 October 2008Revised 27 January 2009Accepted 24 February 2009Available online 3 March 2009

Keywords:MarsMars, climateMars, surfaceIces

At martian mid-to-high latitudes, the surfaces of potentially ice-rich features, including concentric craterfill, lobate debris aprons, and lineated valley fill, typically display a complex texture known as “brainterrain,” due to its resemblance to the complex patterns on brain surfaces. In order to determine thestructure and developmental history of concentric crater fill and overlying latitude-dependent mantle(LDM) material, “brain terrain” and polygonally-patterned LDM surfaces are analyzed using HiRISE imagesfrom four craters in Utopia Planitia containing concentric crater fill. “Brain terrain” and mantle surfacetextures are classified based on morphological characteristics: (1) closed-cell “brain terrain,” (2) open-cell“brain terrain,” (3) high-center mantle polygons, and (4) low-center mantle polygons. A combined glacialand thermal-contraction cracking model is proposed for the formation and modification of the “brainterrain” texture of concentric crater fill. A similar model, related to thermal contraction cracking anddifferential sublimation of underlying ice, is proposed for the formation and development of polygonallypatterned mantle material. Both models require atmospheric deposition of ice, likely during periods ofhigh obliquity, but do not require wet active layer processes. Crater dating of “brain terrain” and mantledsurfaces suggests a transition at martian mid-latitudes from peak “glacial” conditions occurring withinthe past ∼10–100 My to a quiescent period followed by a cold-desert “periglacial” period during the past∼1–2 My.

© 2009 Elsevier Inc. All rights reserved.

1. Introduction

Glaciation, and ice-related processes, have shaped martian mid-dle and high latitudes during the late Amazonian (Mustard etal., 2001; Kreslavsky and Head, 2002; Head et al., 2003, 2006a;Kuzmin, 2005; Forget et al., 2006; Fastook et al., 2008; Head andMarchant, 2009). The record of variable Amazonian climate con-ditions is indicated by a variety of martian landforms (Marchantand Head, 2007), including the latitude-dependent mantle (LDM)(Mustard et al., 2001; Head et al., 2003; Milliken et al., 2003;Schon et al., 2008); concentric crater fill, lobate debris aprons,and lineated valley fill (Squyres, 1979; Lucchitta, 1984; Squyresand Carr, 1986; Head et al., 2006b); polygonally patterned ground(Mangold, 2005; Levy et al., 2009; Mellon et al., 2008); andpedestal craters (Kadish and Barlow, 2006; Kadish et al., 2008).Questions arise as to the nature and timing of Amazonian climatechange, and whether conditions that might have led to meltinghave occurred (Kreslavsky et al., 2008; Soare et al., 2008). Further,unusual textures are observed on the surfaces of many Amazonian

* Corresponding author. Fax: +1 401 863 3978.E-mail addresses: [email protected] (J.S. Levy), [email protected]

(J.W. Head), [email protected] (D.R. Marchant).

examples of concentric crater fill, lineated valley fill, and lobate de-bris aprons. These textures include pit-and-butte texture (Mangold,2003), “knobs—brain coral” (Williams et al., 2008), “brain coral ter-rain” (Dobrea et al., 2007), or, succinctly, “brain terrain” (Levy etal., 2009); the origin, age, and climate conditions represented bythese surface textures remain an area of active research.

We examine concentric crater fill surface textures in UtopiaPlanitia in order to assess initial emplacement conditions and pro-cesses (Figs. 1 and 2). We document relationships between con-centric crater fill surfaces and overlying latitude-dependent mantle(LDM) material (hereafter referred to simply as mantle material)in four Utopia Planitia craters. Finally, we date these surfaces us-ing crater retention ages and propose a model for their forma-tion, and modification history. Although several hypotheses havebeen suggested for the origin of concentric crater fill, ranging fromaeolian modification (Zimbelman et al., 1989) to rock-glacier pro-cesses (Mangold and Allemand, 2001), lineated and lobate concen-tric crater fill surface patterns strongly indicate glacier-like flow(Head et al., 2006a; Levy et al., 2007). “Brain terrain” appears to bea modification of concentric crater fill lobe and lineation patterns.In contrast, the latitude-dependent mantle (LDM) is described as aflat-lying, or draped surface unit, meters to tens of meters thick,which has a variety of characteristic surface textures, including

0019-1035/$ – see front matter © 2009 Elsevier Inc. All rights reserved.doi:10.1016/j.icarus.2009.02.018


Concentric crater fill: Glacial and periglacial change 463

Fig. 1. Context map of study area in Utopia Planitia. Individual HiRISE image locations are shown. Base map is MOLA shaded relief topography.

Fig. 2. Image location maps for subsequent figures. (a) Concentric crater fill and mantle material from PSP_002782_2230 over CTX image P05_002782_2232. North to imagetop and illumination is from the lower left. (b) Concentric crater fill and mantle material from PSP_002175_2210 over CTX image P03_002175_2211. North to image top andillumination is from the lower left.


464 J.S. Levy et al. / Icarus 202 (2009) 462–476

Fig. 3. (a) Closed-cell “brain terrain.” Sinuous, elevated “cells” are approximately 20 m wide, and up to ∼100 m long. Image is 200 m wide. North to image top. Portionof PSP_002175_2210. (b) Open-cell “brain terrain.” Sinuous ridges of positive topography outline flat-floored “cells.” Images is 200 m wide. North to upper right. Portion ofPSP_002175_2210. (c) High-center mantle polygons. Image is 200 m wide. North to image top. Portion of PSP_002175_2210. (d) Low-center mantle polygons. Image is 100 mwide. North to image top. Portion of PSP_002782_2230.

polygonal patterning, pitting, and scalloping (Mustard et al., 2001;Head et al., 2003; Milliken et al., 2003; Schon et al., 2008). Inthe Utopia Planitia study region, both “brain terrain” and polyg-onally patterned mantle are present and in some locations, “brainterrain” underlies the polygonally patterned mantle. Detailed anal-ysis of polygon morphology at the mantle surface provides insightinto the processes of mantle emplacement, and in places, of inter-actions and modification of the underlying “brain terrain.” Thesemorphological and stratigraphic relationships permit us to recon-struct a history of ice-related processes in the martian middle-to-high latitudes during the recent Amazonian. These processesrange from debris-covered glacial activity (i.e., the formation ofalpine and debris-covered glaciers that remain long-lived due tothe preservative effects of capping sublimation lags) (Marchant etal., 2002; Kowalewski et al., 2006) to cold desert “periglacial” pro-cesses (cold-climate, non-glacial geomorphological processes suchas permafrost development, thermal contraction cracking, etc.)(Washburn, 1973).

2. Morphology

2.1. “Brain terrain”

At HiRISE resolution (∼30 cm/pixel), the surface of concentric-crater-fill “brain terrain” displays a complex morphology composedof smaller, discrete surface structures that we term “cells.” Twodistinct textures are commonly present in “brain terrain” observedin Utopia Planitia concentric crater fill: closed-cell “brain terrain”

and open-cell “brain terrain” (Figs. 3–5). Similar features have beenobserved elsewhere, notably on lineated valley fill and lobate de-bris apron surfaces (Williams et al., 2008; Dobrea et al., 2007).

Closed-cell “brain terrain” in Utopia Planitia exhibits arcuate,mounded cells with both flat and rounded upper surfaces. Cells arecommonly ∼10–20 m wide, ∼10–100 m long, and ∼4–5 m high(based on HiRISE image shadow measurements). Some closed-cell“brain terrain” cells have surface grooves or furrows located nearthe centerline of the long axis (Figs. 6–8). Closed-cell “brain ter-rain” cells occur singly, or in linked groups (Figs. 6–8). Closed-cell“brain terrain” cells commonly form lineations that are orientedconcentrically to the crater in which the unit is present (Figs. 5,9 and 10). Spacing between closed-cell “brain terrain” lineationsis variable, but is commonly ∼20 m (Figs. 5–8). Closed-cell “brainterrain” is commonly present on undulating topography, at the topof concentric ridges (and sometimes in the concentric valleys be-tween ridges) (Figs. 5, 9 and 10).

Some closed-cell “brain terrain” units have a strongly polygo-nal surface texture (compare closed-cell “brain terrain” mounds inFig. 6 to Figs. 7 and 8). Polygonal “brain terrain” commonly fea-tures a surface furrow oriented axially at the center of the mound-shaped cell (Figs. 7 and 8).

Open-cell “brain terrain” is composed of arcuate and cuspatecells that are delimited by a convex-up boundary ridge, commonly∼4–6 m wide and ∼2 m high (based on HiRISE image shadowmeasurements), surrounding a flat-floored depression (Figs. 3–5).Open-cell “brain terrain” cells are of similar dimensions to closed-cell “brain terrain” cells. Open-cell “brain terrain” boundary ridges



Fig. 4. An elevated region (center) surfaced by closed-cell “brain terrain” (CC-BT) and surrounded by open-cell “brain terrain” (OC-BT). Positive topography can be seen inthe stereo view in Fig. 9. The “brain terrain” is ringed by mantling (LDM) surfaces. High-center mantle polygons (HC-MP) are visible immediately surrounding the open-cell“brain terrain.” Low-center mantle polygons (LC-MP) are present to the lower left. Illumination is from the left. North to image top. Portion of PSP_002175_2210.

Fig. 5. Alternating open-cell “brain terrain” (OC-BT) in topographic lows between ridges surfaced by closed-cell “brain terrain” (CC-BT). A schematic topographic profilerepresenting surface undulations across the middle of the image is included in white, with increasing elevation towards image top. Illumination is from the left. North toimage top. Portion of PSP_002175_2210.

are commonly parallel along the long axis, but may be tightlyrounded or gradually tapered along the short axis (Figs. 3–5).Open-cell “brain terrain” cells occur singly, or in linked groups(Figs. 3–5). Open-cell “brain terrain” cells commonly form lin-eations that are oriented concentrically to the crater in whichthe unit is present (Figs. 5, 9 and 10). Open-cell “brain terrain”lineation spacing is variable, but is commonly ∼20 m. Open-cell“brain terrain” is commonly present at the lateral contact between“brain terrain” and the mantle (Figs. 3–5), and in topographicallylow valleys between “brain terrain” ridges (Figs. 9 and 10). Open-cell “brain terrain” is also common in topographic lows betweenclosed-cell “brain terrain”-patterned ridges and hills (Figs. 3–5and 10).

Boulders are present on the surface of both open-cell andclosed-cell “brain terrain,” and are typically less than 2 m in di-

ameter (Figs. 3 and 5). Boulders are most commonly found atopclosed-cell mounds and on open-cell boundary bands, althoughsome boulders are present in topographic lows.

2.2. Polygonalized mantle

In the analyzed craters, mantle material is present in a continu-ous ring around the crater interior wall, and as patchy occurrencesin topographic lows between crater fill ridges. Mantle materialpresent in proximity to “brain terrain” is polygonally patterned anddisplays two distinct textures (Fig. 3): high-center mantle poly-gons and low-center mantle polygons (Figs. 11 and 12). High-centermantle polygons are bounded by depressed surface troughs whichintersect at both near-orthogonal and near-hexagonal intersections,forming polygonal patterns with topographically high interiors rel-



Fig. 6. Closed-cell “brain terrain” present in PSP_002175_2210. Axial furrows are present along some “brain terrain” mounds (arrows). North to image top. Illumination isfrom the left.

Fig. 7. Contact between strongly lineated closed-cell “brain terrain” (left) and polygonal closed-cell “brain terrain” (right). Axial furrows are present in many closed-cell “brainterrain” mounds (Arrows). Portion of PSP_002782_2230. Illumination is from the lower left. North to image top.

ative to their boundaries (Figs. 11 and 12). High-center mantlepolygons are commonly ∼10 m in diameter, with slightly convex-up interiors. High-center mantle polygon troughs are commonly∼2–3 m across. The mantle material in which high-center poly-gons are present can be up to ∼40 m thick, based on MOLA pointmeasurements, but thins and pinches out at lateral contacts with“brain terrain” and steep crater wall surfaces. On average, mantlematerial in the examined craters must be thick enough to smooth-over topographic undulations in underlying “brain terrain”—likelyindicating a typical depth of 10–20 m. The mantle unit is gener-ally flat, and is bounded by gently sloping margins, as well as bysteeply scarped, scalloped margins (Figs. 11 and 12).

Low-center mantle polygons (Figs. 3, 11 and 12) are com-posed of troughs with raised shoulders that intersect at near-orthogonal and near-hexagonal intersections, forming polygonswith depressed centers, relative to their raised rims (Figs. 11

and 12). Low-center mantle polygons are commonly ∼10 m in di-ameter (but may be as small as ∼5 m in diameter, e.g., Fig. 12),and have smooth, flat, depressed interiors. Low-center mantlepolygon troughs are commonly ∼3–4 m wide. Low-center man-tle polygons are found at the fringes of the mantle, at both gradualand scalloped margins (Figs. 11 and 12).

2.3. Spatial relationships between brain terrain and mantle polygons

Whereas the superposition relationship between mantle mate-rial and “brain terrain” is generally clear, detailed mapping revealslocally complex lateral contacts between “brain terrain” and man-tle material (Fig. 13). At lateral contacts between mantle and “brainterrain” units, mantle material is commonly draped on, and inter-fingered between, “brain terrain” “cells,” suggesting that mantlematerial superposes, and in places, embays “brain terrain” units



Fig. 8. Closed-cell “brain terrain” with pronounced axial furrows (arrows) at the confluence of two concentric crater fill lobes in PSP_002782_2230. North to image top.Illumination is from the left.

Fig. 9. Three-dimensional surface structure of concentric crater fill and mantle ma-terial. “Brain terrain” visible at full resolution appears as knobby surface roughnessand is strongly lineated concentric to the crater, with concentric ridges and val-leys observable. “Brain terrain”-covering LDM material is located around the craterinterior wall, and in low areas between “brain terrain”-patterned ridges. Crater is∼10 km in diameter. North to image top. Red–blue (left/right) anaglyph is a por-tion of stereo pair PSP_002175_2210 and PSP_001410_2210. Anaglyph produced byJames Dickson.

Fig. 10. Close-up view of concentric crater fill ridges and valleys present in Fig. 9(right side, below middle). Ridges are surfaced with closed-cell and some open-cell “brain terrain.” Valleys are surfaced primarily by open-cell “brain terrain.”Red–blue (left/right) anaglyph is a portion of stereo pair PSP_002175_2210 andPSP_001410_2210, and is ∼2 km wide. North to image top. Anaglyph produced byJames Dickson.

(Figs. 4, 14, and 15). Mantle material is commonly found at thefoot of crater interior wall slopes, and in topographic lows between“brain terrain”-surfaced concentric ridges within craters (Figs. 9and 10). Exposures of underlying “brain terrain” microtopographycrop out through the mantle, suggesting that the mantle thinsand pinches out at lateral contacts with “brain terrain” (Figs. 14and 15). Topographically high exposures of closed-cell “brain ter-rain” are commonly ringed by open-cell “brain terrain,” which is inturn ringed by mantled surfaces patterned with low-center, and/orhigh-center mantle polygons (Figs. 4 and 5). Concentric crater fillridges surfaced by closed-cell “brain terrain” are commonly flankedby open-cell “brain terrain” in the lows between ridges, particu-larly in lows that also have exposures of mantle material (Figs. 10,



Fig. 11. Low-center mantle polygons (LC-MP) present in a scalloped depression in PSP_002782_2230. High-center mantle polygons (HC-MP) surround the low-center mantlepolygons, both within, and outside the depression. Illumination is from the lower left. North to image top.

Fig. 12. Low-center mantle polygons (LC-MP) present along a gently sloping contact between high-center mantle polygons (HC-MP) and “brain terrain” (CC-BT) inPSP_002782_2230. Illumination is from the lower left. North to image top.

14 and 15). Open-cell “brain terrain” is rarely observed withoutnear-by and/or overlying mantle material, and low-center mantlepolygons are not observed in isolation from extensive regions pat-terned by high-center mantle polygons.

Complete transitions between surfaces dominated by high-center mantle polygons, low-center mantle polygons, open-cell“brain terrain,” and closed-cell “brain terrain” occur on lengthscales of ∼100–500 m in the analyzed images (Fig. 15). Lateralcontacts between mantle surfaces patterned with high-center andlow-center mantle polygons are gradational on gentle slopes andabrupt on steeply scalloped slopes (Figs. 11 and 12). Lateral con-tacts between closed-cell “brain terrain” and open-cell “brain ter-rain” are gradational, consisting of closed-cell “brain terrain” cellsthat are partially depressed, or that transition into open-cell “brainterrain”-like boundary ridges (Figs. 4 and 5). A schematic stratig-

raphy of “brain terrain” and polygonally patterned mantle surfacetextures is shown in Fig. 16.

3. Surface ages

Crater counting on concentric crater fill and mantle terrainsis complicated by the obscuration and disruption of craters bybrain terrain and mantle polygons (Mangold, 2003; Kostama etal., 2006). In this study, 168 craters were counted on mantle and“brain terrain” surfaces present in four HiRISE images. Only freshcraters showing no evidence of subsequent polygon formationwithin the crater were counted on mantle surfaces (e.g., Fig. 17a).Given that thermal contraction crack polygons can form duringpermafrost formation (syngenetic wedges) or after substrate degra-dation begins (epigenetic and anti-syngenetic wedges) (Mackay,



Fig. 13. Sketch map of “brain terrain” and mantle in PSP_002782_2230. Closed-Cell BT indicates closed-cell “brain terrain;” Open-Cell BT indicates open-cell “brain terrain;”High-Center MP indicates high-center mantle polygons; and Low-Center MP indicates low-center mantle polygons. Closed-Cell “brain terrain” is more widely distributed thanopen-cell “brain terrain,” and high-center mantle polygons are more widely distributed than low-center mantle polygons. Mapping was conducted on the HiRISE image only,overlaid on a CTX background image.

1990), counting only fresh, unfractured craters gives us a min-imum age for the cessation of mantle emplacement. Counts onmantle surfaces indicate a minimum age of ∼1.5 My (best-fit agesspan 1.3 to 1.7 My using the Neukum (Neukum and Ivanov, 2001)and Hartmann (2005) production functions, respectively) (Fig. 18)—a youthful age consistent with earlier estimates of mantle ageselsewhere in the northern hemisphere (Head et al., 2003). Formantle surfaces displaying small thermal contraction crack poly-gons, a reduction in the abundance of ∼20 m diameter craters

on mantle surfaces is typical, and may result from the removal orobscuration of craters with diameters comparable to polygon di-ameters (Levy et al., 2009).

Crater counts on “brain terrain” are likewise complicated, owingto the obscuration of small craters by “brain terrain” cells, the de-struction of small and medium-sized craters by “brain terrain” celldevelopment, and the presence of craters with unusually modifiedmorphologies (e.g., Mangold, 2003) (Fig. 17). Combined counts offresh and modified craters (including “ring-mold” and/or “oyster



Fig. 14. (a) Closed-cell “brain terrain” (center) surrounded by open-cell “brain ter-rain” on the flanks of a crater wall. Open-cell “brain terrain” grades into high-centermantle polygons, in a relationship suggesting overprinting of “brain terrain” bymantle material. Illumination is from the bottom of the image. North to right.(b) A mantle surface with high-center mantle polygons (image center) transitioningto open-cell “brain terrain,” and to closed-cell “brain terrain.” Exposure of open-cell“brain terrain” bounding ridges within the high-center mantle polygon surface sug-gests overprinting of “brain terrain” by mantle material. Illumination is from imageleft. North to image top. (a) and (b) are portions of PSP_002175_2210.

shell” craters (e.g., Mangold, 2003; Kress and Head, 2008) (Fig. 18),show a roll-off at small crater sizes typical of counts on lobatedebris aprons and lineated valley fill. This is unlikely to be domi-nantly a resolution effect, as counts were made on HiRISE imageswith spatial resolution of ∼30 cm/pixel. Deformed and extensivelymodified craters were included in the “brain terrain” surface countin order to capture the minimum age of glacial activity (see be-low) in concentric crater fill during which massive ice was presentshallowly enough to modify impact craters (Kress and Head, 2008).Critically, even deformed and modified craters are present withinthe brain terrain, rather than underlying it (as would be the casefor large craters superposed by “brain terrain”). Counts of craterslarger than 125 m in diameter suggest a surface age for “brainterrain” of ∼10–100 My, consistent with, if slightly younger, thancounts on lobate debris aprons (Mangold, 2003).

The difference between surface ages of ∼1.5 My for mantle, and∼10–100 My for “brain terrain,” suggests that “brain terrain” andmantle materials are two stratigraphically distinct units that weredeposited at markedly different times, even when scaling effectsfrom impacts into ice are considered (Kress and Head, 2009). Theaerially small exposures of open-cell “brain terrain” and low-centermantle polygons compared to the more common closed-cell “brainterrain” and high-center mantle polygon surfaces (e.g., Fig. 13) pre-clude separate age determinations.

4. Discussion

4.1. Brain terrain origin

What processes result in the formation of “brain terrain” atthe surface of concentric crater fill? To answer this question, itis first necessary to consider potential origins for concentric craterfill itself. First-order lineation patterns on concentric crater fill (onwhich “brain terrain” has developed) have been interpreted to beflow lineations and lobes analogous to those formed by the flow ofterrestrial debris-covered glaciers (Head and Marchant, 2006; Headet al., 2006a, 2006b; Shean et al., 2007). The presence of “ring-mold” craters on concentric crater fill surfaces further suggeststhat ice-rich material may be present within meters to tens ofmeters of the surface beneath a relatively ice-free lag layer (Kressand Head, 2008). Based on MOLA point topography measurementsand typical martian crater depth-diameter ratios (Garvin et al.,2002), the analyzed craters are up to 80% filled, accounting forvolumes of crater-fill material up to 800 m thick (Fig. 19). Recentradar observations of lobate debris aprons with analogous flow, lin-eation, and crater structures raises the possibility that much of theconcentric crater fill material may be ice (e.g., Holt et al., 2008;Plaut et al., 2008), consistent with predictions from global circula-tion models (Madeleine et al., 2007). An ice-rich concentric craterfill (e.g., Fig. 20a) is consistent with mapping of 10–25 km diam-eter craters in the martian northern plains indicating widespreadfilling of large impact craters with hundreds of meters of ice-richmaterials (Kreslavsky and Head, 2006).

Thick accumulations of atmospherically deposited, ice-rich,dusty material (Forget et al., 2006; Head et al., 2005, 2006a)(Fig. 20a) deposited in crater interiors could readily flow undercurrent martian conditions (Milliken et al., 2003), sufficient to pro-duce observed strains (e.g., ∼60% crater deformation in Fig. 17). Upto eight broad-scale concentric crater fill ridges can be observedper crater analyzed in this study, suggesting a complex historyof deformation of ice-rich materials. Boulder-sized clasts presenton the concentric crater fill surface could represent rock-fall en-trained at crater wall ice-accumulation zones, and transported totheir present location by glacial flow (e.g., Marchant et al., 2002;Head and Marchant, 2006; Marchant and Head, 2007).

Gravitational stresses that drive brittle deformation of near-surface glacier ice, coupled with thermal stresses generated byseasonal heating and cooling (Mellon, 1997) would fracture theice-rich crater fill, resulting in the generation of oriented fracturenetworks, analogous to those observed on flowing debris-coveredglaciers on Earth (Fig. 20b) (Levy et al., 2006). This combinationof stresses—thermal and glacial—can account for the variety ofsurface patterns and textures observed in “brain terrain.” Inter-nal deformation of flowing ice-rich material generates the broad,first-order concentric lineation patterns observed in “brain terrain.”Lateral flow orients the near-surface stress field (Benn et al., 2003),aligning thermal contraction cracks both normal and orthogonal tothe flow direction; alternatively, cracks may initially form in ran-dom patterns, but are oriented during subsequent flow of subsur-face ice and debris (Marchant et al., 2002; Levy et al., 2006). Theseoriented fractures will develop into strongly lineated elongate andmound-shaped closed-cell “brain terrain” (see below). Thermalcontraction crack fractures will be less strongly oriented on “brainterrain” that has experienced minimal flow, resulting in only mod-erately oriented, and largely hexagonal fractures (Levy et al., 2006).In addition to the above, concentric crater fill might be thickestnear the center of crater floors where transported debris overly-ing glacier ice is typically at a maximum (e.g., Head et al., 2008;Kowalewski, 2008) and where the temperature conditions are mostfavorable for ice preservation (Russell et al., 2004). These factorscould result in a rise in the elevation of the crater fill at the center



Fig. 15. Contacts between “brain terrain” and mantle polygon textures. A complete transition between high-center mantle polygons (HC-MP), low-center mantle polygons(LC-MP), open-cell “brain terrain” (OC-BT) and closed-cell “brain terrain” (CC-BT) is present in each panel. All panels are excerpted from PSP_002782_2230, with illuminationfrom the left and north to image top in all panels.

Fig. 16. Schematic stratigraphy of “brain terrain” and mantle textures. The upper line is MOLA point topography. Lower lines delimit estimated subsurface boundaries between“brain terrain” and mantle. Brackets illustrate surface exposures of closed-cell “brain terrain” (CC-BT), open-cell “brain terrain” (OC-BT), high-center mantle polygons and low-center mantle polygons. MOLA is derived from a segment of orbit 12303 passing over HiRISE image PSP_002782_2230.

of craters relative to surrounding, lower-elevation, ablated surfacesof crater fill.

If an ice-free sublimation lag deposit is present on concen-tric crater fill surfaces (generated by a combination of aeolian

sediment deposition, sublimation of underlying debris-rich ice,and desert pavement formation), interactions between thermal-contraction surface fractures and overlying lag deposits couldresult in the observed complex morphology of “brain terrain”



Fig. 17. (a) Fresh crater in LDM with high-center mantle polygons. Portion ofPSP_002782_2230. Illumination is from the left. North to image top. (b) “Ring-mold”-like crater (Kress and Head, 2008) in closed-cell “brain terrain.” Portion ofPSP_002175_2210. Illumination is from the left. North to image top. (c) Elongatecrater in closed-cell “brain terrain.” If the crater was deformed by glacier-like flow,linear strain of ∼60% is documented. Portion of PSP_002175_2210. Illumination isfrom the left. North to image top.

(e.g., Marchant et al., 2002; Schorghofer and Aharonson, 2005;Schorghofer, 2007). Fractures would initially represent sites of de-pressed surface troughs (Fig. 20c) (Marchant et al., 2002). Thisis because enhanced contact with the dry atmosphere at poly-gon fractures would result in greater sublimation of subsurfaceice along polygon cracks (Marchant et al., 2002; Kowalewski etal., 2006, 2007), generating widened, deepened troughs analo-gous to those outlining terrestrial sublimation polygons (Marchantet al., 2002). Thermal fractures would accumulate sediment de-rived from winnowing of overlying lag deposits (e.g., Marchant etal., 2002) and infiltration of aeolian sediments, forming wedges

Fig. 18. Crater counts for fresh craters on mantle material (open circles) and allcraters on “brain terrain” surfaces (filled circles) using Hartmann (2005) isochrons.Counts combine data from four concentric crater fill surfaces in Utopia Planitia, andstrongly indicate a surface at of ∼1.5 My for mantle polygons. Brain terrain countsshow a roll-off at small crater sizes typical of lobate debris apron and lineated valleyfill surfaces. Craters in the 50–250 m diameter range suggest an age of ∼10–100 Myfor brain terrain.

Fig. 19. MOLA shot-point profile across a concentric crater fill-filled crater (presentin PSP_002175_2210). Inset. The MOLA shot-point profile reprojected to show theactual profile (near-horizontal line) and the modeled depth of the crater, assuminga parabolic profile, and depth-diameter ratios documented by Garvin et al. (2002).Up to ∼80% of the crater has been filled by concentric crater fill material.

analogous to sand-wedge structures common in cold and arid ter-restrial environments (Fig. 20d) (Péwé, 1959; Berg and Black, 1966;Marchant et al., 2002; Marchant and Head, 2007). Lateral trans-port of surface materials into deepening troughs would result inever-thicker accumulations of sediment relative to “stable” poly-gon interiors (e.g., Marchant et al., 2002).

Inversion of sublimation-polygon-like topography (polygonswith convex-up centers) could occur if sublimation continued toremove near-surface ice at polygon centers, but slowed at poly-gon troughs due to the presence of thickened accumulations ofsediment along polygon troughs (Fig. 20e) (e.g., Marchant et al.,2002). Cementation of wedge sediments at polygon margins byice deposited during periods of reversed vapor flux would en-hance the protection of ice located beneath the sediment wedges



Fig. 20. Schematic illustration of “brain terrain” development by fracturing, differ-ential sublimation, and topographic inversion. (a) Ice-rich deposits (white) are em-placed by atmospheric deposition. Sublimation of near-surface ice results in the for-mation of a sublimation lag deposit (light gray), consisting of dust entrained duringdeposition, wind-blown deposits, rock-fall, etc. (e.g., Marchant et al., 2002). (b) Thecombined action of annual thermal contraction cracking and glacio-tectonic frac-turing crack the ice-rich surface. Overlying fines winnow in, forming sand-wedges(dark gray). (c) Enhanced sublimation at fractures may result in lowering of troughsrelative to polygon interiors. (d) Continued fracture expansion, wedge infilling, andfracture-dominated sublimation further depresses troughs, and thicken wedge sed-iments. (e) Continuing sublimation results in a lowering of the depth of the icetable, possibly caused by obliquity-driven lowering of the depth of ice stability (e.g.,Mellon and Jakosky, 1995). Thickened sediment accumulations at wedges may pref-erentially preserve underlying ice. Original wedge structures may be preserved insome locations. (f) Where glacier-like flow occurs, polygons/fracture networks maybe deformed by flow. (g) Continued removal of subsurface ice, in places enhancedby the presence of thin, dark, dessicated mantle remnants (black), results in the col-lapse of closed-cell “brain terrain,” and the formation of open-cell “brain terrain.”

(Kowalewski, 2008). Such an inversion could result in the forma-tion of raised cells and chains of raised cells at the loci of relictpolygon troughs. Some raised cells may preserve the original, ax-ial surface trough (e.g., the axial furrows observed in some “brainterrain” cells, Figs. 6–8). These cells may remain ice-cored. This

process could account for the formation of closed-cell “brain ter-rain” contemporaneously with and continuing after, the period ofconcentric crater fill flow.

The development of open-cell “brain terrain” could be ac-counted for in this topographic inversion model by the contin-ued removal of ice from beneath closed-cell “brain terrain” cells.Continued removal of subsurface ice over time may explain thepresence of open-cell “brain terrain” in topographic lows betweenclosed-cell-patterned concentric ridges (e.g., Fig. 5). Additionally,where thin, dessicated remnants of low-albedo mantle materialare draped over “brain terrain” cells at the pinched-out marginsof the mantle (Fig. 15) we hypothesize that enhanced sublimationof residual ice beneath “brain terrain” cells (Williams et al., 2008)could further promote the collapse of closed-cell “brain terrain”cells, generating open-cell “brain terrain” cells (Fig. 20g).

This process for the formation and modification of “brain ter-rain” textures differs from that outlined by Mangold (2003) inthat it invokes both glacial stress orientation, as well as thermalcontraction cracking and sand-wedge development, to account forthe detailed microrelief of “brain terrain” textures. It differs fromthe Dobrea et al. (2007) model in that the preservation of broad-scale glacial flow features and of thermal contraction crack polygonwedges are preserved in the “brain terrain”—which would be lost ifcryoturbation or thermokarst formation (melting) had resulted inwidespread reworking of the concentric crater fill surface. Our pro-posed process is illustrated schematically in Fig. 20. Lastly, somefractures, particularly those located near the margins of the crater,may have formed more recently than the observed “brain terrain”textures. Such fractures (e.g., Fig. 13) are interpreted to representlate-stage mechanical failure of concentric crater fill glacial rem-nants.

4.2. Mantle origin

What is mantle material, and how might its deposition andmodification relate to “brain terrain”? In the analyzed images, thepresence of mantle material at the inner margins of crater walls,and in topographic lows between concentric crater fill ridges, sug-gests that mantle material could be an atmospherically emplaced,ice-rich deposit, that preferentially accumulates in shadowed ar-eas (Mustard et al., 2001; Hecht, 2002; Head et al., 2003). Surfaceages of ∼1.5 My suggest that mantle material in Utopia Plani-tia concentric crater fill craters is temporally associated with re-cent hemisphere-wide latitude-dependent mantle (LDM) deposi-tion events (Mustard et al., 2001; Head et al., 2003; Kreslavskyand Head, 2006): deposits which are thought to contain sufficientdusty material to generate a surficial lag deposit during sublima-tion of near-surface ice (Marchant et al., 2002).

4.3. Polygonalized mantle

Seasonal thermal contraction cracking generates contraction-crack polygons in ice-rich sediment on Earth and Mars (Washburn,1973; Mellon, 1997; Mangold, 2005; Marchant and Head, 2007).As in the formation of “brain terrain,” sublimation may initially belocally enhanced at polygon margins, generating high-center man-tle polygons, which may be analogous to terrestrial sublimationpolygons (Washburn, 1973; Mellon, 1997; Marchant et al., 2002;Kowalewski et al., 2006; Marchant and Head, 2007). The lackof strongly oriented mantle polygons suggests that young man-tle material has not flowed significantly (Zimbelman et al., 1989;Milliken et al., 2003), consistent with thicknesses of <40 m—although a preferred orientation may arise where mantle materialoverlies sloped crater walls, and where it occurs proximally towell-developed brain terrain. For the latter, contraction cracks may



Fig. 21. Examples of “brain terrain” and mantle stratigraphy in Eastern Hellas (a and b), and in Deuteronilus Mensae (c and d). (a) Lineated fill in “The Hourglass” (Head et al.,2005) crater on the eastern rim of the Hellas basin. White box indicates location of panel North to left. (b) Portion of PSP_002782_1405. (b) Closed-cell “brain terrain” (CC-BT);sinuous, open-cell “brain terrain” (OC-BT); and high-center mantle polygons (HC-MP) present in a similar arrangement to that observed in Utopia Planitia. (c) A lobate debrisapron in Deuteronilus Mensae. White box indicates location of panel (d). North to image top. Portion of P06_003246_2178. (d) Closed-cell “brain terrain” on lobate debrisapron ridges; open-cell “brain terrain” in topographic lows between ridges. Dunes partially obscure both surfaces. Portion of PSP_003246_2180.

be aligned with those of nearby or underlying brain terrain. Infill-ing of polygonal fractures with overlying lag deposit fines couldgenerate subsurface wedges, similar to the initial steps in the for-mation of “brain terrain.”

As with “brain terrain” discussed earlier, inversion of sublim-ation-polygon-like topography (polygons with convex-up centers)could occur if sublimation continued to remove near-surface iceat polygon centers, but slowed at polygon troughs due to thepresence of thickened accumulations of sediment along polygontroughs (e.g., Marchant et al., 2002). This differential sublimationprocess could also account for the formation of low-center man-tle polygons with relatively flat, low-lying interiors, and elevatedmargins. This shared inversion process accounts for the similarityin morphology between low-center mantle polygons and open-cell “brain terrain” cells, although open-cell “brain terrain” cellsare generally larger than mantle polygons, and form in gradationalcontacts with closed-cell “brain terrain,” rather than in mantle ma-terial.

Across the martian mid-to-high latitudes, low-center polygonsare exceptionally uncommon in HiRISE images (Levy et al., 2009),suggesting that unique conditions may exist at the margins ofmantle surfaces in concentric crater fill terrains. The concentra-tion of low-center polygons within scalloped depressions in themantle (that may have a sublimation origin) (Kadish et al., 2008;Zanetti et al., 2008; Lefort et al., 2009), and in regions where themantle thins and pinches out, suggests that enhanced removal ofsubsurface ice by locally intense sublimation may be critical forthe formation of low-center mantle polygons.

4.4. Climate implications: Local and global

What do the proposed mechanisms for “brain terrain” andmantle formation and modification suggest about climate condi-tions in Utopia Planitia at ∼1.5 My and ∼10–100 My timescales?Debris-covered-glacier-like landforms have been documented ex-tensively in martian mid-latitudes, and have been interpreted tobe geomorphic evidence of cold and arid climate conditions, domi-nated by sublimation of ground ice during periods of low obliq-uity, and dramatic redistribution and deposition of ice on thesurface during periods of high obliquity (Mustard et al., 2001;Kreslavsky and Head, 2002, 2006; Laskar et al., 2002, 2004; Headet al., 2003). Glacier-like deformation of thick accumulations ofice-rich material during high-obliquity (∼35–45◦) (Madeleine etal., 2007) redistribution/depositional periods during the past ∼10–100 My may account for the emplacement of concentric crater fill,while more subtle obliquity excursions over the past ∼1–2 My (upto ∼35◦) (Head et al., 2003) may account for the emplacement ofthinner, static mantle material. Near-surface modification of both“brain terrain” and mantle is consistent with cold and arid cli-mate conditions and sublimation-driven processes. These lines ofevidence suggest an Amazonian hydrological cycle dominated bysolid-vapor transitions rather than by widespread melting and flowof liquid water.

Recent work (e.g., Soare et al., 2007, 2008) has concluded thatwidespread melting and thermokarst erosion has been instrumen-tal in the formation of some Utopia Planitia surfaces (e.g., scallopedterrain present in mantle material). What do mantle polygons and



“brain terrain” indicate about the presence and duration of satu-rated active layers in Utopia Planitia during the recent Amazonian?From terrestrial experience, the formation of low-center polygonson Earth is most dramatic in ice-wedge polygons, which requireseasonal input and freezing of liquid water (Washburn, 1973;Root, 1975; Marchant and Head, 2007). However, raised shoul-ders also form on sand-wedge polygons that develop in cold andarid climates in which liquid water is not available in significantquantities (Péwé, 1959; Berg and Black, 1966; Washburn, 1973;Root, 1975; Marchant and Head, 2007). If ephemeral liquid waterwas present during the development of low-center mantle poly-gons, driving the formation of pronounced polygon-margin shoul-ders, then its spatial extent was limited to concentrated occur-rences at the tapering margins of the mantle, and to both thefloors and steep slopes within scalloped depressions. This rangeof locations in which low-center mantle polygons are observed isinconsistent with simple ponding and saturation of sediments. Fur-ther, exposures of pristine “brain terrain” in locations where themantle has been completely removed from the surface stronglysuggest a cold and dry (e.g., sublimation) mechanism for the re-moval of mantle material to form windows down to the “brainterrain;” had these pits been water-saturated to form ice-wedgepolygons, it is likely that the underlying “brain terrain” would havebeen reworked and extensively modified. Lastly, the preservationof original “brain terrain” cell axial furrows suggests that subse-quent, widespread cryoturbation has not disrupted these fine-scalesurface features. Rather, we suggest that the morphology of “brainterrain” and mantle material can be accounted for by atmosphericdeposition of ice during periods of high obliquity, and modifica-tion of ice-rich units by a suite of cold-desert processes includingglacial deformation, thermal contraction cracking, and differentialsublimation in the absence of abundant near-surface liquid water.

Finally, these analyses provide a framework for understandingrelationships between “brain terrain” textures present on lobatedebris aprons and lineated valley fill elsewhere on Mars and thelatitude-dependent mantle (LDM). Contacts between “brain ter-rain” and LDM occur extensively at martian midlatitudes, includ-ing transects in eastern Hellas, Deuteronilus Mensae and NilosyrtisMensae (Fig. 21). The wide distribution of contacts between “brainterrain” and LDM material suggests that the transition betweenmid-Amazonian glacial periods, and more recent Amazonian, dry“periglacial” conditions are not restricted to local occurrences andmay indicate global climate processes.

5. Summary and conclusions

“Brain terrain” and polygonally-patterned latitude-dependentmantle (LDM) surfaces were analyzed in Utopia Planitia. Strati-graphic relationships and crater-count ages show that the mantlesurface postdates concentric crater fill “brain terrain” by ∼10–100 My. Two unique surface textures were identified in eachunit, respectively, closed-cell “brain terrain” and open-cell “brainterrain,” and high-center mantle polygons and low-center man-tle polygons. Lateral contacts between textures are shown to begradational, suggesting modification of “brain terrain” and mantlematerial into the current range of surface morphologies. A com-bined glacial-stress/thermal-contraction and differential sublima-tion mechanism is proposed for the formation and modification of“brain terrain” on concentric crater fill. A similar model, driven bythermal contraction cracking and differential sublimation of under-lying ice, but without glacier-like ice flow, is also proposed for theformation and development of the polygonally patterned mantle.This explanation for examined “brain terrain” and polygonalizedmantle material requires two different styles of atmospheric depo-sition of ice: an older, “glacial” deposition period for the formationof concentric crater fill “brain terrain,” and a more recent “ice

age” mantling period for the deposition of LDM. Both depositionstyles result in the formation of near-surface excess ice (ice vol-ume exceeding available pore space). Glacier-like concentric craterfill with closed-cell “brain terrain” is interpreted to have formedduring a mid-latitude peak-glaciation period that ended at ∼10–100 My. This was followed by quiescent cold desert conditions,preceding the most recent “ice age” (Head et al., 2003) period at∼1–2 My, responsible for the deposition of mid-high latitude de-pendent mantle material, the formation of mantle polygons, andthe modification of closed-cell “brain terrain” present at mantlemargins into open-cell “brain terrain.” Understanding the devel-opment of “brain terrain” on Utopia Planitia concentric crater fillprovides insight into analysis of “brain terrain” present on lobatedebris aprons and lineated valley fill across the martian midlati-tudes, suggesting a dynamic history of ice redistribution under colddesert conditions during the late Amazonian.

Acknowledgments

We gratefully acknowledge support from the NASA Mars DataAnalysis Program (NNX07AN95G to J.W.H.), the Mars FundamentalResearch Program (NNX06AE32G to D.R.M. and J.W.H.), the AppliedInformation Systems Research Program (NNG05GA61G to J.W.H.),and the Mars Express HRSC Participating Scientist Program (JPL1237163 to J.W.H.), and the National Science Foundation, Office ofPolar Programs (through grant ANT-0338291 to D.R.M. and J.W.H.).Special thanks to Caleb Fassett for assistance in processing and in-terpreting crater count information and for image processing; toJames Dickson for coordination of image processing and anaglyphproduction; and to Dr. Matt Balme and one anonymous reviewerfor their efforts in improving the manuscript.

References

Benn, D.I., Kirkbride, M.P., Owen, L.A., Brazier, V., 2003. Glaciated valley landsystems.In: Evans, D.J.A. (Ed.), Glacial Landsystems. Edward Arnold, London, UK, pp. 372–406.

Berg, T.E., Black, R.F., 1966. Preliminary measurements of growth of nonsorted poly-gons, Victoria Land, Antarctica. In: Tedrow, J.C.F. (Ed.), Antarctic Soils and Soil-Forming Processes. American Geophysical Union, Washington, DC, pp. 61–108.

Dobrea, E.Z.N., Asphaug, E., Grant, J.A., Kessler, M.A., Mellon, M. T., 2007. Patternedground as an alternative explanation for the formation of brain coral texturesin the mid latitudes of Mars: HiRISE observations of lineated valley fill textures.In: Proc. 7th International Conference on Mars. Abstract #3358.

Fastook, J.L., Head, J.W., Marchant, D.R., 2008. Dichotomy boundary glaciation mod-els: Implications for timing and glacial processes. Lunar Planet. Sci. 39. Abstract#1109.

Forget, F., Haberle, R.M., Montmessin, F., Levrard, B., Head, J.W., 2006. Formationof glaciers on Mars by atmospheric precipitation at high obliquity. Science 311,368–371.

Garvin, J.B., Sakimoto, S.E.H., Frawley, J.J., Schnetzler, C., 2002. Global geometricproperties of martian impact craters. Lunar Planet. Sci. 33. Abstract #1255.

Hartmann, W.K., 2005. Martian cratering 8: Isochron refinement and the chronologyof Mars. Icarus 174, 294–320.

Head, J.W., Marchant, D.R., 2006. Modification of the walls of a Noachian craterin northern Arabia Terra (24E, 39N) during northern mid-latitude Amazonianglacial epochs on Mars: Nature and evolution of lobate debris aprons and theirrelationships to lineated valley fill and glacial systems. Lunar Planet. Sci. 37. Ab-stract #1126.

Head, J.W., Marchant, D.R., 2009. Inventory of ice-related deposits on Mars: Evidencefor burial and long-term sequestration of ice in non-polar regions and implica-tions for the water budget and climate evolution. Lunar Planet. Sci. 40. Abstract#1356.

Head, J.W., Mustard, J.F., Kreslavsky, M.A., Milliken, R.E., Marchant, D.R., 2003. Recentice ages on Mars. Nature 426, 797–802.

Head, J.W., and 13 colleagues, 2005. Tropical to mid-latitude snow and ice accumu-lation, flow and glaciation on Mars. Nature 434, 346–351.

Head, J.W., Marchant, D.R., Agnew, M.C., Fassett, C.I., Kreslavsky, M.A., 2006a. Ex-tensive valley glacier deposits in the northern mid-latitudes of Mars: Evidencefor late Amazonian obliquity-driven climate change. Earth Planet. Sci. Lett. 241,663–671.

Head, J.W., Nahm, A.L., Marchant, D.R., Neukum, G., 2006b. Modification of thedichotomy boundary on Mars by Amazonian mid-latitude regional glaciation.Geophys. Res. Lett. 33, doi:10.1029/2005GL024360.



Head, J.W., Marchant, D.R., Kreslavsky, M.A., 2008. Formation of gullies on Mars:Link to recent climate history and insolation microenvironments implicate sur-face water flow origin. Proc. Natl. Acad. Sci. 105, 13,258–13,263.

Hecht, M.H., 2002. Metastability of water on Mars. Icarus 156, 373–386.Holt, J.W., and 10 colleagues, 2008. Radar sounding evidence for ice within lobate

debris aprons near Hellas Basin, mid-southern latitudes of Mars. Lunar Planet.Sci. 39. Abstract #2441.

Kadish, S.J., Barlow, N.G., 2006. Pedestal crater distribution and implications for anew model of formation. Lunar Planet. Sci. 37. Abstract #1254.

Kadish, S.J., Head, J.W., Barlow, N.G., 2008. Pedestal craters on Mars: Distribution,characteristics, and implications for Amazonian climate change. Lunar Planet.Sci. 39. Abstract #1766.

Kostama, V.-P., Kreslavsky, M.A., Head, J.W., 2006. Recent high-latitude icy mantle inthe northern plains of Mars: Characteristics and ages of emplacement. Geophys.Res. Lett. 33, doi:10.1029/2006GL025946.

Kowalewski, D.E., 2008. Vapor diffusion through sublimation till: Implications forpreservation of ancient glacier ice in the McMurdo dry valleys, Antarctica.Boston University, Boston.

Kowalewski, D.E., Marchant, D.R., Levy, J.S., Head, J.W., 2006. Quantifying summer-time sublimation rates for buried glacier ice in Beacon valley, Antarctica. Antarc-tic Sci. 18, 421–428.

Kowalewski, D.E., Marchant, D.R., Head, J.W., Levy, J.S., 2007. Modeling vapor dif-fusion in sublimation tills of the Antarctic dry valleys: Implications for thepreservation of near-surface ice on Mars. Lunar Planet. Sci. 38. Abstract #2143.

Kreslavsky, M.A., Head, J.W., 2002. Mars: Nature and evolution of young,latitude-dependent water-ice-rich mantle. Geophys. Res. Lett. 29, doi:10.1029/2002GL015392.

Kreslavsky, M.A., Head, J.W., 2006. Modification of impact craters in the northernplains of Mars: Implications for the Amazonian climate history. Meteorit. Planet.Sci. 41, 1633–1646.

Kreslavsky, M.A., Head, J.W., Marchant, D.R., 2008. Periods of active permafrostlayer formation during the geological history of Mars: Implications for circum-polar and mid-latitude surface processes. Planet. Space Sci. 56, doi:10.1016/j.pss.2006.1002.1010.

Kress, A.M., Head, J.W., 2008. Ring-Mold Craters on Lineated Valley Fill (LVF) andLobate Debris Aprons (LDA) on Mars (I): Evidence for the presence of subsurfaceice. Lunar Planet. Sci. 39. Abstract #1273.

Kress, A.M., Head, J.W., 2009. Ring-mold craters on lineated valley fill, lobate debrisaprons, and concentric crater fill on Mars: Implications for near-surface struc-ture, composition, and age. Lunar Planet. Sci. 40. Abstract #1379.

Kuzmin, R.O., 2005. Ground ice in the martian regolith. In: Tokano, T. (Ed.), Wateron Mars and Life. Springer-Verlag, Berlin, pp. 155–189.

Laskar, J., Levrard, B., Mustard, J.F., 2002. Orbital forcing of the martian polar layereddeposits. Nature 419, 375–377.

Laskar, J., Correia, A.C.M., Gastineau, M., Joutel, F., Levrard, B., Robutel, P., 2004.Long term evolution and chaotic diffusion of the insolation quantities of Mars.Icarus 170, 343–364.

Lefort, A., Russel, P.S., Thomas, N., McEwen, A.S., Dundas, C.M., Kirk, R.L., 2009.HiRISE observations of periglacial landforms in Utopia Planitia. J. Geophys.Res. 114, doi:10.1029/2008JE003264.

Levy, J.S., Marchant, D.R., Head, J.W., 2006. Distribution and origin of patternedground on Mullins valley debris-covered glacier, Antarctica: The roles of ice flowand sublimation. Antarctic Sci. 18, 385–397.

Levy, J.S., Head, J.W., Marchant, D.R., 2007. Lineated valley fill and lobate debrisapron stratigraphy in Nilosyrtis Mensae, Mars: Evidence for phases of glacialmodification of the dichotomy boundary. J. Geophys. Res. 112, doi:10.1029/2006JE002852.

Levy, J.S., Head, J.W., Marchant, D.R., 2009. Thermal contraction crack polygons onMars: Classification, distribution, and climate implications from HiRISE observa-tions. J. Geophys. Res. 114, doi:10.1029/2008JE003273.

Lucchitta, B., 1984. Ice and debris in the fretted terrain, Mars. J. Geophys. Res. 89,B409–B418.

Mackay, J.R., 1990. Some observations on the growth and deformation of epigenetic,syngenetic and antisyngenetic ice wedges. Permafrost Periglacial Process. 1, 15–29.

Madeleine, J.-B., Forget, F., Head, J.W., Levrard, B., Montmessin, F., 2007. Exploringthe northern mid-latitude glaciation with a general circulation model. In: Proc.Seventh International Conference on Mars, Pasadena, CA, July 9–13. Abstract#3096.

Mangold, N., 2003. Geomorphic analysis of lobate debris aprons on Mars at MarsOrbiter Camera scale: Evidence for ice sublimation initiated by fractures. J. Geo-phys. Res. 108, 8021–8033.

Mangold, N., 2005. High latitude patterned grounds on Mars: Classification, distri-bution and climatic control. Icarus 174, 336–359.

Mangold, N., Allemand, P., 2001. Topographic analysis of features related to ice onMars. Geophys. Res. Lett. 28, 407–410.

Marchant, D.R., Head, J.W., 2007. Antarctic dry valleys: Microclimate zonation, vari-able geomorphic processes, and implications for assessing climate change onMars. Icarus 192, 187–222.

Marchant, D.R., Lewis, A.R., Phillips, W.M., Moore, E.J., Souchez, R.A., Denton, G.H.,Sugden, D.E., Potter Jr., N., Landis, G.P., 2002. Formation of patterned groundand sublimation till over Miocene glacier ice in Beacon valley, southern Victorialand, Antarctica. Geol. Soc. Am. Bull. 114, 718–730.

Mellon, M.T., 1997. Small-scale polygonal features on Mars: Seasonal thermal con-traction cracks in permafrost. J. Geophys. Res. 102, 25,617-625,628.

Mellon, M.T., Jakosky, B.M., 1995. The distribution and behavior of martian groundice during past and recent epochs. J. Geophys. Res. 100, 11,781–711,799.

Mellon, M.T., Boynton, W.V., Feldman, W.C., Arvidson, R.E., Titus, T.N., Bandfield,J.L., Putzig, N.E., Sizemore, H.G., 2008. A prelanding assessment of the ice ta-ble depth and ground ice characteristics in martian permafrost at the Phoenixlanding site. J. Geophys. Res. 113 (E12), doi:10.1029/2007JE003067.

Milliken, R.E., Mustard, J.F., Goldsby, D.L., 2003. Viscous flow features on the surfaceof Mars: Observations from high-resolution Mars Orbiter Camera (MOC) images.J. Geophys. Res. 108 (E6), doi:10.1029/2002JE002005.

Mustard, J.F., Cooper, C.D., Rifkin, M.K., 2001. Evidence for recent climate change onMars from the identification of youthful near-surface ground ice. Nature 412,411–414.

Neukum, G., Ivanov, B.A., 2001. Crater production functions for Mars. Lunar Planet.Sci. 32. Abstract #1757.

Péwé, T.L., 1959. Sand-wedge polygons (tessellations) in the McMurdo Sound region,Antarctica—A progress report. Am. J. Sci. 257, 545–552.

Plaut, J.J., Safaeinili, A., Holt, J.W., Phillips, R.J., Campbell, B.A., Carter, L.M., Leuschen,C., Gim, Y., Seu, R., the Sharad Team, 2008. Radar evidence for ice in lobate de-bris aprons in the mid-northern latitudes of Mars. Lunar Planet. Sci. 39. Abstract#2290.

Root, J.D., 1975. Ice-wedge polygons, Tuktoyaktuk area, north–west territories. Geo-logical Survey of Canada, Canada, p. 181.

Russell, P.S., Head, J.W., Hecht, M.T., 2004. Evolution of ice deposits in the local envi-ronment of martian circum-polar craters and implications for polar cap history.Lunar Planet. Sci. 35. Abstract #2007.

Schon, S.C., Head, J.W., Milliken, R.E., 2008. Layered morphology of the latitude-dependent mantle: A potential late-Amazonian paleoclimate signal. LunarPlanet. Sci. 39. Abstract #1873.

Schorghofer, N., 2007. Dynamics of ice ages on Mars. Nature 449, 192–194.Schorghofer, N., Aharonson, O., 2005. Stability and exchange of subsurface ice on

Mars. J. Geophys. Res. 110, doi:10.1029/2004JE002350.Shean, D.E., Head, J.W., Marchant, D.R., 2007. Shallow seismic surveys and ice thick-

ness estimates of the Mullins valley debris-covered glacier, McMurdo dry val-leys, Antarctica. Antarctic Sci. 19, 485–496, doi:10.1017/S0954102007000624.

Soare, R.J., Kargel, J.S., Osinksi, G.R., Costard, F., 2007. Thermokarst processes and theorigin of crater-rim gullies in Utopia and western Elysium Planitia. Icarus 191,95–112.

Soare, R.J., Osinksi, G.R., Roehm, C.L., 2008. Thermokarst lakes and ponds on Marsin the very recent (late Amazonian) past. Earth Planet. Sci. Lett. 272 (1–2), 382–393.

Squyres, S.W., 1979. The distribution of lobate debris aprons and similar flows onMars. J. Geophys. Res. 84, 8087–8096.

Squyres, S.W., Carr, M.H., 1986. Geomorphic evidence for the distribution of groundice on Mars. Science 231, 249–252.

Washburn, A.L., 1973. Periglacial Processes and Environments. St. Martin’s Press,New York, pp. 1–2, 100–147.

Williams, K.E., Toon, O.B., Heldmann, J.L., McKay, C., Mellon, M.T., 2008. Stability ofmid-latitude snowpacks on Mars. Icarus 196, 565–577.

Zanetti, M., Hiesinger, H., Reiss, D., Hauber, E., Neukum, G., 2008. Scalloped de-pressions in Malea Planum, southern Hellas Basin, Mars. Lunar Planet. Sci. 39.Abstract #1682.

Zimbelman, J.R., Clifford, S.M., Williams, S.H., 1989. Concentric crater fill on Mars:An aeolian alternative to ice-rich mass wasting. Lunar Planet. Sci. 19, 397–407.

Documents

Author's personal copy - Brown University · Author's personal copy Concentric crater ll: Glacial and periglacial change 465 Fig. 4. An elevated region (center) surfaced by closed-cell