Late Hesperian to early Amazonian midlatitude Martian valleys: Evidence from Newton and Gorgonum basins

Transcription

1 JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 116,, doi: /2010je003782, 2011 Late Hesperian to early Amazonian midlatitude Martian valleys: Evidence from Newton and Gorgonum basins Alan D. Howard 1 and Jeffrey M. Moore 2 Received 23 November 2010; revised 15 February 2011; accepted 24 February 2011; published 14 May [1] Numerous shallowly incised valleys extend from the upper interior rims of Newton and Gorgonum basins across smooth interior basin deposits. These valleys are a few meters to 300 m wide and may have experienced discharges over all or most of their width, implying that they are, in fact, incised channels. In Newton Crater these valleys extend up to 75 km, to near the center of the basin floor. In Gorgonum basin the valleys terminate at what we interpret to be a former ice covered lake. These valleys appear to be examples of scattered, shallowly incised valleys found throughout the midlatitudes of Mars superimposed upon the widespread mantling deposits within the region. On the basis of crater count age dating, the interior valleys in Newton and Gorgonum basins were formed at about the Hesperian to Amazonian transition. The runoff through the valleys may have occurred due to episodic melting of snow and ice that had accumulated on the crater rims. Temperatures warm enough to cause extensive melting may have occurred during optimal orbital and obliquity configurations because of intensive volcanism releasing greenhouse gasses or as a result of a brief episode of warming from a large impact somewhere on Mars. The valleys were formed at about the same time as major outflow channels were active along the highlands lowlands boundary. Water delivered to the northern lowlands by the outflow channels may have been recycled as snow and ice deposits within the Martian midlatitudes. The episodic melting of such deposits may have formed the midlatitude valleys. Citation: Howard, A. D., and J. M. Moore (2011), Late Hesperian to early Amazonian midlatitude Martian valleys: Evidence from Newton and Gorgonum basins, J. Geophys. Res., 116,, doi: /2010je Introduction [2] The midlatitudes of Mars feature distinctive landforms, including mantling deposits, glacial and periglacial landforms, Amazonian age gullies on steep slopes, and sparse, shallowly incised, fresh appearing valleys that are the focus of this study [e.g., Dickson et al., 2009; Fassett et al., 2010]. These midlatitude valleys (MLVs) are distinct from the older, late Noachian to early Hesperian valley systems which are deeply dissected, are generally of much larger spatial extent, and are more degraded [Howard et al., 2005; Irwin et al.,2005; Fassett and Head, 2008;Hynek et al., 2010]. Although some MLVs involve rejuvenation of older valley networks, many MLVs are eroded into smooth or rolling slopes and intercrater terrain. The MLVs range from a few meters to more than 300 m 1 Department of Environmental Sciences, University of Virginia, Charlottesville, Virginia, USA. 2 Space Sciences Division, NASA Ames Research Center, Moffett Field, California, USA. Copyright 2011 by the American Geophysical Union /11/2010JE in width, with nearly parallel valley walls and planforms that are locally sinuous. [3] Particularly well presented MLVs occupy parts of the floors of the 300 km Newton (41 S 202 E) and the 240 km Gorgonum (37 S 189 E) basins (Figures 1 4). We examine the morphology of these valleys, determine age relationships using crater counting, and discuss the local and global environmental conditions that may have formed these valleys. 2. Basin and Valley Morphology 2.1. Newton Basin [4] The MLVs in Newton basin (Figure 2) are atypically long for post Noachian valleys seen elsewhere. Most valleys (red in Figure 3) source on the steep slopes of the upper interior crater rim and flow up to 75 km across the smooth basin interior almost to the basin center. Segments of incised valleys alternate with smooth, inferred alluvial deposits on lower gradient parts of the basin floor (purple in Figure 3, examples in Figures 5 7). Some of the alluvial portions exhibit braided fluvial textures or distributaries (Figure 6, pink in Figure 3). Some of the mapped braided textures are indistinct (Figure 8a) but are mapped as depositional fluvial plains because they connect to incised valley segments and they define a smooth surface. 1of10

2 contour (Figures 10a and 10b). A distinctive low ridge follows the 0 m contour around much of the basin floor (Figure 10b). Figure 1. Shaded relief map of a portion of the Martian southern cratered highlands from 150 E to 210 E and 15 S to 50 S, identifying Newton Crater (N), Gorgonum basin (G), and Ma adim Valles (M). 3. Age Relationships 3.1. Relative Ages [8] MLVs in Newton postdate formation of the smooth surface of the basin interior and most of the large craters superimposed on the smooth deposits. The upper rims of Newton and Gorgonum craters exhibit incised, locally dendritic valleys that we infer to have formed during the late Noachian to Early Hesperian episode of widespread valley incision on Mars. Unlike the MLVs, these valleys do not extend downward onto the smooth basin interior deposits. Some of these valleys appear to have been reactivated during the later formation of the MLVs. The MLVs postdate wrinkle ridge formation on the floor of Newton because they are deflected by the ridges. In Gorgonum basin, MLVs are [5] MLV valley floors appear to be nearly flat, sometimes exhibiting faint lineations. These features suggest that the MLVs in many cases are incised channels that were occupied at least intermittently by flows over the entire valley bottom. In locations where the valleys exhibit a sinuous pattern (e.g., Figures 2, 5, and 6), the ratio of the bottom width to meander wavelength averages about 7.5. This is similar to the wavelength/width ratio in terrestrial meandering channels [e.g., Leopold and Wolman, 1957] and likewise suggests flows occupying the entire valley width. We retain, however, the conservative designation of these features as valleys rather than as channels. Deposits with a planar surface (Figure 8b) occupy the lowest part of the western Newton floor and part of the floor of the crater on Newton s north rim (green in Figure 3). [6] MLV density is highest on the steep slopes of the northern and western basin floor. Their absence in eastern basin might be due to extensive aeolian activity in that location that could have buried or eroded preexisting features. Alternatively it could be due to uneven spatial distribution of regional precipitation Gorgonum Basin [7] Gorgonum basin also exhibits MLVs (red in Figure 4), but these are limited to the upper interior slopes above the 0 m contour relative to the Martian datum. Flat topped benches at about 300 m (Figure 9 and situated between the pink and green lines in Figure 4) exhibit a wandering scarp front at the basin interior edge, dropping sharply to the basin center at about 400 m [Howard and Moore, 2004]. Fields of kilometer scale, light colored knobs [Moore and Howard, 2003] occupy portions of the basin floor, which is typical of basin centers in this region (excepting Newton). Knobs interior of the 0 m contour appear blanketed by dark toned deposits and individual knobs often appear fractured (Figures 9 and 10a). In addition, fields of small knobs (Figure 10a, yellow in Figure 3) may be remnants of the larger knobs that have been fractured and perhaps translocated. Impact craters that straddle the 0 m contour exhibit eroded ejecta interior of the Figure 2. Fresh appearing, shallowly incised valleys (MLVs) on the floor of Newton basin. Smooth areas between and upstream from incised sections are probably alluvial deposits where the valleys cross low gradient portions of the basin floor (see Figure 5). Part of CTX image P11_005376_1389. Image centered at S and E. 2of10

3 Figure 3. Late Hesperian to Early Amazonian fluvial features in Newton basin. Red, incised MLV channels; purple, inferred fluvial deposits; pink, fluvial deposits with bed forms; green, possible lacustrine sediment. Contour interval 100 m. Map centered at S, E. Blue boxes with numbers indicate footprints of figures. superimposed on smooth basin floor deposits but occur only above the 0 m contour. The deposits interior of the 0 m contour bury and thus postdate Sirenum Fossae grabens. In summary, the MLV formation and formation of Gorgonum features below 0 m occurred late in Martian history. Crater counting provides further constraints on the age of the MLVs Crater Count Ages [9] The interior of Newton basin is fully covered by 6m/ pixel CTX images, permitting counting of craters larger than a few tens of meters. In presenting the results or our crater count dating, we use the noncumulative, variable diameter bin size method of Hartmann [2005] and the absolute ages and geologic boundary positions presented in that paper. We establish the upper age of the MLV system in Newton Crater by dating the smooth interior deposits of the basin into which the valleys are incised. Two crater counts were undertaken and are presented in Figure 11a. A count of craters larger than 0.1 km in diameter ( Newton Interior ) covers the entire extent of the smooth basin deposits in Newton Crater using a 20 m/pixel base map assembled from downsampled CTX camera images. The western margin of the Newton floor deposits is relatively unaffected by aeolian modification, and we utilized a 5 m/pixel mosaic to count craters as small as 0.01 km diameter ( CTX Count ). By both counts, craters greater than 1 km in diameter indicate a basin floor age between the Noachian Hesperian boundary and the Early to Mid Hesperian boundary. Craters less than 1 km in diameter show a decrease in apparent age with decreasing size, including a downturn in the Hartmann frequency plot for craters less than about 0.2 km in diameter (Figure 11a). This decrease in age is consistent with continuing crater degradation that is most effective at eliminating small craters. [10] Using the buffered counting technique of Fassett and Head [2008], we estimated the age of about 1100 km of valleys/channels on the interior of Newton basin using 5 m/pixel CTX images. We separated these counts into two classes, the first including all craters within the size dependent buffer radius as defined by Fassett and Head [2008] ( all craters in Figure 8b) and a second class ( unambiguous craters in Figure 11b) that includes only those craters showing a clear superposition relationship to the valleys. Both classes indicate a late Hesperian to early Amazonian MLV age based on superimposed craters >1 km in diameter. As with the basin floor crater counts, the apparent age of craters <1 km in size decreases with size due to degradation processes. [11] In the crater counting of the smooth deposits on the floor of Newton Crater Craters from CTX images we noted all craters less than 4 km in diameter on the floor of Newton Crater that exhibit well defined ejecta or sharply defined 3of10

4 Figure 4. Map of fluvial and hypothesized lacustrine features of Gorgonum basin. Red, MLV valleys; blue, hypothesized shoreline at about 0 m contour; pink, inner edge of flat topped benches at about 300 m contour; green, outer (interior basin facing) edge of flat topped benches; yellow, fields of small, 100 m knobs found only interior to hypothesized shoreline. Map centered at S, E. Contour interval 100 m. Brown boxes with numbers indicate footprints of figures. rims. These relatively unmodified craters suggest an effective age at about mid to early Amazonian (Figure 11c, fresh craters ) as compared to the mid to early Hesperian age indicated by larger craters. This again indicates active crater degradation processes at these latitudes extending to at least this later time period, which is consistent with the post Noachian mantling processes acting in the Martian midlatitudes [Soderblom et al., 1973; Mustard et al., 2001; Berman et al., 2009]. Crater counts on the benches and interior floor of Gorgonum basin likewise suggests an early Amazonian age (Figure 11a, Gorgonum Interior ) with strong degradation of small craters. This age is approximately equivalent to the Newton basin MLVs. 4. Discussion and Interpretation [12] The smooth interior deposits of Newton and Gorgonum basins date to about the Noachian Hesperian boundary 4of10

5 wrinkle ridge (the formation of which could have triggered the flow) and its location in the center of the basin where the sediment infill is presumably thickest. The offset of the dome from the basin center suggests that it is not a central peak. The deposition and presumed erosion of the kilometer scale knobs in Gorgonum basin appear to predate at least the last episode of deep lakes in the basin because the smooth basin floor deposits onlap and partially or wholly mantle the knobs at elevations above the 0 km contour. [13] During the late Hesperian to early Amazonian, fluvial activity briefly returned to the Newton Gorgonum region in the form of MLV incision. Most of the MLVs within Newton and Gorgonum basins are sourced from the upper basin rims. A shallow lake may have briefly occupied the Newton basin below 1200 m depositing the sediment indicated in green in Figure 3 (also see Figure 8b). The unusual geomorphic features at elevations below 0 m in Gorgonum basin have been interpreted as indicative of occupation by an ice covered lake [Howard and Moore, 2004]. Ice in this lake may have been up to 300 m thick, with the convolute, flat topped benches at Figure 5. Examples of alluvial connections between reaches of incised MLV valleys. Fan like forms at the end of incised valleys occasionally exhibit distributaries, A, and are interpreted as alluvial fans. Some incised valleys connect to flat topped benches, B, that we interpret to be channel beds inverted by aeolian erosion. Other incised valleys are connected by wispy linear features, C, often lineated and with slight positive or negative relief that we interpret to be alluvial channel segments crossing lower gradient portions of the basin floor. The inferred alluvial connections are not everywhere as clear as in this illustration, but they display smooth, nearly planar surfaces between incised valley segments. Part of CTX image P06_003530_1404. Image centered at E and S. See Figure 3 for relative location. as do most of the classic highland valley networks [Fassett and Head, 2008; Hynek et al., 2010]. Valley networks on the upper interior rims of Newton and Gorgonum terminate at these smooth deposits at about the 0 m contour in Newton and the +1 km level in Gorgonum. The abruptness of valley network terminations at these levels and the smooth interior deposits we interpret to indicate that these basins were occupied by deep lakes at the time of major fluvial activity. Gorgonum and other large, interconnected midlatitude lakes around the 180 E longitude range (e.g., Atlantis and Ariadnes basins) may have overflowed to deepen Ma adim Valles [Irwin et al., 2002, 2004]. Subsequently these lake deposits became desiccated and deformed by wrinkle ridge formation and formation of the Sirenum Fossae grabens. The 1 km high dome in the center of the Newton basin floor (Figure 3) straddles a north south striking prominent wrinkle ridge. We suggest that this dome may be a diapiric structure deep within the lake sediments. This interpretation is based on its location along the prominent Figure 6. Braided or anastomosing MLV on the floor of Newton basin. Channels dissect a wrinkle ridge oriented N S in the western half of the image. Part of CTX image P06_003253_1390. Image centered at S, E. See Figure 3 for relative location. 5of10

6 Figure 7. Examples of distinct channel patterns mapped as inferred fluvial deposits (purple) in Figure 3. (a) The ridges (arrows) continue down gradient from the braided channels in Figure 6 and are interpreted as inverted channel deposits. Part of CTX image P06_003253_1390. Image centered at E S. (b) The shallowly incised or wispy, lineated dark bands (arrows) are interpreted as alluvial channel complexes. Part of CTX image P15_006721_1379. Image centered at E 41.68S. 300 m forming at the contact between the ice and unfrozen water/brine below (Figure 9). The ice may have formed the low bench at 0 m at the lake margins (Figure 10), eroded ejecta of craters straddling this contour (Figure 10), deposited sediment on knobs interior to this contour (Figures 9 and10), and may have fractured and deformed submerged knobs (Figures 9 and 10). Because of the restriction of MLVs in Gorgonum to above the 0 m contour (Figure 4), we conclude that the hypothesized ice covered lake postdates, or more likely, was concomitant with MLV activity and that the lake was at least partially supplied from runoff through the MLVs. [14] The width and meander wavelength of the MLVs can be used to provide a rough estimate of the formative discharges assuming that the valleys were, in fact, occupied by flows spanning their width. The bottom width of the valleys in Figure 2 generally range from about 30 to 55 m. Based upon formulas of Irwin et al. [2008], gravity corrected estimates for discharges, Q, using channel width, W, (assuming sand bed channels) range from 76 to 159 m 3 /s using the Figure 8. (a) Indistinct, interconnected dark bands I connect to segments of incised valleys C. This smooth surface is interpreted to be an alluvial plain draining to the incised valleys at C. Some of the dark lines could be dust devil tracks. Part of CTX image P13_003253_1390. Image centered at E, S. (b) Level surface in positive relief at the lowest part of the western portion of Newton basin. Arrows point to the outward facing scarp. Although no distinct layering can be seen in CTX images, the location at the terminus of incised valleys, the low elevation, and the flat surface suggest that this is a playa or lacustrine deposit subsequently inverted by Aeolian erosion. Part of CTX image P06_003253_1390. Image centered at E, S. See Figure 3 for relative locations. 6of10

7 Figure 9. Flat topped benches in Gorgonum basin. Benches ring the basin at an elevation of 300 m with a complex basin facing planform, dropping steeply at their inner edge to the basin floor about 100 m below. White arrows point to examples of scarp edges. Higher hills are knobs of light toned deposits that appear to be fractured remnants of 1 2 km knobs typically occupying basin interiors in the region (black arrows). The benches have been hypothesized to be formed by soft sediment deformation beneath an ice covered lake [Howard and Moore, 2004]. Part of CTX image P19_008343_1424. Image centered at S, E. See Figure 4 for relative location. empirical relationship Q = 1.2W The meander wavelengths, l, range from about m, which from the gravity scaled relationship l 2 based upon the work by Irwin et al. [2008] suggest discharges in the m 3 /s range. Although these estimates carry a high degree of uncertainty, flows through the MLVs probably involved several tens of cumecs. [15] The sourcing of most MLVs at the upper basin interior rims indicates water was primarily derived from precipitation limited to the basin rims. We suggest that this pattern is most consistent with precipitation as snow or frosts rather than convective rainstorms. As in terrestrial mountain ranges, snow or frost could accumulate seasonally or possibly over multiyear periods during obliquities favorable for ice accumulation in the midlatitudes [e.g., Head et al., 2003; Head et al., 2005] with seasonal or episodic melting of the snow or ice deposits feeding the MLVs. The absence of obvious glacial features on the rims of these two basins suggests either that the inferred ice deposits were only tens of meters thick or were cold based but subject to episodic melting. [16] Valley systems younger than the main episode of valley network incision at the Noachian Hesperian transition [Howard et al., 2005; Irwin et al., 2005; Fassett and Head, 2008; Hynek et al., 2010] have been described in a number of settings, including the flanks of volcanoes [Gulick and Baker, 1990; Gulick, 2001; Fassett and Head, 2006, 2007], local dendritic valley networks (often now expressed as inverted channels) [Mangold et al., 2004; Quantin et al., 2005; Mangold et al., 2008; Weitz et al., 2008; Burr et al., 2009, 2010], in the interior of craters [Dickson et al., 2009], and associated with glacial features (lobate debris aprons, concentric crater fills, viscous flow features, lineated valley fills) [Fassettetal., 2010]. In addition, Irwin et al. [2005] Figure 10. Hypothesized lacustrine features in Gorgonum basin. (a) Possible shoreline (arrows) along the 0 m contour. Lake interior to the right. Note that ejecta from the crater straddling the shoreline is eroded on the lake facing (east) side. Part of CTX image P16_007262_1408. Image centered at S, E. (b) Hypothesized shoreline exhibiting a low positive relief in the Gorgonum basin (black arrow) along the 0 m contour. Lake interior to the top. Note that ejecta from the crater straddling the shoreline is eroded on the lake facing side. White arrow points to a MLV that terminates near the shoreline. The smooth surface near F may be a terminal fan deposit of the MLV at the shoreline. Part of CTX image P19_008343_1424. Image centered at S, E. See Figure 4 for relative location. 7of10

8 presents evidence that some valley networks became reactivated subsequent to the main period of incision near the Noachian Hesperian boundary, locally terminating at small deltas in basins. Channels emanating from the ejecta of a young impact crater [Morgan and Head, 2009] and small interior fans of Mojave Crater [Williams and Malin, 2008] may have been formed by water mobilized by the impact event itself. Our nonsystematic examination of THEMIS VIS Figure 11 8of10

9 ( 20 m/pixel) and CTX camera ( 6 m/pixel) images in the Martian midlatitude regions suggests that MLVs are more widespread than the published literature to date would suggest. [17] A variety of mechanisms have been suggested for producing runoff that formed post Noachian Martian valleys, including hydrothermal circulation for the volcanic channels [Gulick and Baker, 1990; Gulick, 2001], accumulation of snowpack with melting due to volcanic heating [Fassett and Head, 2006, 2007], melting of ice rich deposits due to a favorable climate at low relative elevations [Dickson et al., 2009], melting of glacial ice [Fassett et al., 2010] and the direct influence of an impact cratering event [Morgan and Head, 2009; Williams and Malin, 2008]. The dendritic MLV networks and those in Newton and Gorgonum basins appear to require at least regionally extensive precipitation, either producing direct runoff or melting of accumulated snow and ice [Mangold et al., 2004, 2008; Burr et al., 2010]. [18] Explanations for the post Noachian fluvial features that require precipitation or ice melting raise the possibility that fluvial activity during the later Hesperian and Amazonian may have may have been the consequence of one or more global episodes of enhanced precipitation. With the exception of the present study, most such fluvial features are too limited in areal extent or are too modified by subsequent degradation (e.g., the inverted dentritic fluvial systems) to afford reliable ages from crater counting of the channel systems themselves by the buffered crater counting technique [Fassett and Head, 2008]. Some bracketing of ages can be established from superposition relationships, including ages of the deposits into which the deposits are incised [e.g., Dickson et al., 2009; Fassett et al., 2010]. Fassett et al. [2010] use such techniques to establish the relative ages of the young valleys sourced from melting of glacial ice, with the results suggesting valley systems in different locations formed at distinct times during the Amazonian. [19] The occurrence of widespread MLVs suggest the possibility of their formation during one or more regional to global climatic episodes, perhaps due to melting of seasonal to long term accumulations of snow and ice. Temperatures warm enough to cause extensive melting may have occurred during optimal orbital and obliquity configurations, perhaps in conjunction with intensive volcanism releasing greenhouse gasses, or as a result of a brief episode of warming from a large impact somewhere on Mars. The location of most MLVs to the northern and western basin slopes of Newton and Gorgonum basins suggests a possible aspect control to ice accumulation or melting. [20] MLV activity occurred about at the same time as formation of the major outflow channels [e.g., Tanaka, 1986]. A possible scenario is that delivery of water to the northern lowlands provided, through evaporation and sublimation, water that temporarily accumulated in the midsouthern latitudes as widespread ice deposits whose partial melting formed the MLVs and small, dominantly ice covered lakes such as occurred in Gorgonum basin [e.g., Baker et al., 1991; Moore et al., 1995]. [21] Although the late Hesperian to Amazonian valleys discussed here occur in the midlatitudes of Mars, their recognition as distinct young features may be enhanced by their superimposition on the post Noachian midlatitude mantling deposits characteristic of these latitudes. Reactivation of the deeply incised valley networks of the Martian equatorial region, formed during the late Noachian and early Hesperian, may have occurred at the same time that MLVs were created. Recognition of the later fluvial activity, however, may be hindered by the relatively minor incision occurring during the flows forming the MLVs. Alternatively, the precipitation responsible for forming the MLVs might have been restricted to the midlatitudes and related to volatile redistribution during obliquity cycles [e.g., Head et al., 2003]. 5. Conclusions [22] Numerous shallowly incised valleys extend from the upper interior rims of Newton and Gorgonum basins and cross their smooth interior basin deposits (Figures 2 4). These valleys are a few meters to 300 m wide and may have experienced discharges over all or most of their width, implying that they are, in fact, incised channels (e.g., Figure 6). Rough estimates of discharges based upon valley width and meander wavelength suggest formative discharges of several tens of cubic meters per second. [23] In Newton Crater the MLV valleys extend up to 75 km to near the center of the basin floor (Figure 3). In Gorgonum basin the valleys terminate at what we interpret to be a former ice covered lake (Figures 4 and 9 10) [Howard and Moore, 2004]. These valleys appear to be examples of scattered, shallowly incised valleys (MLVs) found throughout the midlatitudes of Mars and which are superimposed upon the widespread mantling deposits within the region. Based upon our crater count age dating, the interior valleys in Newton and Gorgonum basins were formed at about the Hesperian to Amazonian transition. [24] The Newton and Gorgonum interior valleys originate at the upper interior rims of the basins, with few tributaries occurring with the basin interior. We speculate that the runoff through the valleys may have occurred due to episodic melting of snow and ice accumulated on the crater rims. Temperatures warm enough to cause extensive melting may have occurred during optimal orbital and obliquity configurations, perhaps Figure 11. Crater frequency diameter counts for Newton and Gorgonum basins. Counting uses the binning technique and absolute age curves of Hartmann [2005] (light gray lines with ages in years). Heavy lines show definition of geologic age boundaries for Mars [Tanaka, 1986]. (a) Newton interior, crater count based on a 20 m/pixel mosaic of CTX images covering the smooth interior deposits of Newton basin. Gorgonum interior, crater count of the Gorgonum basin benches at 300 m and smooth portions of the adjacent basin floor for limited areas covered by 5 m/pixel CTX images. CTX count, crater count from part of the Newton basin floor using 5 m/pixel CTX images. (b) Unambiguous craters, buffered crater counts [Fassett and Head, 2008] for a total of 1100 km of MLVs on the floor of Newton basin using only craters with unambiguous superposition on the MLVs. All craters, as above but including craters of uncertain superposition status. (c) Fresh craters, crater count of the same region as CTX count in Figure 11a but utilizing only fresh appearing craters. 9of10

10 due to or in conjunction with intensive volcanism releasing greenhouse gasses, or as a result of a brief episode of warming from a large impact somewhere on Mars. [25] The valleys were formed at about the same time as major outflow channels were active along the highlandslowlands boundary. Modeling by Moore et al. [1995] and Kreslavsky and Head [2002] and indicates that such flood waters would rapidly freeze and sublimate. Water delivered to the northern lowlands by the outflow channels, therefore, may have been recycled as snow and ice deposits within the Martian midlatitudes as suggested by Baker et al. [1991] and Moore et al. [1995]. The episodic melting of such deposits may have formed the shallowly incised midlatitude valleys. References Baker, V. R., R. G. Strom, V. C. Gulick, J. S. Kargel, G. Komatsu, and V. S. Kale (1991), Ancient oceans, ice sheets and the hydrological cycle on Mars, Nature, 352(6336), , doi: /352589a0. Berman, D. C., D. A. Crown, and L. F. I. Bleamaster (2009), Degradation of mid latitude craters on Mars, Icarus, 200, 77 95, doi: /j.icarus Burr, D. M., M. T. Enga, R. M. E. Williams, J. R. Zimbelman, A. D. Howard, and T. A. Brennand (2009), Pervasive aqueous paleoflow features in the Aeolis/Zephyria Plana region, Mars, Icarus, 200, 52 76, doi: /j. icarus Burr, D. M., R. M. E. Williams, K. D. Wendell, M. Chojnacki, and J. P. Emery (2010), Inverted fluvial features in the Aeolis/Zephyria Plana region, Mars: Formation mechanism and initial paleodischarge estimates, J. Geophys. Res., 115, E07011, doi: /2009je Dickson, J. L., C. I. Fassett, and J. W. Head (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, doi: /2009gl Fassett, C. I., and J. W. Head III (2006), Valleys on Hecates Tholus, Mars: Origin by basal melting of summit snowpack, Planet. Space Sci., 54(4), , doi: /j.pss Fassett, C. I., and J. W. Head (2007), Valley formation on Martian volcanoes in the Hesperian: Evidence for melting of summit snowpack, caldera lake formation, drainage and erosion on Ceraunius Tholus, Icarus, 189, , doi: /j.icarus Fassett, C. I., and J. W. I. Head (2008), Timing of Martian valley network activity: Constraints from buffered crater counting, Icarus, 195, 61 89, doi: /j.icarus Fassett, C. I., J. L. Dickson, J. W. Head, J. S. Levy, and D. R. Marchant (2010), Supraglacial and proglacial valleys on Amazonian Mars, Icarus, 208, , doi: /j.icarus Gulick, V. C. (2001), Origin of the valley networks on Mars: A hydrological perspective, Geomorphology, 37(3 4), , doi: /s x(00) Gulick, V. C., and V. R. Baker (1990), Origin and evolution of valleys on Martian volcanos, J. Geophys. Res., 95, 14,325 14,344, doi: / JB095iB09p Hartmann, W. K. (2005), Martian cratering 8: Isochron refinement and the chronology of Mars, Icarus, 174, , doi: /j.icarus Head, J. W., J. F. Mustard, M. A. Kreslavsky, R. E. Milliken, and D. R. Marchant (2003), Recent ice ages on Mars, Nature, 426, , doi: /nature Head, J.W., et al. (2005), Tropical to mid latitude snow and ice accumulation, flow and glaciation on Mars, Nature, 434, , doi: / nature Howard, A. D., and J. M. Moore (2004), Scarp bounded benches in Gorgonum Chaos, Mars: Formed beneath an ice covered lake?, Geophys. Res. Lett., 31, L01702, doi: /2003gl Howard, A. D., J. M. Moore, and R. P. Irwin III (2005), An intense terminal epoch of widespread fluvial activity on early Mars: 1. Valley network incision and associated deposits, J. Geophys. Res., 110, E12S14, doi: / 2005JE Hynek, B. M., M. Beach, and M. R. T. Hoke (2010), Updated global map of Martian valley networks and implications for climate and hydrologic processes, J. Geophys. Res., 115, E09008, doi: /2009je Irwin, R. P., III, T. A. Maxwell, A. D. Howard, R. A. Craddock, and D. A. W. Leverington (2002), A large lake basin at the head of Ma adim Vallis, Mars, Science, 296, , doi: /science Irwin, R. P., III, A. D. Howard, and T. A. Maxwell (2004), Geomorphology of Ma adim Vallis, Mars, and associated paleolake basins, J. Geophys. Res., 109, E12009, doi: /2004je Irwin, R. P., III, T. A. Maxwell, A. D. Howard, R. A. Craddock, and J. M. Moore (2005), An intense terminal epoch of widespread fluvial activity on early Mars: 2. Increased runoff and paleolake development, J. Geophys. Res., 110, E12S15, doi: /2005je Irwin, R. P., III, A. D. Howard, and R. A. Craddock (2008), Fluvial valley networks on Mars, in River Confluences, Tributaries and the Fluvial Network, edited by S. P. Rice et al., pp , John Wiley, Chichester, U. K., doi: / ch19. Kreslavsky, M. A., and J. W. Head (2002), Fate of outflow channel effluents in the northern lowlands of Mars: The Vastitas Borealis Formation as a sublimation residue from frozen ponded bodies of water, J. Geophys. Res., 107(E12), 5121, doi: /2001je Leopold, L. B., and M. G. Wolman (1957), River channel patterns: Braided, meandering and straight, U.S. Geol. Surv. Prof. Pap., 282 B, 85 pp. Mangold, N., C. Quantin, V. Ansan, C. Delacourt, and P. Allemand (2004), Evidence for precipitation on Mars from dendritic valleys in the Valles Marineris area, Science, 305, 78 81, doi: /science Mangold, N., V. Ansan, P. Masson, C. Quantin, and G. Neukum (2008), Geomorphic study of fluvial landforms on the northern Valles Marineris plateau, J. Geophys. Res., 113, E08009, doi: /2007je Moore, J. M., and A. D. Howard (2003), Ariadnes Gorgonum knob fields of north western Terra Sirenum, Mars, Lunar Planet. Sci., XXIV, Abstract Moore, J. M., G. D. Clow, W. L. Davis, V. C. Gulick, D. R. Janke, C. P. McKay, C. R. Stoker, and A. P. Zent (1995), The circum Chryse region as a possible example of a hydrologic cycle on Mars: Geologic observations and theoretical evaluation, J. Geophys. Res., 100, Morgan, G. A., and J. W. Head (2009), Sinton crater, Mars: Evidence for impact into a plateau icefield and melting to produce valley networks at the Hesperian Amazonian boundary, Icarus, 202, 39 59, doi: /j. icarus Mustard, J. F., C. D. Cooper, and M. K. Rifkin (2001), Evidence for recent climate change on Mars from the identification of youthful near surface ground ice, Nature, 412, , doi: / Quantin,C.,P.Allemand,N.Mangold,G.Dromart,andC.Delacourt (2005), Fluvial and lacustrine activity on layered deposits in Melas Chasma, Valles Marineris, Mars, J. Geophys. Res., 110, E12S19, doi: / 2005JE Soderblom, L. A., T. J. Kreidler, and H. Masursky (1973), Latitudinal distribution of a debris mantle on the Martian surface, J. Geophys. Res., 78, , doi: /jb078i020p Tanaka, K. L. (1986), The stratigraphy of Mars, Proc. 17th Lunar Planet. Sci. Conf., Part1,J. Geophys. Res., 91, E139 E158, doi: / JB091iB13p0E139. Weitz, C. M., R. E. Milliken, J. A. Grant, A. S. McEwen, R. M. E. Williams, and J. L. Bishop (2008), Light toned strata and inverted channels adjacent to Juventae and Genges chasmata, Mars, Geophys. Res. Lett., 35, L19202, doi: /2008gl Williams, R. M. E., and M. C. Malin (2008), Sub kilometer fans in Mojave Crater, Mars, Icarus, 198, , doi: /j.icarus A. D. Howard, Department of Environmental Sciences, PO Box , University of Virginia, Charlottesville, VA , USA. (ah6p@ virginia.edu) J. M. Moore, Space Sciences Division, NASA Ames Research Center, MS 245 3, Moffett Field, CA 94035, USA. 10 of 10