Icarus 206 (2010) 669–684

Contents lists available at ScienceDirect

Icarus journal homepage: www.elsevier.com/locate/icarus

The age of the Medusae Fossae Formation: Evidence of Hesperian emplacement from crater morphology, stratigraphy, and ancient lava contacts Laura Kerber *, James W. Head Department of Geological Sciences, Brown University, Providence, RI 02912, USA

a r t i c l e

i n f o

Article history: Received 23 June 2009 Revised 29 September 2009 Accepted 2 October 2009 Available online 9 October 2009 Keywords: Mars Mars, Surface Geological processes Volcanism

a b s t r a c t The Medusae Fossae Formation (MFF), covering about 2.1  106 km2 (with an estimated volume of 1.4  106 km3) and straddling the equatorial region of Mars east of Tharsis, has historically been mapped and dated as Amazonian in age. Analysis of the MFF using a range of new observations from recent mission data at multiple resolutions reveals evidence that the formation is older than previously hypothesized, with parts of the MFF having formed in the Hesperian and parts having been reworked and reformed throughout the Amazonian, up to the present. Ancient outcroppings of the MFF, edged with jagged yardangs, became a ‘‘mold” for embaying Hesperian-aged lavas. The erosion of the MFF left solidified lava ‘‘casts” in the embaying lava unit. This lava edge morphology permits the identification of ancient contacts between the MFF and Hesperian-aged lava terrain. Additionally, the flanking fan of the Hesperian-aged Apollinaris Patera volcano embays the formation at its foot, indicating that parts of the MFF were formed in the Hesperian. Erosion has erased and inverted many of the superposed craters in the region, showing that very young Amazonian ages derived from impact crater size–frequency distributions are resurfacing ages, and not emplacement ages. We find abundant evidence that the formation is extremely mobile and continuously reworked. We conclude that a significant part of the MFF may have originally been emplaced in the Hesperian. These observations place new constraints on the mode of origin of the MFF. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction The Medusae Fossae Formation (MFF) is a complicated and discontinuous formation located in southern Elysium and Amazonis Planitias and northern Memnonia and Aeolis Planitias (130–230°E and 12°S–12°N), covering an area of approximately 2.1  106 km2 with an estimated volume of 1.4  106 km3 (Bradley et al., 2002) (Fig. 1). The formation is characterized by large accumulations of fine-grained, friable deposits (e.g., Scott and Tanaka, 1986; Greeley and Guest, 1987; Zimbelman et al., 1996). The boundaries of the deposit correlate with some parts of the ‘‘stealth” and ‘‘greater stealth” regions defined by Muhleman et al. (1991) and Butler (1994) on the basis of their unusual radar properties (extremely low return when probed with 3.5-cm wavelength radar, indicating a very low-density material with very few rocks), though whether the units are related is unresolved (Edgett et al., 1997). Later radar analysis conducted with the Mars Advanced Radar for Subsurface and Ionospheric Sounding (MARSIS) at longer

* Corresponding author. Address: Geological Sciences, Brown University, 324 Brook St., Box 1846, Providence, RI 02912, USA. Fax: +1 401 863 3978. E-mail address: [email protected] (L. Kerber). 0019-1035/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.icarus.2009.10.001

wavelengths confirmed that the MFF had dielectric properties consistent with either relatively clean water ice or dry, low-density materials (Watters et al., 2007). The age of the MFF has been determined mainly using crater counting techniques and stratigraphic relationships. Scott and Tanaka (1982) divided the formation into seven separate units and counted the number of >1 km diameter craters on each unit, yielding cumulative crater counts between 729 and 63.6 > 1 km craters/106 km2. The majority of the formation was thus placed in the mid-Amazonian, with a few units in the lower Amazonian and one unit in the upper Amazonian. These units were later combined into three main geologic units based on their color and states of erosion: an upper member (Amu) characterized by smooth and rolling light-colored plains, a middle member (Amm) characterized by more progressed erosion, and a lower member (Aml) characterized by extensive erosion and a darker color (Scott and Tanaka, 1986; Greeley and Guest, 1987). These units were placed in the middle to late Amazonian based on cumulative superposed craters (0–200 > 2 km craters/106 km2). Later crater counts done by Werner (2005) on areas near Gusev Crater and southwest Amazonis Mensae were interpreted to suggest a global formation age for the MFF of 1.6 Ga (early to midAmazonian according to Hartmann and Neukum (2001)).

670

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

Fig. 1. A regional MOLA hillshade of the Medusae Fossae Formation, stretched to show detail within the formation. Bottom panel shows the locations of the figures. Black outline indicates the boundary of the Medusae Fossae Formation as mapped by Scott and Tanaka (1986) and Greeley and Guest (1987). Lat.: 27°S–24°N, Lon.: 140 to 130°E.

There is evidence of significant wind-related erosion; yardangs are abundant in the region (Ward, 1979), and their frequent shifts in orientation (depending on stratigraphic level exposed) has been cited as evidence for changes in wind direction in the past (Zimbelman et al., 1997; Zimbelman and Griffin, 2009), or changes in structurally controlled jointing in the material itself (Scott and Tanaka, 1982; Bradley et al., 2002), though any joints would have to be subparallel to the prevailing wind in order to initiate yardang formation (Livingstone and Warren, 1996; Inbar and Risso, 2001). Evidence for aeolian modification of the unit suggests that much of the impact cratering record may have been erased as friable units were eroded and long-buried terrains were exhumed (Schultz and Lutz, 1988; Schultz, 2006, 2007). For this reason, it is important to attempt to distinguish between the original formation age of the MFF and the modification or resurfacing age(s) of the MFF. There have been many hypotheses regarding the origin of the MFF, including ash flow tuffs or ignimbrites (Scott and Tanaka, 1982, 1986), pyroclastic or aeolian materials (Scott and Tanaka, 1986; Greeley and Guest, 1987; Tanaka, 2000), paleopolar deposits or loess (Schultz and Lutz, 1988; Schultz, 2002), carbonate platforms (Parker, 1991), ice-rich dusty mantling deposited during high obliquity (Head and Kreslavsky, 2004), or ash fall (Tanaka, 2000; Bradley et al., 2002; Hynek et al., 2003). As image resolution has improved, the carbonate platform hypothesis and the rafted pumice deposit hypothesis have been deemed less likely (Mandt et al., 2007, 2008). Paleopolar deposits have been rejected based on the young age of the formation: Grimm and Solomon (1986) have suggested that any significant true polar wander would leave

a tectonic signature, which is not found. This result would suggest that no spin axis reorientation has taken place since at least the end of the heavy bombardment. Additionally, the locations and ages of the south-pole pitted terrains suggest that no significant polar wandering has taken place since the early Hesperian (Tanaka, 2000). Mantling during high obliquity could result in deposits that resemble polar deposits without the need for true polar wander (Head and Kreslavsky, 2004); however, shallow radar (SHARAD) analysis of the MFF has failed to detect the fine-scale layering that is characteristic of the polar deposits (Carter et al., 2009). Data from SHARAD suggest that unlike the polar deposits, MFF layers have low permittivity contrasts or the layers are discontinuous (Carter et al., 2009). Several analyses of these hypotheses have been done, reaching the conclusion that the formation is most likely composed of volcanic ash, ignimbrites, or aeolian dust (Zimbelman et al., 1997; Mandt et al., 2007, 2008). The composition and mode of emplacement of the deposits have broad implications for the evolution of the surface and atmosphere of Mars in the Amazonian. A volcanic hypothesis would require the eruption of large amounts of pyroclastic material from a nearby volcanic center. Both the Tharsis Montes and Elysium Montes have been active in the Amazonian and could potentially provide a source for the needed pyroclastics for the deposit. The surfaces of these volcanoes are currently dominated by effusive lava flows; however, some evidence of recent ash has been noted at the summit of Arsia Mons (Mouginis-Mark, 2002). Explosive volcanic eruptions would be able to emplace blankets of material relatively instantaneously over large areas (Wilson and Head, 1994). If the deposits were poorly lithified, they could be steadily eroded

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

over long periods of time in the Mars environment. Different yardang orientations and bidirectional yardang patterns would form as a result of changing wind directions (Zimbelman et al., 1997; Zimbelman and Griffin, 2009). Emplacement as atmospheric dust would require a periodic climate change sufficient to transform the region alternately from a dust sink into an area of intense aeolian erosion, and for a sufficiently long time to accumulate up to 3 km of airborne dust. Periods of accumulation would have to be interrupted by periods of erosion to form internal layers of yardangs with different orientations (Sakimoto, 1999; Zimbelman and Griffin, 2009). This process may require a longer time period than that of volcanic emplacement, as volcanic emplacement does not require long periods of accumulation. Some modified and inverted fluvial channels have also been found within the deposit (Zimbelman et al., 2000; Burr et al.,

671

2009; Griffin and Zimbelman, 2009; Zimbelman and Griffin, 2009), indicating that there was some fluvial activity during the emplacement or modification of the Medusae Fossae Formation (Fig. 2). If the MFF is among the youngest surficial deposits on Mars, it is implied that meandering, channelized flow of water must have extended into the Amazonian (Burr et al., 2006). If the age is correct, this is a significant observation in that such fluvial activity in the Amazonian is very rare (e.g., Dickson et al., 2009; Fassett et al., 2009). Because of the significant implications that these findings have for the evolution of Mars and the martian atmosphere and climate, it is important to re-examine the evidence for the Amazonian age of the Medusae Fossae Formation. The current conclusion that the MFF is of Amazonian age comes from two main lines of argument, as discussed earlier. First, the relatively few impact craters superposed on the unit led to an Amazonian age assignment for

Fig. 2. Evidence of fluvial reworking of the MFF. (A) A channel near Arsia Mons (3°N, 140°W) described by Zimbelman et al. (2000). (B) Small, dendritic channels flow towards the larger channel in (A). (C) Inverted channels in southeast Aeolis Planum described by Burr et al. (2009) (HSRC images h2146, h2135, h2124; HSRC image h2146; CTX image P03_002279_1737, respectively).

672

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

the MFF in global mapping (e.g., Scott and Tanaka, 1986; Greeley and Guest, 1987; less than 200 craters greater than 2 km in diameter in 106 km2). In addition, impact crater size–frequency distribution analyses provide a similar age (1.6 Ga; Werner, 2005). The second line of argument for the Amazonian age comes from stratigraphic relations. The MFF is superposed on Amazonian-aged

lowland terrain and directly overlies Hesperian-aged outflow channels such as Labou Vallis (Bradley and Sakimoto, 2001; Bradley et al., 2002). Using high resolution data from the Mars Global Surveyor Mars Orbiter Camera (MOC) and Mars Orbiter Laser Altimeter (MOLA), the Mars Odyssey Thermal Emission Imaging System (THEMIS), the Mars Express High Resolution Stereo Camera (HRSC),

Fig. 3. A comparison of MFF and mid-latitude pedestal craters. Mid-latitude pedestal craters are very symmetrical with subdued edges. MFF pedestal craters are often quite asymmetric with zigzagging pedestal margins (see Kadish et al., 2009a,b). From top right: V18359003, V19230010, V09954010, V13935010.

Fig. 4. Sea-urchin-like pedestal craters in CTX and HRSC images. The indurated portions of the crater ejecta deposits extend much further than is typical for such craters, and the sinuosity of the pedestal outline is anomalously high. From top right: P06_003544_1703, h0998, P02_001843_1716 (bottom two panels).

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

and the Mars Reconnaissance Orbiter Context Imager (CTX) and High Resolution Imaging Science Experiment (HiRISE), we reexamined the relationships both within the MFF and with respect to adjacent units. In this contribution, we examine first the state of crater preservation in the MFF, and thus the basis for impact crater size–frequency distribution ages, second, the nature of the MFF itself and its mode of emplacement and modification, and third, the stratigraphic relationships between the MFF and other units formed prior to, subsequent to, and contemporaneously with the MFF. On the basis of this analysis, we conclude that the MFF is a dynamic unit, constantly being eroded, re-deposited, and exhumed, making impact crater size–frequency distribution dating difficult, and making most stratigraphic relationships not definitive of a formation age. We conclude first that a significant part of the MFF is Hesperian in age, much older than previously hypothesized, and second, that it has been eroded, re-deposited and exhumed throughout the Amazonian. 2. Observations 2.1. Crater preservation The method of age-dating by crater counting requires several preconditions (e.g., Hartmann and Neukum, 2001). Each unit being dated must be emplaced during one, geologically short event. After

Fig. 5. Twenty-kilometer diameter impact crater showing contrasting ejecta deposit morphologies. The left half of the crater displays a double-layered lobate ejecta deposit. The perched, jagged edges of the right half of the ejecta deposit are typical of MFF pedestals, indicating that the projectile struck the edge of a previous MFF deposit. HRSC images h2965_0000.nd2.03.04 and h2976_0000.nd2.04.04, supplemented by the THEMIS global mosaic.

673

emplacement, the unit must remain unmodified except for craters formed by a well-known population of incoming projectiles. Lastly, the population of craters must not become so great as to reach saturation. If these conditions are met, the crater count age of a unit may be said to be synonymous with the emplacement age, and uncertainties are limited to the character of the incoming projectile population. Of these conditions, the Medusae Fossae Formation meets neither the condition of a geologically short emplacement nor the condition of a subsequently unmodified exposure surface. In particular, the MFF has been drastically modified by aeolian processes, which have continuously eroded the exposure surface and masked the cratering record through erasure, modification, and burial of craters. Because of these modifications, the crater record preserved on the MFF (i.e., the crater retention age) is not likely to represent its emplacement age. The relative paucity of craters on its surface yields far more information about its level of activity through time than it does about when the formation was emplaced. This is true of the MFF and any other terrain which has been subject to significant erosion, an issue that has been noted by others in regard to the MFF as well as the polar regions and Arabia Terra (Schultz and Lutz, 1988; Tanaka, 2000; Greeley et al., 2001). By studying the degradation and selective preservation of the MFF cratering

Fig. 6. Progressive crater modification in the MFF. (A) Craters are formed in a finegrained unit. (B) The heavily cratered unit is eroded by the wind, leaving armored remnant ejecta deposits. (C) The ejecta deposit is finally eroded away, leaving only the strongly welded crater floor. (D) In time, the wind erodes the remaining crater floor into a knob. HRSC images h0987 (A and C), h0998 (B and D).

674

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

record, information may be extracted about the material properties of the unit and the qualities that may (or may not) be used to characterize its age. Pedestal craters (Fig. 3) are observed both within the MFF and around its margins (Schultz and Lutz, 1988). Pedestal craters are distinct from other craters in that their ejecta deposits are perched above the surrounding terrain. They are believed to originally form at the same level as the surrounding plains and are often associated with terrain which is interpreted to be volatile rich (e.g., Schultz and Lutz, 1988; Wrobel et al., 2006; Kadish et al., 2008). Over time, areas around the crater are deflated but the crater bowl and pedestal remain at their original elevations, because they are indurated (perhaps during the impact process as a result of atmospheric interactions, substrate gardening, or impact melt generation, e.g., Wrobel et al., 2006; Barlow, 2006). While this type of crater is common poleward of 40° (Mouginis-Mark, 1979), it is rare at low latitudes (Kadish and Barlow, 2006; Kadish et al., 2008, 2009a,b), and the pedestal craters that appear in the Medusae Fossae Formation differ from those found at higher latitudes (e.g., Kadish et al., 2009a) in several ways (Fig. 3). Pedestal craters found at higher latitudes have both perched ejecta deposits and perched crater bowls, meaning that the bottom of the crater bowl is also perched above the surrounding terrain (Kadish et al., 2009a). The MFF pedestal crater-bowls extend lower than the surrounding terrain, despite evidence for extensive erosion. Barlow (1993) has noted that craters formed in this deposit have unusual depth to diameter ratios (approximately 79% deeper than similar-sized simple craters elsewhere), suggesting that projectiles may penetrate

deeper into this unit compared to other units on Mars (perhaps due to more efficient compression of the low-density material). One of the most striking differences between the MFF pedestal craters and those in higher latitudes is that their pedestals usually have a crisp and asymmetric morphology when compared to polar pedestal craters, which exhibit a far more subdued, circular shape (Fig. 3) (Kadish et al., 2009a). The character of the MFF pedestals is interpreted to be an indication that unlike polar pedestal craters, the surrounding terrain is most likely being removed by the wind rather than by sublimation. MFF pedestal craters have sinuous pedestal margins and long, thin rays. Like other pedestal craters, MFF pedestals have a larger extent than that expected for the ejecta blanket given the crater bowl diameter (Kadish and Barlow, 2006). These distinctive pedestal morphologies may be a function of either the erosion of the pedestal craters or their substrate and are very different than their higher latitude counterparts (compare Fig. 3, top, to Fig. 3, bottom). This morphology is found both in large MFF pedestal craters (such as those shown in Fig. 3) and in a class of small pedestal craters (100 m in diameter) which exhibit an irregular sea-urchin-like morphology (Fig. 4). The effect of the substrate on crater morphology can be seen best in Fig. 5, where a crater formed on the edge of an MFF deposit. The western half of the crater displays a lobate crater morphology commonly seen on Mars, while the eastern half has long, thin, perched rays of material characteristic of MFF pedestals. The unique morphologies of MFF craters may thus be used as a diagnostic tool to determine whether the MFF was present at the time that a crater was formed.

Fig. 7. Crater degradation in the MFF. THEMIS IR mosaic of Memnonia Sulci (southwestern Lucus Planum) showing craters in various stages of the inversion process. Letter A indicates a crater that is still embedded in its original layer. Letter B indicates pedestal craters whose original substrate has been removed. Letter C indicates a crater which has been completely inverted and now stands as a mesa. The exclusion of these craters in a crater-derived age would produce an erroneously young age for the deposit.

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

The presence of pedestal craters in the MFF suggests a history of aeolian erosion and the removal and redistribution of a thickness of material at least up to the level of the pedestal heights (which average 115 m in height (Kadish et al., 2009a), with maximum heights up to 2 km (Schultz, 2007)) in and around the formation (Schultz and Lutz, 1988). Any smaller craters whose imprint into the substrate is shallower than this layer of removed material would be erased completely as the surrounding layer was removed (e.g. Schultz, 2007). For example, secondary craters from 10.1 km Zunil crater, while abundant on adjacent terrain, were absent in MFF, leading Preblich et al. (2007) to estimate that some parts of the MFF are eroding by at least 0.08 m per year. This erosion would have two effects: it would make counts of smaller craters unreliable for age-dating and it would skew the crater size–frequency data towards larger crater sizes. Over time, the relatively resistant inner deposits of pedestal craters are also eroded. This often occurs from the distal edges of the craters, where unarmored materials become exposed to the wind as the plains are deflated. This erosional process tends to leave the resistant bowl of the crater perched above the surrounding plain. In time, this most resistant part of the crater is also degraded and eroded, leaving only a remnant knob. This progression was described by Schultz (2002, 2006) and is seen throughout the MFF (Fig. 6). While small craters erode to form remnant knobs, larger craters erode to leave pedestal craters and finally remnant mesas (Fig. 7). The resulting knob-like features are distinct from other

675

knobby terrains, rootless cones, pingos, and mounds in that they are generally circular (or slightly elongated with wind direction), free-standing, non-overlapping, and lacking moats. Confident identification of the most eroded inverted craters is difficult in the absence of other nearby pedestal craters as context. If many of the knobs and mesas in the MFF are indeed modified and inverted impact craters (Schultz and Lutz, 1988), the number of craters superposed on the deposit would increase dramatically,

Fig. 9. Complex stratigraphic relationships in Gordii Dorsum. MFF yardangs are both covered by and superposed on lavas from Arsia Mons. These stratigraphic relationships make the MFF difficult to date through traditional means. THEMIS VIS image (V12699008).

Fig. 8. Stratigraphic relationships. Medusae Fossae Formation yardangs both overlie (CTX image P02_001791_1852) and are embayed by (CTX image P07_003756_1822) Late Amazonian Cerberus Aec3 lavas. Top, the northwestern tip of Zephyria Planum. Bottom, northeastern Aeolis Planum.

Fig. 10. Redistribution of MFF. This large pedestal crater, preserving an outlying section of the MFF in northern Zephyria Planum below its ejecta deposit, is embayed by the young Cerberus lavas at (A). Outlying MFF sheds unconsolidated fine particles on top of the Cerberus lavas after their emplacement (B). The unconsolidated material is transported by the wind into adjacent terrain, where it forms blankets, wind streaks, and new yardangs that lie on top of young units (C).

676

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

suggesting that the deposit is older than the dates obtained by counting recognizable superposed craters only. Furthermore, even if crater counts included these knobs and mesas (inverted craters), this would still yield a minimum age, because the crater counts would be biased toward a young age by the final erasure of knobs and mesas (Schultz and Lutz, 1988). The observed processes of crater modification, burial, exhumation and destruction will produce not a formation age, but a cumulative exposure age (or crater retention age) due to repeated burial and shielding of units

interrupted by periods of exhumation (Schultz and Lutz, 1988; Schultz, 2002). Both of these factors will result in the underestimation of the formation age of the Medusae Fossae Formation (e.g., Greeley et al., 2001; Schultz, 2006). The cratering record of the MFF thus has three signatures: (1) superposed unmodified fresh craters, (2) remnant craters (knobs, mesas, and pedestal craters) remaining after the removal of material by aeolian processes, and (3) preserved impact craters exhumed from older surfaces below. As a result, crater counts of either the MFF or the underlying terrain will yield ages that are anomalously young, since craters in the MFF were removed and the underlying terrain was shielded by the MFF during a significant part of its existence (Schultz, 2002). These observations and relationships suggest that any age for the MFF derived from crater counts will be a minimum age for its formation and that it will represent a modification age, not a formation age, as pointed out by Schultz and Lutz (1988). In summary, the characteristics of the subunits of the MFF are unsuited to be age dated, using the crater size–frequency distribution method, and other methods of dating must be considered. An alternative technique is to use stratigraphic relationships between the formation and surrounding units, such as lava flows, which may themselves be better suited for deriving crater age dates

Fig. 11. A roughness map created by Kreslavsky and Head (2000). Darker areas denote smoother surfaces; brighter areas denote rougher surface. The young Cerberus lavas (A, black) comprise the smoothest unit in the study region. There is a trough (B) between the MFF of Aeolis Planum and the adjacent Hesperian lavas (C). The smooth Cerberus lavas embay the Hesperian lavas and fill in the trough, flowing in through its southeast end (D) (see Fig. 12, extent outlined in dotted line). The image was created as an RGB composite. Each channel corresponds to a roughness length scale, in this case 9.6, 2.4, and 0.6 km. Lat.: 12°S–21°N, Lon.: 137° to 176°E.

Fig. 12. The progression of erosion and embayment in northeast Aeolis Planum. The MFF boundary originally lay at the northeast margin of the trough. The MFF was embayed by Hesperian lavas. The MFF then eroded towards the southwest. An impact at B armored the MFF. Further erosion caused continued recession of the MFF toward the southwest. In the Late Amazonian Cerberus lavas flooded and embayed the area from the northeast. A channel (A) is present where the Amazonian Cerberus lavas enter the trough. The Amazonian lavas embay the pedestal crater and yardangs at B (Fig. 13). Serrated lava margins characterize the northeast wall of the trough at C (Fig. 16D). The MFF can be seen eroding from the Hesperian lava contact at D (Fig. 15). THEMIS IR mosaic of greater northeastern Aeolis Planum overlaid with MOLA gridded data (red = high, blue = low). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 13. The contact between the Medusae Fossae Formation and the young Cerberus lavas in Aeolis Planum (142.8°E, 2.8°N). The lavas embay the formation and bury the yardangs, which can be seen faintly beneath the thin veneer. Several inverted craters can be seen in the frame; two smaller craters have been transformed into remnant knobs and the larger crater in the lower left has been eroded to leave a circular mesa. These relationships indicate that the MFF predates the embaying Late Amazonian Cerberus lavas (AEc3, from Tanaka et al. (2005)). CTX image (P05_002833_1819).

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

due to their more coherent and less erodible nature. We now investigate regional and local stratigraphic relationships within and adjacent to the MFF. 2.2. Stratigraphy 2.2.1. Relationships with lava units Because individual lava units are commonly emplaced in a geologically short period of time, cease to evolve after emplacement, and are relatively resistant to erosion, they are useful age markers in relation to adjacent overlying and underlying units, such as the Medusae Fossae Formation, which are more difficult to date. However, close examination of the MFF by previous workers yields ambiguous and sometimes contradictory stratigraphic relationships with nearby lava units. For example, early descriptions noted that lowland Amazonian lava flows embay the MFF in places (Scott and Tanaka, 1982). Indeed, the Amazonian-aged Cerberus plains lavas in Elysium Planum embay the MFF at Aeolis Planum (Fig. 8). Later works have placed the MFF at the top of the

Fig. 14. Yardang-molds and lava flow casts. Top: the MFF is eroded into yardangs by wind coming from the northwest, creating ‘‘molds”. Middle: the yardangs are embayed by lavas flowing from the northwest, filling the molds. Bottom: subsequent aeolian erosion of the MFF from the contact leaves serrated lava margins (casts).

677

stratigraphic column because it overlies lowland Amazonian lavas (Bradley and Sakimoto, 2001). This is also correct; for example, the same Amazonian-aged Cerberus lavas are overlain by the MFF in northern Eumenides Dorsum and Zephyria Planum (Fig. 8). In one particularly striking example, a lava unit near southwestern Gordii Dorsum (Fig. 9), both superposes and is superposed by MFF yardangs in close proximity. These ambiguous stratigraphic relationships lead to two possible conclusions: first, that the entire MFF was not formed at the same time, leading to different ages for different parts of the formation, and second, that the location and margins of the MFF have not remained static or constant over time. According to analysis of the data surveyed in this study, both of these conclusions are likely to be true. In favor of the first (the emplacement of the MFF at different times), is the observation that there are many sequential layers of yardangs with different orientations (Bradley et al., 2002; Zimbelman and Griffin, 2009), implying successive periods of emplacement and erosion. In support of the second idea (that the deposits and its margins have not remained constant over time), is the erosional state of the MFF, and the evidence that its eroded material is often deposited on top of adjacent units. For example, in northern Zephyria Planum, there are seemingly contradictory juxtaposed stratigraphic relationships (Fig. 10). Here, Amazonian Cerberus lavas embay a large pedestal crater that originally formed in MFF material (using the criteria discussed in Fig. 5), but MFF slightly to the east of this crater is superposed

Fig. 15. MFF eroding from lava cast boundary. The MFF continues to erode away from the margin of Hesperian unit HBu2 (Tanaka et al., 2005). Light colored finegrained material collects as dunes on top of the younger embaying Cerberus lavas. This eroded material is now stratigraphically above the Cerberus lavas. Because of this cycle of erosion and deposition, the MFF often appears younger than adjacent units. HRSC image h4136.

678

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

on the young lava surface. In context, it may be seen that the material eroded from the ejecta deposit of the crater has likely been transported to the east in the form of loose material, creating wind streaks and dunes. This loose, unconsolidated material rests on top of the same Cerberus lavas that embay the large pedestal crater. This is not because the original emplacement of the MFF postdates the plains, but rather because the MFF has been eroded, modified, and redistributed subsequent to the emplacement of the lavas. Much of the MFF that lies on top of the lavas at this contact consists of dunes, suggesting the lateral aeolian transport of previously emplaced MFF, rather than primary formational emplacement. Even the local presence of superposed yardangs does not require that this part of the deposit was formed by primary deposition, as a large percentage of the yardangs on Earth are composed of indurated sediments and could thus be secondary, not primary (Ward, 1979). The part of the MFF that today rests on top of the Cerberus flows was likely transported there by the wind and indurated, later eroding to form yardangs. Further to the south, below the region of unconsolidated wind-streak material, the Cerberus lavas once again embay the MFF. We thus conclude that the boundaries of the Medusae Fossae Formation as they appear today are a combination of original boundaries defining where the MFF was initially emplaced and new boundaries where the formation was eroded back or re-emplaced as a result of aeolian processes. We conclude on the basis of our analyses (Figs. 2–10) that the MFF is very mobile, initially forming and then eroding from one place to be re-deposited in another within the region. In order to constrain the history of the formation and modification of the MFF, new tools must be developed to glean more information from the obvious contacts and to explore and understand hidden or ambiguous contacts. To this end, we have studied lava flow-front morphology (likely to be more well preserved than the friable MFF) near unit contacts to establish whether the lava flows interacted with the Medusae Fossae Formation at the time of lava emplacement. This technique can be used to establish stratigraphic relationships in places where there is no direct contact between the MFF and nearby lava flows due to the MFF’s erosion.

However, there is an unusual lava-flow margin that composes the northeast wall of the trough into which the young Cerberus lavas poured in northwest Aeolis Planum (Fig. 12C). While normally a

2.3. Three type locations: Aeolis Planum, Amazonis Planitia, and Apollinaris Patera 2.3.1. Aeolis Planum First, we study the relationship between the MFF and the Cerberus lavas at Aeolis Planum. The roughness map compiled by Kreslavsky and Head (2000) (Fig. 11) provides evidence to reconstruct the history of this area. The young lava flows from Cerberus constitute some of the smoothest units in the region, a combination of their low viscosity (Jaeger et al., 2007) and young age (Late Amazonian according to Tanaka et al., 2005). The roughness map reveals the location where the Cerberus lavas flows fill a trough. This trough lies between the current MFF boundary and the adjacent Hesperian-aged lavas (mapped as unit HBu2 in Tanaka et al. (2005)). With the addition of MOLA gridded altimetry data (Fig. 12), a channel can be seen where the new lava entered the trough at its southeast margin and filled it. In this area, the Cerberus flows embayed and buried MFF yardangs, as evidenced by their subdued topography still visible below the veneer of lava. Fig. 13 shows in detail the relationship between the Amazonian lavas and the MFF. Inverted craters and mesas are also embayed by the lavas, indicating that the MFF previously extended further to the east than it does at the present. This evidence suggests that the MFF is older than the Late Amazonian Cerberus lavas, which is consistent with recent mapping (Tanaka et al., 2005).

Fig. 16. Observed formation progression for yardang cast formation, from top to bottom: (A) a field of yardangs is formed, (B) the yardangs are embayed by lava (the mold is filled), (C) the yardangs erode away from the boundary through time, (D) the resulting lava-cast morphology (casts). (A) P22_00909_1674, (B and D) P05_002833_1819, (C) h2965.

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

lava flow forms a lobate pattern and thins at its margins, the lava at the edge of the trough forms a cliff face punctuated by serrated, zigzagging ridges. Given the proximity of the MFF and the similarity to yardang morphology, we interpret the serrated edge of the older lava flows as indicating that the lava flows embayed the MFF yardangs during their emplacement (see Fig. 14). The jagged MFF yardangs served as a ‘‘mold” into which the embaying lavas were ‘‘poured”. Subsequent to the emplacement of the lavas, the MFF was further eroded towards the southwest, leaving the steep-sided, serrated ‘‘lava casts”. In this scenario, the cliff that forms the edge of the trough was created by the abutment of lava against the side of the MFF deposits. This shape would remain after the erosion of the adjacent MFF, as shown diagrammatically in Fig. 14. Indeed, a closeup of the trough wall (Fig. 15) shows a section of the lava margin where the Medusae Fossae material continues to be eroded from the lava contact, even while being embayed by the younger, trough-filling lavas. During the emplacement of the Hesperian lava flows (HBu2), the lavas were deflected around the former edges of the Aeolis Planum MFF. This critical relationship shows that the MFF was present and undergoing erosion and yardang formation in the Hesperian, prior to the emplacement of HBu2. The serrated edges of MFF yardangs provide a unique boundary that is diagnostic of the MFF throughout the region where the MFF occurs. As fluid lavas embay the MFF, they fill the spaces between adjacent yardangs in a manner similar to the filling of a mold. When the MFF is later eroded away by the wind, the solidified and durable lavas remain, just as a cast remains after its mold has been removed. The result is a lava flow cast whose edge mimics the serrated pattern of the former yardangs. Yardang lava casts and true yardangs can be distinguished in that the upper surface of lava flow casts is a smooth flat plane. In contrast, the top surface of a yardang is shaped like the overturned hull of a ship and faceted by the wind. The temporal progression of yardang-mold and lavacast morphology is shown in Fig. 16. 2.3.2. Amazonis Planitia and northern Memnonia Planitia We describe two more examples of cast-and-mold relationships in the context of the general geological unit boundaries in the area

679

Fig. 18. Northwest Gordii Dorsum. A region in northwest Gordii Dorsum demonstrates a multi-staged history of emplacement and erosion. First a crater is formed in older plains (A). Second, the MFF is emplaced (B), shielding the old plains from further impacts. Third, newer lavas (C) embay the Medusae Fossae Formation and the old crater (A), surrounding outlying outcrops of MFF and embaying yardangs at the edge of the deposit. Fourth, the MFF is eroded away, leaving pedestal craters, inverted craters, yardangs, and lava casts (D). The MFF can still be seen retreating from the contact (E). The cross-hatched pattern in the eroding MFF suggest that more recent winds have been bidirectional, both aligned with yardang casts and (perhaps more recently) oblique to them. Indurated parts of the formation remain (F), and troughs surround these outliers, indicating the extent of the MFF at the time of lava emplacement (G). Secondaries from crater (A) are visible where old volcanic terrain is exhumed (H). HRSC image h2965.

(Scott and Tanaka, 1986) (Fig. 17). Near the northwestern tip of Gordii Dorsum (Fig. 17 Box A; Fig. 18), the full erosional progression of the MFF from embayed yardangs to free-standing lava casts

Fig. 17. Geological context for Amazonis region. Each color represents a separate unit. Units Amm and Amu represent the middle and upper members of the MFF; Aa1 and Aa3 are Amazonian lavas; Aoa1 is the Olympus Mons aureole; AHt3 is an Amazonian–Hesperian lava unit from Arsia Mons; Hch are Hesperian channels; Npl1 and Nplf are ancient Noachian terrains. Box A (Fig. 18) shows the boundary between the Amm unit of Gordii Dorsum with Amazonian lavas. Box B (Fig. 19) shows the boundary between the Amm of Eumenides Dorsum with the Amazonian–Hesperian lavas from Arsia. Geological map from Scott and Tanaka (1986), overlaid on the THEMIS IR mosaic. Lat.: 6°S–10°N, Lon.: 160° to 140°E.

680

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

can be found. Initially, a 10-km crater was formed (Fig. 18A). The crater is characterized by a common, lobate crater morphology (as described in Fig. 5) and does not show the unique MFF crater morphology described earlier (Figs. 3–5). This indicates that the crater did not impact into MFF material. Second, the Medusae Fossae Formation was emplaced across the terrain, also filling the crater. Third, the MFF eroded away, its edge moving southeast (bottom right of Fig. 18), leaving some of the deposit stranded in the crater where it was sheltered from the wind. Fourth, flood lavas (Fig. 18C) embayed the edge of MFF (Fig. 18D and G), which was made up of yardangs, pedestal craters, and outlying mesas (Fig. 18E and F). Fifth, the MFF continued to erode away from the contact, leaving lava casts at lava flow-front edges, moats around pedestal craters (which are eroding radially inwards), and hollows where mesas once stood. During the latest activity of the MFF, it underwent further erosion, leaving inverted craters to the east and stripping away the last of the yardangs that still remained at the contact (Fig. 18E). The embaying lava unit (Fig. 18C) appears to be part of long lava lobes that are flowing from south to north, perhaps from the Tharsis region (Fuller and Head, 2002). These lavas are thought to be middle to late Amazonian in age. Thus, the

embayment of the MFF by these lavas does not impose stringent time constraints for the initial formation of the MFF. It does, however, indicate that the MFF has continued to be actively eroded before, during, and after the emplacement of the middle to late Amazonian lava flows (Fuller and Head, 2002; Tanaka et al., 2005). Further to the southeast, lobes of lava, originating from the Arsia Mons region as part of the Amazonian–Hesperian AHt3 unit (Scott and Tanaka, 1986), flow between the ancient Noachian terrain to the south and Amazonis Planum, coming to a stop near an outcropping of the MFF that makes up part of Eumenides Dorsum (Fig. 17 Box B; Fig. 19). Most of the flows have lobate margins and flow-front patterns, typical of lava flows and different from the lava-yardang casts and molds. However, where the flows approach the outcropping of MFF, the flow edges are characterized by the serrated, yardang-like edges seen in Aeolis Planum and Gordii Dorsum (Figs. 16D and 18). The MFF is exposed on the opposite side of a small depression. This example shows the distinctive difference in flow-front morphology displayed by adjacent flows: serrated near the outcrop of MFF, and lobate away from the MFF. It is likely therefore that the MFF formerly abutted the lava flows, causing the lava to flow between the MFF yardangs. The yardangs were then eroded away, leaving the interfingering lava ‘‘cast” pattern that is distinct from normal flow-fronts. This relationship suggests that the MFF was present and being actively eroded to produce yardangs at the time when the Arsia lavas were emplaced, placing this eroded part of the MFF (the middle member according to Scott and Tanaka (1986)) in the early Amazonian, late Hesperian, or earlier. 2.3.3. Apollinaris Patera Apollinaris Patera is a volcano approximately 200 km in diameter, which shows evidence of explosive eruptions (Robinson et al., 1993) (Fig. 20). Scott et al. (1993) classified Apollinaris as Hesperian in their map of the volcano and the surrounding region. Its earliest flows were extruded in the early Hesperian or late Noachian and the latest flows (associated with the southeastern fan) in the mid-Hesperian. Crater size–frequency distribution ages by Werner (2009) suggest that the active period of Apollinaris ceased around 3.71 Ga. The Apollinaris edifice is associated with a magnetic anomaly, indicating that its construction preceded

Fig. 19. Yardang-molds and lava casts between Eumenides Dorsum and Amazonis Mensae. Lavas coming from Arsia Mons (from the bottom right of the image, Lava Flow 2, unit AHt3 from Scott and Tanaka, 1986) stop short of an outcropping of the MFF. Lobate lava flow-fronts are indicated with white arrows (blue boundaries). This flow-front morphology indicates natural, unimpeded lava flow. This morphology sharply contrasts with the serrated lava casts indicated by black arrows (and red boundaries in lower panel). The presence of these casts indicates that the MFF has retreated from this unit boundary to its present position to the northwest. THEMIS Daytime IR mosaic. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 20. A regional view of Apollinaris Patera. The volcano (early Hesperian unit Ha1) is embayed by the MFF (unit Amm) at A (Fig. 21). In the southeast (B), the volcano’s younger fan material (mid-Hesperian unit Ha4) embays the lower MFF member (unit Aml) (Fig. 22). This suggests an emplacement of the lower member of the MFF in the mid-Hesperian or before and an emplacement of the middle member of the MFF in the early Hesperian or later. Map units from Scott et al. (1993). HRSC images h1009, h0998, h0987, h0024, h0335.

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

681

the end of the martian dynamo (Langlais and Purucker, 2007). The paleopole responsible for the magnetic anomaly at Apollinaris Patera appears to be close to the current rotational pole, suggesting that polar wander subsequent to the construction of Apollinaris was minimal (Langlais and Purucker, 2007). There are two main contacts between the MFF and the Apollinaris edifice. One contact is on the northern side of the edifice, where the unit Amm (the middle member of the MFF) comes into contact with the main Apollinaris edifice. The other is on the southeastern flank of Apollinaris, where the large fan of the volcano comes into contact with the unit Aml (the lower member of the MFF). Scott and Tanaka (1982) observed that while the volcano is embayed by MFF unit Amm on the northern flank, the younger Apollinaris fan flows overlap MFF unit Aml in the southwest. However, later maps (e.g. Scott et al., 1993) placed the MFF unit Aml stratigraphically above the southwestern Apollinaris fan. Although

Fig. 22. Southeastern contact between the MFF and Apollinaris. The southeastern contact between the Apollinaris Patera fan (mid-Hesperian unit Ha4) and the Medusae Fossae Formation (unit Aml). The fan material embays and buries the yardangs of the MFF (arrows). MOC image (m2100205). Map units from Scott et al. (1993).

the contacts between Apollinaris Patera edifice and the MFF are somewhat complex, with evidence for multiple episodes of emplacement, the edifice clearly underlies MFF yardangs (unit Amm) along its northern margin (Fig. 21). At the contact between the MFF and the southeastern Apollinaris fan, the fan material appears to bury and embay yardangs associated with MFF unit Aml (Fig. 22). This embayment relationship supports the original conclusions of Scott and Tanaka (1982) and suggests that the MFF (Aml according to Scott and Tanaka, 1986) was already present and being eroded during the construction of the Apollinaris fan, thus placing emplacement of the formation in the Hesperian or earlier. A summary of the stratigraphic constraints provided by lava units is shown in Fig. 23. The relationships that we have documented and discussed are consistent with initial emplacement of the MFF beginning in the mid-Hesperian (Kerber et al., 2009) or perhaps even earlier. Recognition of yardang-mold and lava-cast morphology in areas where the MFF is present has shown that the MFF is an active and dynamic unit that has undergone largescale erosion, causing migration of its borders and re-deposition of eroded material in adjacent areas. A diagram of the major processes at work within the MFF is shown in Fig. 24. Analysis of the cast-and-mold structures can lead to improved knowledge of stratigraphic relationships between the MFF and adjacent units which are otherwise ambiguous. The relationships that we have documented and discussed are consistent with initial emplacement of the MFF beginning in the mid-Hesperian or perhaps even earlier. It is plausible that the bulk of the MFF was deposited at this time and continuously modified since then. However, because of the ambiguous stratigraphic relations caused by erosion and redeposition of the MFF, we cannot rule out the possibility that some parts of the MFF may have been deposited episodically over an extended period of time, with perhaps some primary emplacement taking place into the Amazonian. More detailed analysis of stratigraphic relationships, combined with assessment of the ages of exhumed craters, may help to resolve this outstanding question. Fig. 21. Northern contact between the MFF and Apollinaris (A). In this case, the MFF (unit Amm) superposes the Apollinaris edifice (early Hesperian unit Ha1) (A, B). An older MFF unit with degraded yardangs with a different orientation is also visible, indicating that there may have been multiple episodes of MFF emplacement (C). CTX images (P02_001645_1726, P04_002634_1707). Map units from Scott et al. (1993).

3. Conclusions Previous studies have proposed that the Medusae Fossae Formation is Amazonian in age on the basis of impact crater size–frequency distribution data and stratigraphy (e.g., Scott and Tanaka,

682

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

Fig. 23. A synthesis of present and ancient stratigraphic relationships with adjacent lava units as discussed in the text. The oldest stratigraphic relationship in each MFF region indicates an upper limit for the primary deposition of that part of the formation. Supplementary images are included to illustrate stratigraphic relationships not included in other figures (Zephyria Planum: h2176; Lucus Planum: h2980). Aec3: Amazonian Cerberus lavas; HBu2: Late Hesperian lava plains; Aha: Apollinaris Patera; AHt3: Amazonian/Hesperian lava flows from Arsia Mons; AAa2s: late to middle Amazonian lavas from Amazonis Planitia. Map units from Scott and Tanaka (1986), Greeley and Guest (1987), and Tanaka et al. (2005).

1982, 1986; Greeley and Guest, 1987; Werner, 2005). After examining the characteristics of the Medusae Fossae Formation, we come to several conclusions. First, we conclude that the unit is friable and easily eroded (confirming and supporting previous interpretations). Second, we find that there are three classes of impact craters associated with the MFF: relatively fresh, heavily modified (pedestal craters, knobs, and mesas), and exhumed. The nature of this crater population suggests that any attempt to date the surface with impact crater size–frequency distributions (particularly with smaller crater class sizes) will produce minimum (and incorrect) formation ages. Ages derived by this method will be modification ages, not unit formation ages, as discussed by Schultz and Lutz (1988). Third, examination of stratigraphic relationships with associated and adjacent stratigraphic units, particularly interlayered lava flows, shows that the MFF is very mobile and has constantly been eroded to produce yardangs and transported laterally onto younger stratigraphic units. Lava flows have flooded and embayed MFF yardangs, creating yardang-cast and lava-mold structures.

Continued erosion and retreat of yardangs have left isolated lava casts that are further testament to the dynamic nature of the MFF as a geological unit. In summary, on the basis of our analyses, we conclude that the Medusae Fossae Formation is an evolving geological unit, undergoing seemingly continuous erosion and re-deposition since its initial emplacement. On the basis of stratigraphic relationships with lava flows and Apollinaris Patera, the initial formation age of the MFF is most likely to have been in the Hesperian. Amazonian ages assigned to the MFF appear to be modification ages, not formation ages. Currently unknown is whether the MFF has undergone constant and continuous modification since its initial formation in the Hesperian, or whether erosion and modification processes have been episodic. Also unknown is whether there have been some additional primary contributions to the MFF since its initial emplacement in the Hesperian. The source and mode of emplacement for the MFF remain unknown, but initial deposition in the Hesperian allows for the possibility that explosive eruptions from

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

683

Fig. 24. A schematic drawing of ongoing Medusae Fossae processes.

Hesperian sources (such as centrally located Apollinaris Patera) may have contributed to the MFF (Kerber et al., 2009). Acknowledgments We gratefully acknowledge the financial support from the National Aeronautics and Space Administration from grants from the Mars Data Analysis Program (MDAP) (NNG04GJ99G), the Applied Information Systems Research (AISR) Program (NNX08AC63G), and the Mars Express High Resolution Stereo Camera Guest Investigator Program (JPL 1237163). We also thank Nadine Barlow and Rebecca Williams for helpful reviews, and C.I. Fassett and L.M. Garber for fruitful discussions. References Barlow, N.G., 2006. Impact craters in the northern hemisphere of Mars: Layered ejecta and central pit characteristics. Meteorit. Planet. Sci. 41, 1425–1436. Barlow, N.G., 1993. Increased depth–diameter ratios in the Medusae Fossae Formation deposits of Mars. Lunar Planet. Sci. XXIV, 61–62 (abstract). Bradley, B.A., Sakimoto, S.E.H., Frey, H., Zimbelman, J.R., 2002. Medusae Fossae Formation: New perspectives from Mars Global Surveyor. J. Geophys. Res. 107, E8. Bradley, B.A., Sakimoto, S.E.H., 2001. Relationships between the Medusae Fossae Formation (MFF), fluvial channels, and the dichotomy boundary southeast of Nicholson Crater, Mars. Lunar Planet. Sci. XXXII. Abstract 1335. Burr, D.M., Williams, R.M.E., Nussbaumer, J., Zimbelman, J.R., 2006. Multiple, distinct, (glacio?) fluvial paleochannels throughout the western Medusae Fossae Formation, Mars. Lunar Planet. Sci. XXXVII. Abstract 1367. Burr, D.M., Enga, M.T., Williams, R.M.E., Zimbelman, J.R., Howard, A.D., Brennard, T.A., 2009. Pervasive aqueous paleoflow features in the Aeolis/Zephyria Plana region, Mars. Icarus 200, 52–76. Butler, B.J., 1994. The 3.5-cm Radar Investigation of Mars and Mercury: Planetological Implications. PhD Dissertation, California Institute of Technology, Pasadena, California. Carter, L.M., and 11 colleagues, 2009. Shallow radar (SHARAD) sounding observations of the Medusae Fossae Formation, Mars. Icarus 199, 295–302. Dickson, J.L., Fassett, C.I., Head, J.W., 2009. Amazonian-aged fluvial valley systems in a climatic microenvironment on Mars: Melting of ice deposits on the interior of Lyot Crater. Geophys. Res. Lett. 36, L08201. Edgett, K.S., Butler, B.J., Zimbelman, J.R., Hamilton, V.E., 1997. Geologic context of the Mars radar ‘‘Stealth” region in southwestern Tharsis. J. Geophys. Res. 102 (E9), 21545–21567. Fassett, C.I., Dickson, J.L., Head, J.W., 2009. Small, young fluvial features in icy terrains on Mars. Lunar Planet. Sci. XL. Abstract 1185. Fuller, E.R., Head, J.W., 2002. Amazonis Planitia: The role of geologically recent volcanism and sedimentation in the formation of the smoothest plains on Mars. J. Geophys. Res. 107, E10. Greeley, R., Guest, J., 1987. Geologic map of the eastern equatorial region of Mars. USGS Misc. Inv. Series Map I-1802-B.

Greeley, R., Kuzmin, R.O., Haberle, R.M., 2001. Aeolian processes and their effects on understanding the chronology of Mars. Space Sci. Rev. 96 (1–4), 393–404. Griffin, L.J., Zimbelman, J.R., 2009. Geologic mapping of western Medusae Fossae Formation, Mars (MC 23-NW): Redefining unit boundaries and features to reveal a history of tectonism, wind erosion, and episodic water flow. Lunar Planet. Sci. XL. Abstract 1196. Grimm, R.E., Solomon, S.C., 1986. Tectonic tests of proposed polar wander paths for Mars and the Moon. Icarus 65, 101–121. Hartmann, W.K., Neukum, G., 2001. Cratering chronology and the evolution of Mars. Space Sci. Rev. 96 (1–4), 165–194. Head, J.W., Kreslavsky, M., 2004. Medusae Fossae Formation: Ice-rich airborne dust deposited during periods of high obliquity? Lunar Planet. Sci. XXXV. Abstract 1635. Hynek, B.M., Phillips, R.J., Arvidson, R.E., 2003. Explosive volcanism in the Tharsis region: Global evidence in the martian geologic record. J. Geophys. Res. 108, E9. Inbar, M., Risso, C., 2001. Holocene yardangs in volcanic terrains in the Southern Andes, Argentina. Earth Surf. Proc. 26, 657–666. Jaeger, W.L., Keszthelyi, L.P., McEwen, A.S., Dundas, C.M., Russell, P.S., 2007. Athabasca Valles, Mars: A lava-draped channel system. Science 317, 1709– 1711. Kadish, S.J., Barlow, N.G., 2006. Pedestal crater distribution and implications for a new model of formation. Lunar Planet. Sci. XXXVII. 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. XXXIX. Abstract 1766. Kadish, S.J., Barlow, N.G., Head, J.W., 2009a. Latitude dependence of martian pedestal craters: Evidence for a sublimation-driven formation mechanism. J. Geophys. Res. 114, E10001. doi:10.1029/2008JE003318. Kadish, S.J., Head, J.W., Barlow, N.G., 2009b. Mid-latitude pedestal crater heights: A proxy for the thickness of a past climate-related, ice-rich substrate. Lunar Planet. Sci. XL. Abstract 1313. Kerber, L., Head, J.W., Madeleine, J.B., Forget, F., Wilson, L., 2009. The dispersal of Pyroclasts from Apollinaris Patera, Mars. Lunar Planet. Sci. XL. Abstract 2176. Kreslavsky, M., Head, J.W., 2000. Kilometer-scale roughness of Mars: Results from MOLA data analysis. J. Geophys. Res. 105 (E11), 26695–26711. Langlais, B., Purucker, M., 2007. A polar magnetic paleopole associated with Apollinaris Patera, Mars. Planet. Space Sci. 155, 270–279. Livingstone, I., Warren, A., 1996. Aeolian Geomorphology: An Introduction. Addison Wesley Longman Limited, Singapore, pp. 32–34. Mandt, K.E., de Silva, S., Zimbelman, J.R., Crown, D.A., 2007. A synoptic approach to evaluating the origin of the Medusae Fossae Formation, Mars. Lunar Planet. Sci. XXXVIII. Abstract 1823. Mandt, K.E., de Silva, S.L., Zimbelman, J.R., Crown, D.A., 2008. Origin of the Medusae Fossae Formation, Mars: Insights from a synoptic approach. J. Geophys. Res. 113, E12011. doi:10.1029/2008JE003076. Mouginis-Mark, P.J., 1979. Martian fluidized crater morphology – Variations with crater size, latitude, and target material. J. Geophys. Res. 84, 8011–8022. Mouginis-Mark, P.J., 2002. Prodigious ash deposits near the summit of Arsia Mons, Mars. Lunar Planet. Sci. XXXIII. Abstract 1407. Muhleman, D.O., Butler, B.J., Grossman, A.W., Slade, M.A., 1991. Radar images of Mars. Science 253, 1508–1513. Parker, T.J., 1991. A comparison of the martian Medusae Fossae Formation with terrestrial carbonate platforms. Lunar Planet. Sci. XXII. Abstract 1029. Preblich, B.S., McEwen, A.S., Studer, D.M., 2007. Mapping rays and secondary craters from the martian crater Zunil. J. Geophys. Res. 112, E05006. doi:10.1029/ 2006JE002817.

684

L. Kerber, J.W. Head / Icarus 206 (2010) 669–684

Robinson, M.S., Mouginis-Mark, P.J., Zimbelman, J.R., Wu, S.S.C., Ablin, K.K., Howington-Kraus, A.E., 1993. Chronology, eruption duration, and atmospheric contribution of the martian volcano Apollinaris Patera. Icarus 104, 301–323. Sakimoto, S.E.H., 1999. Topography, roughness, layering, and slope properties of the Medusae Fossae Formation from Mars Orbiter Laser Altimeter (MOLA) and Mars Orbiter Camera (MOC) data. J. Geophys. Res. 104 (E10), 22154–24141. Schultz, P.H., 2002. Uncovering Mars. Lunar Planet. Sci. XXXIII. Abstract 1790. Schultz, P.H., 2006. Uncovering Mars. Planetary Chronology Workshop. Abstract 6024. Schultz, P.H., 2007. Hidden Mars. Science 318, 1080–1081. Schultz, P.H., Lutz, A.B., 1988. Polar wandering of Mars. Icarus 73, 91–141. Scott, D.H., Dohm, J.M., Applebee, D.J., 1993. Geological map of science study area 8, Apollinaris Patera region of Mars. USGS Misc. Invest. Ser. Map MTM-10186. Scott, D.H., Tanaka, K.L., 1982. Ignimbrites of Amazonis Planitia region of Mars. J. Geophys. Res. 87 (B2), 1179–1190. Scott, D.H., Tanaka, K.L., 1986. Geologic map of the western equatorial region of Mars. USGS Misc. Invest. Ser. Map I-1802-A. Tanaka, K.L., 2000. Dust and ice deposition in the martian geologic record. Icarus 144 (2), 254–266. Tanaka, K.L., Skinner Jr., J.A., Hare, T.M., 2005. Geologic map of the Northern Plains of Mars. USGS Sci. Invest. Map 2888. Ward, A.W., 1979. Yardangs on Mars: Evidence of recent wind erosion. J. Geophys. Res. 84, 8147–8165.

Watters, T.R., and 12 colleagues, 2007. Radar sounding of the Medusae Fossae Formation Mars: Equatorial ice or dry, low-density deposits? Science 318, 1125–1128. Werner, S.C., 2005. Major Aspects of the Chronostratigraphy and Geologic Evolutionary History of Mars. PhD Dissertation. Fachbereich Geowissens chaften Freie Universität, Berlin. Werner, S.C., 2009. The global martian volcanic evolutionary history. Icarus 201, 44–68. Wilson, L., Head, J.W., 1994. Mars: Review and analysis of volcanic eruption theory and relationships to observed landforms. Rev. Geophys. 32, 221–263. Wrobel, K., Schultz, P.H., Crawford, D., 2006. An atmospheric blast/thermal model for the formation of high-latitude pedestal craters. Meteorit. Planet. Sci. 41, 1539–1550. Zimbelman, J.R., Griffin, L.J., 2009. HiRISE images of yardangs and sinuous ridges in the lower member of the Medusae Fossae Formation, Mars. Icarus 205, 198– 210. Zimbelman, J.R., Johnston, A.K., Lovett, C.G., 1996. Geologic map of Medusae Fossae Formation within MTM Quadrangle 05142 on Mars. Geol. Soc. Am. Meet. 23 (7), A128 (abstract). Zimbelman, J.R., Crown, D.A., Grant, J.A., Hooper, D.M., 1997. The Medusae Fossae Formation, Amazonis Planitia, Mars: Evolution of proposed hypotheses of origin. Lunar Planet. Sci. XXVIII. Abstract 1482. Zimbelman, J.R., Sakimoto, S.E.H., Frey, H., 2000. Evidence for a fluvial contribution to the complex story of the Medusae Fossae Formation on Mars. Geol. Soc. of Am. Meet. Abstract 50423.