Origin and composition of three heterolithic boulder- and cobble-bearing deposits overlying the Murray and Stimson formations, Gale Crater, Mars

Abstract

Heterolithic, boulder-containing, pebble-strewn surfaces occur along the lower slopes of Aeolis Mons ("Mt. Sharp") in Gale crater, Mars. They were observed in HiRISE images acquired from orbit prior to the landing of the Curiosity rover. The rover was used to investigate three of these units named Blackfoot, Brandberg, and Bimbe between sols 1099 and 1410. These unconsolidated units overlie the lower Murray formation that forms the base of Mt. Sharp, and consist of pebbles, cobbles and boulders. Blackfoot also overlies portions of the Stimson formation, which consists of eolian sandstone that is understood to significantly postdate the dominantly lacustrine deposition of the Murray formation. Blackfoot is elliptical in shape (62 × 26 m), while Brandberg is nearly circular (50 × 55 m), and Bimbe is irregular in shape, covering about ten times the area of the other two. The largest boulders are 1.5-2.5 m in size and are interpreted to be sandstones. As seen from orbit, some boulders are light-toned and others are dark-toned. Rover-based observations show that both have the same gray appearance from the ground and their apparently different albedos in orbital observations result from relatively flat sky-facing surfaces. Chemical observations show that two clasts of fine sandstone at Bimbe have similar compositions and morphologies to nine ChemCam targets observed early in the mission, near Yellowknife Bay, including the Bathurst Inlet outcrop, and to at least one target (Pyramid Hills, Sol 692) and possibly a cap rock unit just north of Hidden Valley, locations that are several kilometers apart in distance and tens of meters in elevation. These findings may suggest the earlier existence of draping strata, like the Stimson formation, that would have overlain the current surface from Bimbe to Yellowknife Bay. Compositionally these extinct strata could be related to the Siccar Point group to which the Stimson formation belongs. Dark, massive sandstone blocks at Bimbe are chemically distinct from blocks of similar morphology at Bradbury Rise, except for a single float block, Oscar (Sol 516). Conglomerates observed along a low, sinuous ridge at Bimbe consist of matrix and clasts with compositions similar to the Stimson formation, suggesting that stream beds likely existed nearly contemporaneously with the dunes that eventually formed the Stimson formation, or that they had the same source material. In either case, they represent a later pulse of fluvial activity relative to the lakes associated with the Murray formation. These three units may be local remnants of infilled impact craters (especially circular-shaped Brandberg), decayed buttes, patches of unconsolidated fluvial deposits, or residual mass-movement debris. Their incorporation of Stimson and Murray rocks, the lack of lithification, and appearance of being erosional remnants suggest that they record erosion and deposition events that post-date the exposure of the Stimson formation.

Keywords: Curiosity rover; Gale crater; Greenheugh pediment; Heterolithic unit; Murray formation; Stimson formation.

Fig. 1

The Curiosity rover field site occurs on the lower northwest slope of the 5-km-high Aeolis Mons (Mt. Sharp), a mountain of predominantly stratified sediments in Gale crater. Gale is a ~154 km diameter impact structure. The white trace indicates the rover traverse from landing until September 2019. The inset shows the location of the field site within Gale crater. The three heterolithic units studied by the rover are indicated by filled circles. Waypoints are indicated by black circles. Targets mentioned in the text that are outside of the heterolithic units are indicated by triangles. Other features mentioned in the text are also indicated.

Fig. 2

(a) The Curiosity rover and the Blackfoot heterolithic unit (within blue dashed outline) as imaged by HiRISE on 04 September 2015. Yellow trace indicates rover traverse, from north toward southwest. Locations of boulders A and B and the sites of two rover stops on Blackfoot are indicated. Locations of ChemCam targets Sunburst, Swan, Lincoln, Jefferson, Madison, and the APXS/MAHLI target Badlands, are indicated in green. (b) Boulder A is dark-toned when viewed from above; the rock is a dark gray sandstone and it has a surface that is rough at a centimeter scale. (c) Boulder B is light-toned when viewed from above (e.g. HiRISE images); the rock is a dark-gray sandstone; bedding is apparent; the skyward-facing surface is smooth at a centimeter scale and coated with eolian dust. (d) Blackfoot as seen from the north on Sol 1094 when the rover was parked at the position shown in (a). Boulders A and B and the locations of (e), (f), and (g) are indicated. (e) The north and east margins of the Blackfoot heterolithic unit overlie dark-gray, cross-bedded eolian Stimson-formation sandstones. (f) North edge of Blackfoot reveals the deposit in cross-section; here, it consists of a jumble of angular cobbles and pebbles; cobbles display a variety of orientations relative to their internal sedimentary structures (bedding). Dark-toned eolian sand has filled some of the gaps between stones. (g) Many of the open fractures which cut through Stimson-formation sandstones exhibit a light-toned, fracture-parallel, altered zone (a “halo;” see Frydenvang et al., 2017 and Yen et al., 2017). Here a fracture and associated halo follow the red trace and are abruptly truncated where covered by the Blackfoot heterolithic unit; this indicates that the fracture and fluids responsible for the “halo” alteration did not penetrate the Blackfoot deposit; i.e., one of the indicators that the deposit is not a lithic unit. (h) At its west-southwest end, the Blackfoot unit directly overlies mudstones of the Murray formation. (i) The cobbles and boulders of Blackfoot display a range of orientations, relative to their internal structure (bedding). Some of them protrude from the surface, others rest on the surface. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 3

(a) Typical Blackfoot surface, with scattered cobbles, small boulders, pebbles, and sand of various orientations (relative to internal sedimentary structure, such as bedding), colors, and lithologies. Two key examples are highlighted: A cross-stratified sandstone boulder (ChemCam target named Sunburst) and a conglomeratic cobble. (b) Dark-gray sandstone target named Sunburst. White and light-gray sand grains are readily distinguished because of their contrast relative to the majority of grains, which are dark gray. (c) Example of a light-gray conglomeratic cobble, protruding from the surface. The orange-brown material on part of this stone is eolian dust. (d) Pebbles in an otherwise light-gray sandy matrix in a conglomeratic stone protruding from the surface at Blackfoot. (e) ChemCam target Madison; inset shows the image location on the rock face. The large, light-toned feature consists of the remains of a fracture wall vein composed of calcium sulfate. More importantly, the image shows light-toned sand grains, which contrast with the bulk of dark-toned sand grains (much more difficult to see) in this rock. Madison is a dark-gray sandstone. (f) APXS/MAHLI target, Badlands. This stone, too, is a dark-gray sandstone. Unlike Madison and Sunburst, Badlands has no light-toned sand grains; they are all dark gray. The sizes of two larger grains are indicated; most grains are smaller and difficult to distinguish from each other and their intergranular cement. Brownish-orange patches are depressions (some of them, perhaps, may be sockets from which sand grains have been removed) containing eolian dust. (g) Two examples of the light-toned sandstone or pebbly conglomerate fragments which occur at Blackfoot. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 4

(a) Brandberg and the Curiosity rover as viewed by HiRISE on Sol 1159. Yellow trace indicates the rover traverse; drive direction was from northeast to southwest. ChemCam targets are indicated by name. (b) Eastern Brandberg. In this scene, Brandberg is downhill from the rover; the near side of Brandberg forms a shallow depression. Rover tracks provide scale. Eolian bedforms explain the faint lineated texture observable in HiRISE images. (c) Brandberg as viewed from the northeast. (d) Examples of steeply dipping cobble-sized angular clasts in Brandberg. (e) Examples of clast variety in northeastern Brandberg. The largest boulder-sized clasts dip toward the center of the circular landform. Boulders with relatively flat, skyward-facing surfaces are coated with dust and thus appear light-toned when seen from above in HiRISE images. The ChemCam target, Gibeon, is indicated. (f) Red/cyan stereo pair anaglyph produced from two HiRISE images; the dashed yellow arc indicates a ridge that might be an indicator that Brandberg occurs within the eroded remains of an impact structure. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 5

(a) Sol 1160 view of the variety of boulders, cobbles, and pebbles which compose the Brandberg deposit. Outlines indicate locations of panels b, c, d, g, h, and i. (b) Dust-coated, fractured Murray-formation lithic clast target named Hoba. (c) Dark-gray sandstone cobble target named Gibeon. The inset shows enlargement of a single grain or concretion, representative of the larger grains (or typical concretion size) in this rock, ~2 mm. (d) Sandstone with concretions. (e) Example sandstone with concretions (target named Maiberg) liberated from Stimson-formation bedrock on the east side of Naukluft Plateau, imaged on Sol 1277; shown here for comparison. (f) Sedimentary structure and texture details of Maiberg. (g) Examples of candidate, rare conglomeratic stones at Brandberg. (h) Example of angular, cobble-sized fragment of cross-bedded sandstone. (i) Example lithic fragment of Murray-formation bedrock incorporated into the Brandberg deposit; white arrows indicate erosion-resistant concretions; compare with (j): Example of intact Murray-formation bedrock south of Brandberg which contains similar concretions (white arrows).

Fig. 6

(a) Local context of Bimbe on the lower slopes of Mt. Sharp, between the Murray buttes to the south, Naukluft Plateau and some of the Bagnold dunes to the east, and Baynes Mountain to the north. The white trace is the Curiosity rover traverse; the drive direction was from northeast (upper right) toward the south (bottom left of center). Blue box indicates the location of (b): Bimbe with ChemCam, APXS, and MAHLI targets indicated in yellow. Green dots indicate ChemCam targets, red dots indicate APXS targets, and black dots indicate rover parking spots. The inset, a 3× expanded view of the area inside the white circle next to target Auchab, shows an example of light- and dark-toned boulders as observed from the orbiting HiRISE camera. The white trace is the Curiosity rover traverse; the drive direction was from the northeast (upper right) toward the south (bottom center). The elevation contours are derived from HiRISE stereo-pair products calibrated to Mars Global Surveyor (MGS) Mars Orbiter Laser Altimeter (MOLA), topography by Parker and Calef (2016). (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 7

A ridge on the northeast side of Bimbe, which trends northeast-southwest (azimuth ~65.5°), is mantled with cobbles and small boulders in Mastcam images (inset). The ridge resembles nearby boulder- and cobble-mantled hills named Bukalo and Bailundo (insets). These are likely examples of buttes that have weathered to the point that they are hills with remnants of the capping rock taking the form of small boulders and cobbles (cf. Migon et al., 2018).

Fig. 8

Stereo pair anaglyph of Bimbe (large, dark-toned patch) and its vicinity. Most of the boulders observed at Bimbe occur on the elevated portions of the south margin of Bimbe. The feature within the dashed ellipse, indicated with a “?,” might be the eroded remains of an impact structure. Definitive impact craters are labeled with “c” and probable impact structures are labeled with “i”. Dark-toned eolian sand, expressed as bedforms, has collected in the impact structure depressions. Features labeled “MB” are a couple of the Murray buttes. (Anaglyph will only appear in 3D on the web version of this article.)

Fig. 9

a) Work site for sols 1405–1410, viewed from the east. Names of targets investigated using ChemCam, Mastcam, APXS, and MAHLI are indicated. The dashed trace approximately indicates the crest of a low ridge line that defines the southern margins of Bimbe. (b) Navcam view of the work site. ChemCam, APXS, and MAHLI targets are indicated (except Mariental, which is to the left of this scene). The stone labeled “angular sandstone,” imaged by MAHLI on Sol 1407, is also shown in Fig. 15 and was not a named target. For scale, the width of the rover wheel (right of lower center) is 40 cm.

Fig. 10

(a) Boulder containing targets Sonneblom (ChemCam, APXS, MAHLI) and Zambezi (APXS, MAHLI). The boulder is a dark-gray, pitted sandstone; the locations of the MAHLI views in (b) and (c) are indicated. (b and c) Close-up views of the surface; the arrows point to several individual grains.

Fig. 11

(a) Context for two ChemCam targets, Seeis and AEGIS_1406a, on two separate stones at Bimbe. (b) Seeis target; outline indicates location of ChemCam RMI coverage, 4× inset indicates examples of the largest (mm-scale) grains. (c) AEGIS_1406a target; outline indicates the location of the ChemCam RMI coverage shown to the right. (d) Close-up view, with crosshairs indicating LIBS observation points on AEGIS_1406a.

Fig. 12

(a) Dark-gray sandstone cobble target named Oranjemund. White box indicates location of expanded view in (b); outline indicates ChemCam RMI coverage in (c). (b) View showing grain-scale bedding, which runs diagonally from upper left to lower right. (c) Close-up view of Oranjemund; inset shows 3× expanded view with measurements on features that could be among the larger sand grains in the rock (i.e., most of the grains are smaller than ~400 μm). Oranjemund is compositionally identical to layered target Chinchimane (Fig. 14).

Fig. 13

(a) Local context for ChemCam target Auchab; yellow outline indicates location of panel (b): closer view of knobby gray target Auchab; yellow outline indicates ChemCam RMI coverage in (c). Note the nearby angular, cross-bedded sandstone clast. Both stones are surrounded by windblown sand and overlie reddish Murray-formation bedrock. (c) ChemCam RMI mosaic showing the nodular texture of Auchab. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 14

(a) Local context for layered gray sandstone cobble target Chinchimane. Note the differences in cobble orientation relative to their bedding structure, as well as the diversity of cobble colors, response to sunlight, and sedimentary structure and texture. (b) Mastcam-100 view. Bedding is approximately parallel and thin (sand grain scale thickness). A fracture filled with a while material (vein) cuts across bedding. The yellow outline indicates the location of the ChemCam RMI coverage shown in (c), where a few individual sand grains can be identified in a well cemented and well sorted sandstone. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 15

Angular sandstone cobble with fine-grained interbeds. (a) Location (white arrow) of the cobble relative to the larger boulders in the rover workspace during sols 1405–1410. (b) Highest spatial resolution view of the cobble. Coarser grains are ~700 μm; the skyward rock face cuts steeply across the original bedding. A recessed interval of ~1 cm thickness consists of grains too small to resolve.

Fig. 16

(a) Example showing diversity of cobble and boulder orientations, sedimentary structures, and colors at Bimbe. The white circles indicate examples of white stones shown in (c); the yellow circles indicate examples of “red” stones in (c). (b) Knobby sandstone target, Canico. (c) Examples of “red” (left) and white (center and right) cobbles which are minor constituents of the Bimbe deposit; locations and relative sizes are shown in (a). The red stone at the lower left in (c) has a white vein cutting across bedding within the stone. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 17

(a) Conglomeratic boulder containing MAHLI targets Tumba and Funda; the latter was also an APXS target. (b) Conglomeratic boulder, ~60 cm wide, that includes veins that cut across the sandy matrix (inset). (c) Conglomeratic boulder that includes the ChemCam target Balombo (outline indicates RMI image coverage).

Fig. 18

(a) Conglomeratic boulder investigated via four ChemCam targets. (b–e) Respective RMI views and LIBS target areas of Seeheim, Bungo, Cabamba, and Wilhelmstal. Note the vitreous luster of Wilhelmstal, and similarity to Cabamba.

Fig. 19

(a) Conglomeratic boulder that includes the APXS/MAHLI target, Funda, and the MAHLI target, Tumba. White arrows indicate recessed white objects similar to Funda; blue arrows indicate light-gray and white protrusive pebble clasts. (b) Close-up view of white, recessed, banded object (clast or void fill) comprising the target Funda. (c) Pebble clast target named Tumba. (d) Close-up view of a portion of the Tumba pebble clast, showing that it is a sandstone; arrows point to a few example grains. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 20

(a) Conglomeratic boulder with ChemCam target, Mariental. Inset shows the entire boulder. (b) Close-up view of Mariental and the locations of ChemCam LIBS observations. Recessive white features are interpreted to be similar to Funda (Fig. 19).

Fig. 21

Major-element abundances, in wt%, of ChemCam targets at Bimbe. Larger symbols indicate averages of individual targets for Aussenkehr, Lucala, Seeis, Sonneblom, AEGIS_post_1406a (“Massive,” red dots); Wilhelmstal, Cabamba (“Vitreous Conglomerate,” blue diamonds, also includes AEGIS_post_1400a); Chinchimane, Oranjemund (“Bimbe Layered,” yellow dots), Auchab and Canico (“Nodular SS,” brown dots). Smaller symbols represent individual observation points for the conglomerates (“Conglomerate Pts”), which have more diverse compositions. Circles indicate groupings of Bimbe layered (yellow), massive (red), and vitreous conglomerate (blue) targets. Standard deviations between individual LIBS observation points within a target are given in Table 2. Upper limits on precision for individual point observations are shown in each panel, taken from 480 Sheepbed-formation measurements (Mangold et al., 2015). Also shown are the mean compositions, and standard deviations of the means, of other types of targets observed nearby and earlier along the traverse. These include Murray, Mars soil, two groups of conglomerates, ChemCam target Pyramid Hills, and a group of massive targets from Bradbury rise. Stimson-formation compositions are shown as contours where each contour represents equal weighting in terms of density of samples (see text and Supplementary material). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 22

Rocks observed on Bradbury rise that match Bimbe float rocks in terms of morphology and composition in some cases, and only in morphology in other cases. (a) Bathurst_Inlet, observed Sol 55, and (b) Nullataktok (Sol 336), found in the same area, have similar compositions and morphologies to the Bimbe layered clasts. Bradbury massive targets (c) Bull Arm and (d) Jake_M do not match the compositions of the Bimbe massive targets, although they bear morphological resemblances. (e) Oscar (observed Sol 516) does match the Bimbe massive target compositions. (f) Pyramid Hills (observed north of Hidden Valley on Sol 692) is a close match to the Bimbe layered compositions, as well as to the Bradbury layered targets, e.g., (a) and (b). Locations of these targets are shown on Fig. 1.

Fig. 23

ChemCam relative reflectance spectra of representative samples of the Bimbe layered and nodular rocks, along with sample spectra from the Bradbury layered rocks. The Bimbe rocks show either flat near-infrared spectra or downturns that begin near 600 nm, whereas the Bradbury rocks have peak reflectance wavelengths near 650–675 nm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 24

Possible scenarios for formation of the heterolithic units. Each panel shows three time steps. The gray stippled top layer is the Stimson formation, the white bedded unit is the Murray, and the orange records the heterolithic unit. (a) Impact penetrates through the Stimson formation into the Murray formation, creating a crater whose walls erosionally retreat through time. (b) Incision into a resistant Stimson cap rock (either by wind or water) that creates buttes and sheds Stimson and Murray blocks into a narrow valley. Further erosional retreat leaves isolated patches. Alternately, buttes decay in place to create the heterolithic units. Sediments are protected from weathering in the original location of the incision, or boulders shed from the mesas are collected in certain areas that are sheltered from wind erosion. (c) Mass transport of debris, either dry or by glacial processes, followed by extensive erosion that obscures the original source region. (d) Large-scale fluvial or debris flow fan deposits followed by erosion. (e) Localized fluvial or debris flow followed by erosion.