Icarus 219 (2012) Contents lists available at SciVerse ScienceDirect. Icarus. journal homepage:

Transcription

1 Icarus 219 (2012) Contents lists available at SciVerse ScienceDirect Icarus journal homepage: Patterns of accumulation and flow of ice in the mid-latitudes of Mars during the Amazonian James L. Dickson a, James W. Head a,, Caleb I. Fassett a,b a Department of Geological Sciences, Brown University, Providence, RI 02912, USA b Department of Astronomy, Mount Holyoke College, South Hadley, MA 01075, USA article info abstract Article history: Received 28 December 2011 Revised 2 March 2012 Accepted 9 March 2012 Available online 29 March 2012 Keywords: Mars Mars, Climate Mars, Surface Mars, Atmosphere Ices Evidence has accumulated that non-polar portions of Mars have undergone significant periods of glaciation during the Amazonian Period. This evidence includes tropical mountain glacial deposits, lobate debris aprons, lineated valley fill, concentric crater fill, pedestal craters, and related landforms, some of which suggest that ice thicknesses exceeded a kilometer in many places. In some places, several lines of evidence suggest that ice is still preserved today in the form of relict debris-coved glaciers. The vast majority of deposit morphologies are analogous to those seen in cold-based glacial deposits on Earth, suggesting that little melting has taken place. Although these features have been broadly recognized, and their modes of ice accumulation and flow analyzed at several scales, they have not been analyzed and wellcharacterized globally despite their significance for understanding the evolution of the martian climate. A major outstanding question is the global extent of accumulation and flow of ice during periods of nonpolar glaciation: As a mechanism to address this question, we outline two end-member scenarios to provide a framework for further discussion and analysis: (1) ice accumulation was mainly focused within individual craters and valleys and flow was largely local to regional in scale, and (2) ice accumulation was dominated by global latitudinal scale cold-based ice sheets, similar in scale to the Laurentide continental ice sheets on Earth. In order to assess these end members, we conducted a survey of ice-related features seen in Context Camera (CTX) images in each hemisphere and mapped evidence for flow directions within well-preserved craters in an effort to decipher orientation preferences that could help distinguish between these two hypotheses: regional/hemispheric glaciation or local accumulation and flow. These new crater data reveal a latitudinal-dependence on flow direction: at low latitudes in each hemisphere (<40 45 ) cold, pole-facing slopes are strongly preferred sites for ice accumulation, while at higher latitudes (>40 45 ), slopes of all orientations show signs of ice accumulation and ice-related flow. This latitudinal onset of concentric flow of ice within craters in each hemisphere correlates directly with the lowest latitudes at which typical pedestal craters have been mapped. Taken together, these observations demarcate an important latitudinal boundary that partitions each hemisphere into two zones: (1) poleward of 45, where net accumulation of ice is interpreted to have occurred on all surfaces, and (2) equatorward of 45, where net accumulation of ice occurred predominantly on pole-facing slopes. These results provide important constraints for deciphering the climatic conditions that characterized Mars during periods of extensive Amazonian non-polar glaciation. Ó 2012 Elsevier Inc. All rights reserved. 1. Introduction Despite desert conditions across the entire planet, Mars has undergone substantial ice accumulation and flow (glaciation) in many non-polar locations during the Amazonian Period (Head and Marchant, 2009). Global circulation models suggest that Mars cyclically transports ice from polar regions to lower latitudes and back, forced primarily by oscillations of the planet s obliquity (Mischna et al., 2003; Levrard et al., 2004; Forget et al., 2006; Corresponding author. address: james_head@brown.edu (J.W. Head). Madeleine et al., 2009). The current obliquity of Mars (25.2 ) is low compared to the last 20 Myr (Laskar et al., 2004), implying that we may be currently in an interglacial period where ice accumulation is concentrated at the poles (Head et al., 2003). Orbital variations from before 20 Myr are not as well known (Laskar et al., 2004), but several lines of evidence suggest more massive and expansive glaciation in non-polar latitudes during previous epochs. For example, the fan-shaped glacial deposits on the flanks of the major Tharsis volcanoes, all found within the present-day tropics, are at the same scale as recent terrestrial ice sheets (Head and Marchant, 2003; Shean et al., 2005; Milkovich et al., 2006; Kadish et al., 2008a). Evidence for mid-latitude /$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.

2 724 J.L. Dickson et al. / Icarus 219 (2012) glaciation includes lobate debris-aprons and lineated valley fill (Squyres, 1979; Lucchitta, 1984; Pierce and Crown, 2003); recent studies have shown that many of these deposits are products of ice accumulation and flow in the form of debris-covered glaciers (e.g., Head et al., 2006a,b, 2010; Kreslavsky and Head, 2006; Levy et al., 2007, 2009, 2010; Morgan et al., 2009; Baker et al., 2010), rather than vapor diffusion of ice into talus aprons forming ice-lubricated debris flows as rock glaciers. Multiple sites reveal evidence for glacial highstands, arguing for past ice thicknesses of at least 1km (Dickson et al., 2008, 2010). Much of this ice is apparently still present, as radar data show stagnant, clean glacial ice in the mid-latitudes of Mars beneath a relatively thin debris cover (Holt et al., 2008; Plaut et al., 2009), akin to debris-covered cold-based glaciers observed in the McMurdo Dry Valleys of Antarctica (Marchant and Head, 2007). Ice deposits of unknown thickness are also found beneath a thin debris cover at high latitudes (Boynton et al., 2002; Feldman et al., 2002; Smith et al., 2009) and in the mid-latitudes (Byrne et al., 2009), perhaps related to the latitude-dependent ice mantle interpreted to have formed in recent history (e.g., Head et al., 2003). Pedestal craters (McCauley, 1973; Arvidson et al., 1979) provide evidence for the past existence of thicker ice-rich units that have largely been removed at higher latitudes (Schultz, 1988; Schultz and Lutz, 1988; Barlow, 2005; Wrobel et al., 2006; Kadish et al., 2008b, 2009, 2010; Kadish and Head, 2011). The distribution of specific types of ice-related features on Mars varies as a function of latitude (Head and Marchant, 2009). At low latitudes (<25 ), ice-related deposits are primarily observed on the northwestern flanks of the Tharsis Montes and Olympus Mons (Lucchitta, 1981; Head and Marchant, 2003; Basilevsky et al., 2005; Shean et al., 2005, 2007; Milkovich et al., 2006; Kadish et al., 2008a) and in a few unusual locations in Sinus Sabaeus (Shean, 2010). In the mid-latitudes (25 60 ), however, ice-related morphologies are ubiquitous (Head and Marchant, 2009). Evidence for ice-related activity within the last 10 Myr in these latitudes is found in the form of (1) viscous flow features (VFF) (Milliken et al., 2003), (2) mantling units interpreted to be a desiccating layer of icy soil (Mustard et al., 2001; Schon et al., 2009) and (3) young gullies (Malin and Edgett, 2000). The distribution pattern for these features broadly agrees with where GCMs predict accumulation over the last several million years (Mischna et al., 2003). Another suite of features in the mid-latitudes suggests more substantial ice-related activity within the last several hundred million years. Historically, these features have been subdivided into three separate classes (Squyres, 1978, 1979): Lobate Debris Aprons (LDA) (Pierce and Crown, 2003; Mangold, 2003; Li et al., 2005; Chuang and Crown, 2005; Chuang et al., 2009), Lineated Valley Fill (LVF) (Head et al., 2006a,b; Baker et al., 2010) and Concentric Crater Fill (CCF) (Levy et al., 2010). This classification reflects the topographic settings within which these features are found in addition to the specific morphologies of the features themselves: LDAs show evidence of flowing from the base of scarps across smooth plains, LVF flow patterns are found within constrained valleys, and CCF is confined to the interior of impact craters. Radar measurements of LDA and LVF support the interpretation that many of these features are nearly pure ice (Holt et al., 2008; Plaut et al., 2009) and the common proximity of LVF and LDA suggest Fig. 1. Type examples of flow orientation in the northern hemisphere. (A) Concentric flow of ice within a crater at 46.2 N (mosaic of HRSC orbit h0038_0000, CTX orbit P17_007740_1357, and HRSC orbit h0300_0000). (B) Interpretation of flow direction, based upon mapping of concentric ridges on the floor of the crater. Measurement of orientation was made at the point on the crater wall where flow appears to commence. (C) Ice-related flow features flowing from south to north (towards the pole) at 31.9 N (mosaic of CTX orbits B02_010484_2122 and B02_010273_2121). (D) Interpretation of flow direction.

3 J.L. Dickson et al. / Icarus 219 (2012) that their patterns of flow may at one time been more integrated (Head and Marchant, 2009; Head et al., 2010; Dickson et al., 2008, 2010). As discussed below, new data show that concentric may not be the best descriptor for some of the ice-related features confined to the interior of craters at all latitudes, because of evidence for preferential directional flow. The broad distribution of ice-related features across the northern and southern mid-latitudes raises questions about the nature of periods of regional non-polar deposition of ice (ice ages) during the Amazonian on Mars. Fundamental among these questions is the scope and style of flow during periods when the rate of ice accumulation in the mid-latitudes exceeded the rate of ablation via sublimation (e.g. Levrard et al., 2004; Madeleine et al., 2009) or melting (Dickson et al., 2009; Fassett et al., 2010). For example, did Mars ever host continental-scale ice sheets in either hemisphere? Did the thickness of ice ever reach high enough values so that flow occurred regionally? Or was the accumulation and flow of ice dominated by ice accumulation differences related to local differences in accumulation (e.g., plateaus, scarps, crater rims and interiors)? Deciphering where Mars might fall on this spectrum has implications for the amount of ice deposited at the surface and will help clarify the evolution of the climate. An important signal that we can measure to test these endmembers is the orientation of flow. Orientation measurements have provided insight regarding the distribution and properties of other potentially ice-related features, like gullies (Malin and Edgett, 2000; Milliken et al., 2003; Heldmann and Mellon, 2004; Bridges and Lackner, 2006; Balme et al., 2006; Dickson et al., 2007; Heldmann et al., 2007). Patterns detected through the measurement of flow direction could potentially lead to the mapping of either (a) regional/hemispheric sources that represent the past existence of continental-scale ice sheets, or (b) localized sources that reflect smaller regions of enhanced accumulation of ice. Images obtained from the Context Camera (CTX) on the Mars Reconnaissance Orbiter (MRO) allows us to make this measurement. 2. Methods Ice-related features on Mars, particularly LDA, LVF and CCF, are most common poleward of 30 in each hemisphere (Head and Marchant, 2009). At latitudes greater than 55 in the north and south, detailed surface morphology is obscured by a blanket of late Amazonian mantling material that inhibits recognition of past flow features (Kreslavsky and Head, 2000; Mustard et al., 2001; Head et al., 2003). Therefore, we surveyed Mars Reconnaissance Orbiter Context Camera (CTX) images through mission phase B09 (May, 2009) to latitudes between 20 and 60 in each hemisphere (10,261 total images). Images that contained evidence for ice-related flow were added to an ArcGIS database to ensure spatial accuracy and prevent redundant measurements of features imaged multiple times. Uneven imaging of slopes of different orientation would introduce bias into our results, so we restricted our measurements of ice-related features to those found within impact craters that provide 360 coverage of orientations within the crater. In order to focus on the flow patterns within craters, craters were only Fig. 2. Type examples of flow orientation in the southern hemisphere. (A) Pole-facing flow of debris-covered ice in an 18 km impact crater at 31.8 S (subframe of CTX orbit P06_003316_1479). Almost all flow directions measured in this latitude band (30 35 S) are pole-facing (89.3%), consistent with flow features in the same latitude band in the northern hemisphere (89.1%). (B) Interpretation of flow directions within the crater. (C) Concentric flow of ice in the southern mid-latitudes (46.2 ) (subframe of CTX orbit P19_008308_2265). Flow at this latitude in each hemisphere is generally concentric, with only a slight preference for pole-facing orientations (58.5% in the south, 57.4% in the north between 45 and 50 latitude). (D) Interpretation of flow direction.

4 726 J.L. Dickson et al. / Icarus 219 (2012) considered if no portion of the rim had been breached or eroded to the level of the adjacent unit. Ice-related flow features found within craters with rims that are breached at any point were segregated into a separate category and assessed for potential infilling of ice/ debris from a regional ice field. A conformal projection was necessary to preserve angles at all latitudes, so all mapping of flow orientation was conducted in a Mercator map projection. Previous studies have determined the necessary requirements for mapping ice-related flow on Mars, so we utilized the criteria listed by Head et al. (2010). Among the 14 criteria that they defined, several provide information not simply about the presence of ice, but the orientation of flow. These include: (1) parallel arcuate ridges trending away from the host crater wall (Fig. 1c); (2) constriction of flow between obstacles; (3) broadening of arcuate ridges on unobstructed plains; (4) folding and deformation of broad lobes into smaller lobes; and (5) integration of multiple smaller lobes into larger lobes (Fig. 2a). Rare instances where flow was apparent but the direction of flow was ambiguous were not included in the orientation measurements. A line indicating the direction of flow was mapped for each individual lobe that we observed. Hence, a suite of smaller individual flows would be statistically amplified in our study, relative to a broader lobe of material. Each line indicative of flow was traced back towards the host crater wall, where the orientation measurement was made. Thus, depending on the original source of the ice, we are either measuring (1) where ice accumulated on the wall and the direction of localized flow, or (2) the direction from which ice overtopped the crater from the surrounding plains, with the steep crater rim providing a source for rock-fall and debris-cover not available on the adjacent plains. Considering that present topography on ice-related flow features is not necessarily indicative of topography at the time of maximum ice-volume (Dickson et al., 2008, 2010), only surface morphology was used to evaluate flow direction. 3. Results Our survey provides the most complete map of the distribution of ice-related landforms yet constructed for the mid-latitudes of Mars. Fig. 3a shows the mid-latitude distribution of all CTX orbits Fig. 3. (A) The distribution of ice-related flow features (LDA, LVF, CCF) in the mid-latitudes of Mars (20 60 in each hemisphere). Each dot represents the center coordinates of a CTX frame that contains any features known to be the product of ice-related flow. A footprint map of all CTX images included in the survey (through mission phase B09) is rendered in white to show where targeting heterogeneities could potentially influence data accumulation and interpretations. Ice-related flow is abundant in each hemisphere poleward of 30, particularly where steep topography occurs. Map projection is Robinson. (B) The distribution of flow features confined to the interior of impact craters with fully intact rims in the mid-latitudes of Mars. Each dot represents a mapped feature from our survey. The distribution of flow features within impact craters is identical to that of ice-related flow features as a whole in the mid-latitudes, suggesting that there is nothing intrinsic to the impact process itself that enhances or inhibits glaciation on Mars. Craters that show evidence for flow out of (yellow) or into (green) are displayed here but were not used for orientation calculations, as they did not contain intact rims. Map projection is Robinson; measurements were made in Mercator projection. Background map is MOLA color topography over a global MOLA hillshade map. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

5 J.L. Dickson et al. / Icarus 219 (2012) Fig. 4. The orientation of flow directions in the mid-latitudes of Mars (25 60 ). Orientation of flow direction was measured for features within impact craters along the crater wall from which the ice was sourced. Flow is almost exclusively pole-facing at lower mid-latitudes (<40 ), where such features are also most abundant. Between 40 and 50 in each hemisphere, a greater number of equator-facing flow features are observed; craters at these latitudes and higher exhibit the most well-developed concentric crater fill.

6 728 J.L. Dickson et al. / Icarus 219 (2012) Table 1 The orientation of ice-related flow features as a function of latitude. In each hemisphere, flow direction is dominantly pole-facing equatorward of 45. Poleward of 45, flow is observed on more equator-facing slopes, yielding more concentric flow. Latitude band Ice-related flow features mapped Pole-facing (%) 45 50N N N N N S S S S S Fig. 5. Neighboring craters in the northern mid-latitudes (34.8 N) showing identical flows, each flowing from south to north (towards the pole) (subframe of CTX orbit P17_007619_2153). that show evidence for ice-related flow (e.g., LVF, LDA or CCF). Flow in the latitude range examined is typically confined to latitudes poleward of 30 in each hemisphere, with exceptions in northern Hellas and in Arabia Terra. Ice-related flow features may have formed poleward of 55, but these surfaces are obscured by a late Amazonian mantling unit that is interpreted as ice-rich itself (Kreslavsky and Head, 2000; Mustard et al., 2001; Head et al., 2003). Ice-related flow in the northern hemisphere is concentrated in regions of steep topography (Deuteronilus Mensae, Protonilus Mensae, Acheron Fossae, Phlegra Montes, Kasei Vallis, etc.; Fig. 3a). Steep walls provide rockfall and hence a debris source to protect ice, in addition to shaded slopes that provide microclimates conducive to the preservation of ice. In the southern hemisphere, where steep slopes are far more abundant (Kreslavsky and Head, 2000), ice-related features are observed at nearly all locations between 30 and 50 S. The distribution of ice-related flow features confined to the interior of impact craters (Fig. 3b) is very similar to the distribution of all ice-related flow features in the mid-latitudes of Mars (Fig. 3a). More individual flow features were mapped in the southern hemisphere (910) compared to the northern hemisphere (690), and the majority are found between 30 and 45 latitude (north = 75.8%; south = 82.3%). Within the latitude band, there is a dominant pole-facing slope trend that is observed in each hemisphere (Figs. 1c, d, 2a, b, and 4). Of ice-related flow features between 30 and 45 N, 77.1% are oriented towards the pole, as are 86.5% of ice-related flow features between 30 and 45 S (Table 1). Frequently, neighboring craters of similar size show identical orientations of flow (Fig. 5). The few flow features observed equatorward of 30 in each hemisphere also show this preference. In both hemispheres, this preference for pole-facing slopes recedes at higher latitudes. Poleward of 45, more equator-facing flow features are found (Figs. 1a, b, 2c, d, and 4), such that in each hemisphere between 45 and 60, there is a near split in orientation preference (58.5% pole-facing in the southern hemisphere, 57.4% pole-facing in the northern hemisphere). Hence, use of the term concentric should also be modified to reflect this latitudedependence for symmetry: true concentric flow within craters is not observed equatorward of 45. This transition from pole-facing to concentric flow appears to occur at a slightly lower latitude in the northern hemisphere than in the south: in the N lat- Fig. 6. The distribution of flow orientations as a function of latitude on Mars. In the lower mid-latitudes of each hemisphere (<45 ), flow is dominantly pole-facing, suggesting that ice accumulates and/or is better preserved on steep, cold pole-facing slopes. In the higher mid-latitudes (45 60 ), conditions are cold enough to allow for flow emanating from all orientations, providing concentric flow of ice within impact craters. In the polar regions (>60 ) a young ice-rich mantle (Mustard et al., 2001) blankets the terrain and inhibits measurements of orientation.

7 J.L. Dickson et al. / Icarus 219 (2012) Fig. 8. The diameters of craters that host ice-related flow features in the midlatitudes of Mars. Ice is likely to accumulate on the walls of craters <1 km in diameter, but their small size preclude them from being included in our survey. Craters >75 km in diameter can also host ice-related features, but they are not included in this survey because (1) they may not be fully imaged in CTX data, or (2) they are less likely to have coherent rims that have not been breached (a prerequisite for this survey). Larger craters are also generally older and are less likely to have steep slopes, which would be more conducive to ice accumulation and flow. Fig. 7. Longitude distribution of ice-related flow features on Mars. Flow features are longitudinally distributed relatively evenly across the planet poleward of 25, with the exception of Deuteronilus/Protonilus Mensae in the northern hemisphere and the Newton Crater/Terra Sirenum and Northern/Eastern Hellas regions in the southern hemisphere. (A) Cumulative flow features vs. longitude poleward of 40 N. (B) Cumulative flow features vs. longitude for the northern hemisphere. (C) Cumulative flow features vs. longitude for the southern hemisphere. (D) Cumulative flow features vs. longitude poleward of 40 S. itude band, 68.5% of flow features are pole-facing, compared to 83.7% between 40 and 45 S. So while the broad latitude-dependent trends are identical in each hemisphere (Fig. 6), there are distinctions that can be drawn that could be due to: (1) absolute elevation differences between the northern lowlands and the southern highlands (ice should be more stable on all slopes at lower elevations) and (2) the availability of steep slopes in each hemisphere (Kreslavsky and Head, 2000). We also examined the distribution of ice-related flow features as a function of longitude (Fig. 7). While distribution is broadly uniform in both hemispheres, each hemisphere shows a major peak in one respective area. For the north, the well-studied features along the dichotomy boundary at Deuteronilus and Protonilus Mensae clearly represent the largest concentration of ice-related morphology north of the equator. In the south, a similar concentration is observed in the Newton Crater region, south of Terra Sirenum. A lower-amplitude but measureable concentration is also observed in the terrain along thenorthernandeasternrimsofthehellasbasin.thesearethesameregions where these types of features were first mapped in the global mosaicsof Vikingimagery(Squyres, 1978, 1979). In our survey, the impact craters that host ice-related flow features range in diameter from 1kmto72 km, with an average of 11.9 km (Fig. 8). It is possible that additional craters with ice-related flow features of both larger and smaller size exist. While we did observe flow features within craters as small as 1.2 km in diameter, features within craters of this size are harder to recognize; our criteria for mapping the existence and orientation of features required recognition of morphological markers of flow. For this reason, our survey does not include many potential features at this scale. For larger craters (>75 km), their size often precludes them from having complete coverage in CTX images, which is a prerequisite for inclusion in our survey. Another possible effect is that large craters are on average older than smaller craters, and thus have more degraded slopes which may be less preferred for ice accumulation. The regions with the highest density of ice-related flow features (Deuteronilus/Protonilus Mensae, Eastern Hellas, Phlegra Montes, etc.) exhibit flow features within more than half of the impact craters in those regions. We excluded flow features within craters with breached rims from our orientation measurements, as the uneven availability of slope angles would bias our statistics. These features, though, could still be valuable as potential sites for regional glaciation, where breaches could have been exploited for infilling of ice/debris from adjacent terrain. Our survey revealed 34 craters that had breached rims and clear evidence for flow (Fig. 3b). Of these 34, 27 showed evidence only for ice flowing out of the crater. Of more interest are the 7 craters observed that show clear evidence for ice entering the crater through a breach in the rim, suggesting that ice did not accumulate solely along the interior of the crater rim. The small sample set precludes any conclusions with regard to regional ice fields that could have sourced these flows, but

8 730 J.L. Dickson et al. / Icarus 219 (2012) Fig. 9. The correlation between equator-facing flow (blue) and the onset of pedestal craters (red) in each hemisphere, as a function of latitude. Using the database produced by Kadish et al. (2009), pedestal craters are only observed poleward of 40 in each hemisphere. This correlates well with the onset of equator-facing flow within impact craters, as mapped in this study. Taken together, this argues that ice was once stable on all slopes poleward of in each hemisphere, while flow in lower latitudes is controlled by microclimate conditions: flow is only observed on cold, pole-facing slopes where ice has the greatest opportunity to accumulate and where deposition was greater than ablation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) these features are observed in the regions predicted to have maximum accumulation potential under high-obliquity conditions (Forget et al., 2006; Madeleine et al., 2009): Eastern Hellas (4 craters), Deuteronilus Mensae (2 craters), and the Phlegra Montes (1 crater). 4. Discussion and conclusions The importance of microclimates to the surface and near-surface accumulation and preservation of ice is demonstrated by the trends we observe in the orientation of ice-related flow in the mid-latitudes of Mars. The symmetry of orientation across both hemispheres shows that while we are likely to be observing local features formed as a function of microclimates, their global distribution reflects how the martian climate behaved during this past period of net ice accumulation on the surface: ice was preserved in colder regions but not in warmer regions, even when those two are found on the opposing walls of the same crater (e.g. Fig. 5). Ice-related accumulation and flow in the mid-latitudes of Mars occurred under specific latitude, orientation and atmospheric conditions. GCMs have shown (e.g. Madeleine et al., 2009) that under periods of generally high obliquity and increased atmospheric dust opacity, ice can be deposited nearly anywhere in the mid-latitudes, similar to the distribution of young gullies and VFF (Milliken et al., 2003); conditions of longer-term ice accumulation and glaciation (Head et al., 2010) require specific conditions within this realm (Madeleine et al., 2009). Within these regions, our study finds that flow occurred at least in the places where ice is most likely to accumulate over long time periods: steep, cold pole-facing slopes in the lower mid-latitudes, and steep slopes of any orientation in the higher mid-latitudes. The possibility remains that flow also occurred elsewhere, but evidence has not been preserved. A necessary condition for these flow features to be detected today is that there was a layer of debris that was deformed in patterns diagnostic of viscous flow (e.g., Fastook et al., 2011) Had a flow occurred with no superposing debris layer, (e.g., Hauber et al., 2008) the ice could have sublimated without leaving any evidence of its past existence. This makes the detection of past ice fields a substantially more difficult question than the one being addressed here. This raises the question of whether these CCF features could have been sourced by regional/hemispheric ice sheets and ice fields (Fastook and Head, 2012) instead of locally on the walls of the craters where they are now observed? If flow was initiated on the plains and flowed into craters, the steep walls would have provided a source of debris not available on the plains, such that evidence for glaciation would only be preserved within the crater, as rim-crest scarps were exposed and debris shed onto the ice. This is perhaps possible for latitudes poleward of 45, where flow within craters is concentric, but unlikely for most locations within the lower mid-latitudes ([45 ), where dominant pole-facing flow orientations are observed (Fig. 4). Rare instances where flow appears to enter craters through breaches in the rim could provide critical clues regarding the past existence of regional ice fields. In the northern hemisphere, this transition from pole-facing flow to concentric flow occurs at the same latitude as the southernmost extent of non-equatorial pedestal craters (Fig. 9) (Kadish et al., 2009), unique features in which the ejecta deposit and the crater itself are perched above the surrounding plains (McCauley, 1973). A similar correlation is observed in the southern hemisphere, though the paucity of landforms indicative of flow poleward of 50 S (Fig. 3a) focuses the correlation directly at the onset of pedestal craters/equator-facing flow. Our data show that the percentage of ice-related flow features that are equator-facing increases significantly between 40 and 50 S, which corresponds to the lowest-latitude where regionally distributed pedestal craters were mapped by Kadish et al. (2009). Ice-related flow features of all orientations are not found poleward of this region in the southern hemisphere, which could be an effect of (1) low density of CTX coverage in the S latitude band, or (2) insufficient accumulation of ice at the high altitudes of the southern highlands. Pedestal craters and ice-related flow features both formed in the Amazonian Period, though strong constraints on their absolute formation age are difficult to ascertain. Kadish et al. (2009) assumed that high-latitude pedestal craters all formed in the same era, which provided enough statistics (1363 craters) to conclude that these features are at least 100 Myr old. This age is comparable to age estimates for LDA (e.g. Crown et al., 1992; Mangold, 2003; Joseph et al., 2011; Berman et al., 2011), LVF (e.g. Morgan et al., 2009), and CCF (e.g. Levy et al., 2009). What are the implications of this correlation? Initially hypothesized to be products of eolian reworking and deflation (McCauley, 1973; Arvidson et al., 1979), recent modeling efforts (Wrobel et al., 2006) and global mapping with recently-obtained high-resolution imaging data (Barlow, 2005; Kadish et al., 2008b, 2009, 2010; Kadish and Head, 2011) bolster experimental evidence (Schultz,

9 J.L. Dickson et al. / Icarus 219 (2012) ) and observations from Viking (Schultz and Lutz, 1988) that pedestal craters preserve evidence of epochs when ice accumulated on high-latitude plains to thicknesses of up to 260 m (Kadish et al., 2010). If this interpretation is correct, then it correlates well with our findings. Both results suggest that at certain times poleward of in each hemisphere, ice accumulated regionally, regardless of slope orientation and specific microenvironments. Provided that slopes were steep enough, this accumulation could induce flow and yield the concentric crater fill that is preserved today. Equatorward of 45, ice accumulation has been limited to pole-facing slopes that lie outside of the present-day tropics. In these locales, the importance of microclimates are amplified; the disparate insolation conditions provided by opposite walls of the same crater rim yield net ice accumulation on the pole-facing slope, with negligible accumulation on the equator-facing slope (e.g. Fig. 5) (Costard et al., 2002; Kreslavsky and Head, 2006; Kreslavsky et al., 2008; Morgan et al., 2010). In summary, these findings show that the most recent phase of significant ice-related flow on Mars is likely to have been focused (1) in cold-traps on steep slopes in the mid-latitudes and (2) over all steep slopes at high-latitudes (>45 ). The areal correlation of concentric flow in CCF at these latitudes and the distribution of pedestal craters suggests that if their formation was synchronous, then there could have been regional ice sheets in place for potentially extended periods of time at these latitudes, analogous in scale to terrestrial continental glaciation. These observations, along with ongoing measurements of the age of ice-rich deposits, will provide important new constraints on the Amazonian climate history of Mars. Acknowledgments We are thankful to the science, engineering and operational teams for the Context Camera at MSSS for acquiring and distributing the primary data in this study. Baerbel Lucchitta and an anonymous reviewer provided helpful and thorough comments. We appreciate support with high-volume data processing from John Huffman and the Center for Computation and Visualization (CCV) at Brown. Input from Joe Levy and David Hollibaugh-Baker helped guide this study. Financial support from NASA Grants from the Mars Data Analysis Program (NNX09A146G) and the Mars Express High-Resolution Stereo Camera (HRSC) team (JPL ) are gratefully acknowledged and appreciated. References Arvidson, R.E., Guinness, E., Lee, S., Differential Aeolian redistribution rates on Mars. Nature 278, Baker, D.M.H., Head, J.W., Marchant, D.R., Flow patterns of lobate debris aprons and lineated valley fill north of Ismeniae Fossae, Mars: Evidence for extensive mid-latitude glaciation in the Late Amazonian. Icarus 207, Balme, M. et al., Orientation and distribution of recent gullies in the southern hemisphere of Mars: Observations from High Resolution Stereo Camera/Mars Express (HRSC/MEX) and Mars Orbiter Camera/Mars Global Surveyor (MOC/ MGS) data. J. Geophys. Res. 111, E JE Barlow, N.G., A New Model for Pedestal Crater Formation. LPI Contribution No. 1273, Basilevsky, A.T. et al., Morphology and geological structure of the western part of the Olympus Mons volcano on Mars from the analysis of the Mars Express HRSC imagery. Solar Syst. Res. 39, Berman, D.C., Crown, D.A., and Joseph, E.C.S., 2011, Determining erosional/ depositional history of Deuteronilus Mensae, Mars using categorized crater size frequency distributions. Lunar Planet. Sci. 42, Boynton, W.V. et al., Distribution of hydrogen in the near surface of Mars: Evidence for subsurface ice deposits. Science 297, Bridges, N.T., Lackner, C.N., Northern hemisphere martian gullies and mantled terrain: Implications for near-surface water migration in Mars recent past. J. Geophys. Res. 111, E Byrne, S. et al., Distribution of mid-latitude ground ice on Mars from new impact craters. Science 325, Chuang, F.C., Crown, D.A., Surface characteristics and degradational history of debris aprons in the Tempe Terra/Mareotis Fossae region of Mars. Icarus 179, Chuang, F.C., Crown, D.A., Geologic Map of MTM 35337, 40337, and Quadrangles, Deuteronilus Mensae Region of Mars, US Geological Survey Scientific Investigations Series Map 3079, Scale 1:1004,000. Costard, F., Forget, F., Mangold, N., Peulvast, J.P., Formation of recent martian debris flows by melting of near-surface ground ice at high obliquity. Science 295, Crown, D.A., Price, K.H., Greeley, R., Geologic evolution of the east rim of the Hellas basin, Mars. Icarus 100, Dickson, J.L., Head, J.W., Kreslavsky, M., Martian gullies in the southern midlatitudes of Mars: Evidence for climate-controlled formation of young fluvial features based upon local and global topography. Icarus 188, Dickson, J.L., Head, J.W., Marchant, D.R., Late Amazonian glaciation at the dichotomy boundary on Mars: Evidence for glacial thickness maxima and multiple glacial phases. Geology 36, Dickson, J.L., Fassett, C.I., Head, J.W., Amazonian-aged fluvial valley systems in a climatic microenvironment on Mars: Melting of ice deposits on the interior of Lyot Crater. Geophys. Res. Lett Dickson, J.L., Head, J.W., Marchant, D.R., Kilometer-thick ice accumulation and glaciation in the northern mid-latitudes of Mars: Evidence for crater-filling events in the Late Amazonian at the Phlegra Montes. Earth Planet. Sci. Lett. 294, Fassett, C.I., Dickson, J.L., Head, J.W., Levy, J.S., Marchant, D.R., Supraglacial and proglacial valleys on Amazonian Mars. Icarus 208, Fastook, J.L., Head, J.W., Mid-latitude amazonian glaciation on Mars: Controls on accumulation and glacial flow patterns. Lunar Planet. Sci. 43, Fastook, J.L., Head, J.W., Marchant, D.R., Formation of ice-rich lobate debris aprons through regional icesheet collapse and debris-cover armoring. Lunar Planet. Sci. 42, Feldman, W.C. et al., Global distribution of neutrons from Mars: Results from Mars Odyssey. Science 97, Forget, F., Haberle, R.M., Montmessin, F., Levrard, B., Head, J.W., Formation of glaciers on Mars by atmospheric precipitation at high obliquity. Science 311, Hauber, E., van Gasselt, S., Chapman, M.G., Neukum, G., Geomorphic evidence for former lobate debris aprons at low latitudes on Mars: Indicators of the Martian paleoclimate. J. Geophys. Res JE Head, J.W., Marchant, D.R., Cold-based mountain glaciers on Mars: Western Arsia Mons. Geology 31, Head, J.W., Marchant, D.R., Inventory of ice-related deposits on Mars: Evidence for burial and long-term sequestration of ice in non-polar regions and implications for the water budget and climate evolution. Lunar Planet. Sci. 40, Head, J.W., Mustard, J.F., Kreslavsky, M.A., Milliken, R.E., Marchant, D.R., Recent ice ages on Mars. Nature 426, Head, J.W., Marchant, D.R., Agnew, M.C., Fassett, C.I., Kreslavsky, M.A., 2006a. Extensive valley glacier deposits in the northern mid-latitudes of Mars: Evidence for late Amazonian obliquity-driven climate change. Earth Planet. Sci. Lett. 241, Head, J.W., Nahm, A.L., Marchant, D.R., Neukum, G., 2006b. Modification of the dichotomy boundary on Mars by Amazonian mid-latitude regional glaciation. Geophys. Res. Lett. 33, L08S03. Head, J.W., Marchant, D.R., Dickson, J.L., Kress, A.M., Baker, D.M., Northern mid-latitude glaciation in the Late Amazonian Period of Mars: Criteria for the recognition of debris-covered glacier and valley glacier landsystem deposits. Earth Planet. Sci. Lett. 294, Heldmann, J.L., Mellon, M.T., Observations of martian gullies and constraints on potential formation mechanisms. Icarus 168, Heldmann, J.L., Carlsson, E., Johansson, H., Mellon, M.T., Toon, O.B., Observations of martian gullies and constraints on potential formation mechanisms. II. The northern hemisphere. Icarus 188, Holt, J.W. et al., Radar sounding evidence for buried glaciers in the southern mid-latitudes of Mars. Science 322, Joseph, E.C.S., Crown, D.A., Berman, D.C., and Chuang, F.C., Using CTX-based crater size frequency distributions to refine the geologic history of Deuteronilus Mensae, Mars. Lunar Planet. Sci. 42, Kadish, S.J., Head, J.W., Impacts into non-polar ice-rich paleodeposits on Mars: Excess ejecta craters, perched craters and pedestal craters as clues to Amazonian climate history. Icarus 215, Kadish, S.J., Head, J.W., Parsons, R.L., Marchant, D.R., 2008a. The Ascraeus Mons fanshaped deposit: Volcano ice interactions and the climatic implications of coldbased tropical mountain glaciation. Icarus 197, Kadish, S.J., Head, J.W., Barlow, N.G., Marchant, D.R., 2008b. Martian pedestal craters: Marginal sublimation pits implicate a climate-related formation mechanism. Geophys. Res. Lett. 35, L Kadish, S.J., Barlow, N.G., Head, J.W., Latitude dependence of martian pedestal craters: Evidence for a sublimation-driven formation mechanism. J. Geophys. Res.. Kadish, S.J., Head, J.W., Barlow, N.G., Pedestal crater heights on Mars: A proxy fort he thicknesses of past, ice-rich, Amazonian deposits. Icarus 210, Kreslavsky, M.A., Head, J.W., Kilometer-scale roughness of Mars: Results from MOLA data analysis. J. Geophys. Res. 105,

10 732 J.L. Dickson et al. / Icarus 219 (2012) Kreslavsky, M.A., Head, J.W., Modification of impact craters in the northern plains of Mars: Implications for the Amazonian climate history. Meteorit. Planet. Sci. 41, Kreslavsky, M.A., Head, J.W., Marchant, D.R., Periods of active permafrost layer formation during the geological history of Mars: Implications for circum-polar and mid-latitude surface processes. Planet. Space Sci. 56, dx.doi.org/ /j.pss Laskar, J., Correia, A.C.M., Gastineau, M., Joutel, F., Levrard, B., Robutel, P., Long term evolution and chaotic diffusion of the insolation quantities of Mars. Icarus 170, Levrard, B., Forget, F., Montmessin, F., Laskar, J., Recent ice-rich deposits formed at high latitudes on Mars by sublimation of unstable equatorial ice during low obliquity. Nature 431, Levy, J.S., Head, J.W., Marchant, D.R., Lineated valley fill and lobate debris apron stratigraphy in Nilosyrtis Mensae, Mars: Evidence for phases of glacial modification of the dichotomy boundary. J. Geophys. Res. 112, E dx.doi.org/ /2006je Levy, J.S., Head, J.W., Marchant, D.R., Concentric crater fill in Utopia Planitia: History and interaction between glacial brain terrain and periglacial mantle processes. Icarus 202, Levy, J., Head, J.W., Marchant, D.R., Concentric crater fill in the northern midlatitudes of Mars: Formation processes and relationships to similar landforms of glacial origin. Icarus, in press. doi: Li, H., Robinson, M.S., Jurdy, D.M., Origin of martian northern hemisphere mid-latitude lobate debris aprons. Icarus 176, Lucchitta, B.K., Mars and Earth: Comparison of cold climate features. Icarus 45, Lucchitta, B.K., Ice and debris in the fretted terrain, Mars. 14th Proceedings of the Lunar and Planetary Science Conference, JGR, 89, B407 B418. Madeleine, J.-B., Forget, F., Head, J.W., Levrard, B., Montmessin, F., Millour, E., Amazonian northern mid-latitude glaciation on Mars: A proposed climate scenario. Icarus 203, Malin, M.C., Edgett, K.S., Evidence for recent groundwater seepage and surface runoff on Mars. Science 288, Mangold, N., Geomorphic analysis of lobate debris aprons on Mars at Mars Orbiter Camera scale: Evidence for ice sublimation initiated by fractures. J. Geophys. Res Marchant, D.R., Head, J.W., Antarctic Dry Valleys: Microclimate zonation, variable geomorphic processes, and implications for assessing climate change on Mars. Icarus 192, McCauley, J.F., Mariner 9 evidence for wind erosion in the equatorial and midlatitude regions of Mars. J. Geophys. Res. JB078i020p Milkovich, S.M., Head, J.W., Marchant, D.R., Debris-covered piedmont glaciers along the northwest flank of the Olympus Mons scarp: Evidence for lowlatitude ice accumulation during the Late Amazonian of Mars. Icarus 181, Milliken, R.E., Mustard, J.F., Goldsby, D.L., Viscous flow features on the surface of Mars: Observations from high-resolution Mars Orbiter Camera (MOC) images. J. Geophys. Res. 108, E Mischna, M.A., Richardson, M.I., Wilson, R.J., McCleese, D.J., On the orbital forcing of Martian water and CO 2 cycles: A general circulation model study with simplified volatile schemes. J. Geophys. Res JE Morgan, G.A., Head, J.W., Marchant, D.R., Lineated valley fill (LVF) and lobate debris aprons (LDA) in the Deuteronilus Mensae northern dichotomy boundary region, Mars: Constraints on the extent, age and episodicity of Amazonian glacial events. Icarus 202, Morgan, G.A., Head, J.W., Forget, F., Madeleine, J.-B., Spiga, A., Gully formation on Mars: Two recent phases of formation suggested by links between morphology, slope orientation and insolation history. Icarus 208, Mustard, J.F., Cooper, C.D., Rifkin, M.K., Evidence for recent climate change on Mars from the identification of youthful near-surface ground ice. Nature 412, Pierce, T.L., Crown, D.A., Morphologic and topographic analyses of debris aprons in the eastern Hellas region, Mars. Icarus 163, Plaut, J.J. et al., Radar evidence for ice in lobate debris aprons in the midnorthern latitudes of Mars. Geophys. Res. Lett. 36, L Schon, S.C., Head, J.W., Milliken, R.E., A recent ice age on Mars: Evidence for climate oscillations from regional layering in mid-latitude mantling deposits. Geophys. Res. Lett. 36, L Schultz, P.H., Impact vaporization of volatile-rich targets; experimental results and implications. Lunar Planet. Sci. 19, Schultz, P.H., Lutz, A.B., Polar wandering of Mars. Icarus 73, Shean, D.E., Candidate ice-rich material within equatorial craters on Mars. Geophys. Res. Lett.. Shean, D.E., Head, J.W., Marchant, D.R., Origin and evolution of a cold-based tropical mountain glacier on Mars: The Pavonis Mons fan-shaped deposit. J. Geophys. Res.. Shean, D.E., Head, J.W., Fastook, J.L., Marchant, D.R., Recent glaciation at high elevations on Arsia Mons, Mars: Implications for the formation and evolution of large tropical mountain glaciers. J. Geophys. Res JE Smith, P.H. et al., H 2 O at the phoenix landing site. Science 325, Squyres, S.W., Martian fretted terrain Flow of erosional debris. Icarus 34, Squyres, S.W., The distribution of lobate debris aprons and similar flows on Mars. J. Geophys. Res. 84, Wrobel, K., Schultz, P., Crawford, D., An atmospheric blast/thermal model for the formation of high-latitude pedestal craters. Meteorit. Planet. Sci. 41,