Earth and Planetary Science Letters 458 (2017) 357–365

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

www.elsevier.com/locate/epsl

Sedimentological evidence for a deltaic origin of the western fan

deposit in Jezero crater, Mars and implications for future exploration

Timothy A. Goudge a,b,∗, Ralph E. Milliken a, James W. Head a, John F. Mustard a, Caleb I. Fassett c

a Department of Earth, Environmental and Planetary Sciences, Brown University, Providence, RI, USAb Jackson School of Geosciences, The University of Texas at Austin, Austin, TX 78712-1722, USAc Department of Astronomy, Mount Holyoke College, South Hadley, MA, USA

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

Article history:Received 10 December 2015Received in revised form 27 April 2016Accepted 27 October 2016Available online 18 November 2016Editor: C. Sotin

Keywords:MarsJezero craterpaleolakedelta deposit

We examine the stratigraphic architecture and mineralogy of the western fan deposit in the Jezero crater paleolake on Mars to reassess whether this fan formed as a delta in a standing body of water, as opposed to by alluvial or debris flow processes. Analysis of topography and images reveals that the stratigraphically lowest layers within the fan have shallow dips (<2◦), consistent with deltaic bottomsets, whereas overlying strata exhibit steeper dips (∼2–9◦) and downlap, consistent with delta foresets. Strong clay mineral signatures (Fe/Mg-smectite) are identified in the inferred bottomsets, as would be expected in the distal fine-grained facies of a delta. We conclude that the Jezero crater western fan deposit is deltaic in origin based on the exposed stratal geometries and mineralogy, and we emphasize the importance of examining the stratigraphic architecture of sedimentary fan deposits on Mars to confidently distinguish between alluvial fans and deltas. Our results indicate that Jezero crater contains exceptionally well-preserved fluvio-deltaic stratigraphy, including strata interpreted as fine-grained deltaic bottomsets that would have had a high potential to concentrate and preserve organic matter. Future exploration of this site is both geologically and astrobiologically compelling, and in situanalyses would be complementary to the ongoing in situ characterization of fluvio-lacustrine sediment in the Gale crater paleolake basin by the Curiosity rover.

© 2016 Elsevier B.V. All rights reserved.

1. Introduction

The martian sedimentary rock record offers unique insight into ancient surface processes that have occurred on the planet by providing information on past sediment transport, deposition, and accumulation in a variety of distinct environments (e.g., Ori et al., 2000; Malin and Edgett, 2003; Fassett and Head, 2005;Howard et al., 2005; Irwin et al., 2005; Lewis and Aharonson, 2006;Di Achille and Hynek, 2010; Ansan et al., 2011; Grotzinger and Mil-liken, 2012; DiBiase et al., 2013; Grotzinger et al., 2015). Constrain-ing the origin of a given sedimentary deposit requires an under-standing of the deposit’s facies and stratigraphic architecture (i.e., the geometry of discrete bundles of exposed strata and bounding unconformities), which together record how the processes respon-sible for its formation evolved in both space and time. This record,

* Corresponding author at: Jackson School of Geosciences, The University of Texas at Austin, 2275 Speedway, Stop C9000, Austin, TX 78712-1722, USA.

E-mail address: tgoudge@jsg.utexas.edu (T.A. Goudge).

http://dx.doi.org/10.1016/j.epsl.2016.10.0560012-821X/© 2016 Elsevier B.V. All rights reserved.

in turn, can provide important information about the depositional environment of the sediment, such as the relative contributions of fluvial, submarine, and aeolian transport and deposition (e.g., Mitchum et al., 1977).

Understanding the stratigraphic architecture of terrestrial sedi-mentary deposits requires the analysis and reconstruction of three-dimensional structures within the deposit, which is commonly achieved by studying well exposed outcrops and seismic reflec-tion imaging (e.g., Mitchum et al., 1977). In a similar manner to the study of stratal geometries exposed in outcrop on Earth, high-resolution orbital topography and orthorectified images of Mars allow for detailed observations of the large-scale stratigraphic ar-chitecture of martian sedimentary deposits that have been ex-humed, exposing their three-dimensional structures (e.g., Lewis and Aharonson, 2006; Ansan et al., 2011; DiBiase et al., 2013).

Many sedimentary fan-shaped deposits have been identified on Mars (e.g., Ori et al., 2000; Malin and Edgett, 2003; Fassett and Head, 2005; Howard et al., 2005; Irwin et al., 2005; Di Achille and Hynek, 2010; DiBiase et al., 2013), including landforms interpreted as both deltas and alluvial fans. The formation of a delta deposit

358 T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365

requires deposition into a standing body of liquid water, whereas an alluvial fan forms subaerially, a distinction that implies sig-nificantly different environmental conditions. Delta deposits also have a much higher potential for concentration and preservation of organic matter in fine-grained intervals when compared with alluvial fan deposits, making them more attractive candidates for future in situ exploration of organic compounds (e.g., Summons et al., 2011).

Although modern terrestrial deltas and alluvial fans show dis-tinct differences in morphology and surface topography (e.g., Blair and McPherson, 1994), interpreting a given sedimentary fan de-posit on Mars as a delta on the basis of morphology and geologic context alone is not necessarily conclusive. Morphology and geo-logic setting are certainly important aspects to consider and can provide compelling arguments for a deltaic over alluvial fan ori-gin (e.g., Fassett and Head, 2005; Schon et al., 2012); however, the complication with relying exclusively on these characteristics is that sedimentary fan deposits on Mars represent exposed por-tions of the rock record (e.g., Grotzinger and Milliken, 2012), and not perfectly preserved depositional surfaces. Therefore, analysis of modern surface morphology and geologic setting of a fan de-posit cannot be uniquely linked to the original depositional surface or geologic setting of that landform, because such analyses often cannot account for all of the effects of potential post-depositional processes (e.g., burial, erosion, compaction, etc.).

It is thus crucial to also examine the stratigraphic architec-ture of sedimentary fan deposits on Mars in order to differentiate deltas from alluvial fans, as the different processes responsible for deposition and accumulation of strata within these depositional environments result in distinct stratal geometries (e.g., Rich, 1951;Bull, 1977; Orton and Reading, 1993; Blair and McPherson, 1994). A characteristic sedimentary structure of delta deposits is a cli-noform, which comprises two prominent slope breaks going from approximately flat-lying topset strata to steeper dipping foresets to shallowly dipping bottomsets (Mitchum et al., 1977; Rich, 1951;Orton and Reading, 1993; Pirmez et al., 1998). Deltaic clino-forms also commonly exhibit downlap, where more steeply dip-ping foresets transition to shallower dips and terminate (lapout) against underlying strata that have shallower dips (Rich, 1951;Mitchum et al., 1977; Pirmez et al., 1998). In contrast, allu-vial fans more commonly show approximately sub-parallel layer-ing and a distinct concave radial profile, indicative of their sub-aerial formation by rapidly decelerating river channels (Bull, 1977;Blair and McPherson, 1994).

These distinctions in stratigraphic architecture provide a key ba-sis upon which deltas can be distinguished from alluvial fans on Mars through studies of exposed stratal geometries using remotely sensed data. However, only in a few instances have the stratal ge-ometries of sedimentary fan deposits on Mars been closely exam-ined and shown to be consistent with a deltaic origin (e.g., Ansan et al., 2011; DiBiase et al., 2013; Grotzinger et al., 2015).

One specific site of interest is the ∼45 km diameter Jezero impact crater that is breached by two inlet valleys and an out-let valley, and once hosted an open-basin lake (Fig. 1; Fassett and Head, 2005). Jezero crater itself is located in terrain mapped as the Hesperian and Noachian transition unit (HNt) by Tanaka et al. (2014), and the fluvial activity that fed the Jezero crater pa-leolake ceased near the Noachian–Hesperian boundary based on buffered crater counting of the associated valley network system by Fassett and Head (2008). Jezero crater contains two sedimen-tary fan deposits, one at the mouth of each inlet valley (Fassett and Head, 2005), which have sampled a large portion of ancient martian crust and contain Fe/Mg-smectite and Mg-rich carbonate (Ehlmann et al., 2008a, 2008b, 2009; Goudge et al., 2015).

The Jezero fan deposits have been interpreted as deltas on the basis of their geologic setting (i.e., within a hydrologically open pa-

Fig. 1. Jezero crater paleolake basin, located at ∼18.4◦N, ∼77.7◦E (Fassett and Head, 2005). North is up in both images. (a) Overview of the Jezero crater pale-olake showing the two inlet valleys and the outlet valley (labeled thin black lines). Thick black outline is the −2395 m MOLA contour, the minimum level to which water must have ponded in the basin (Fassett and Head, 2005). Location of part (b) is indicated by white box. Red box indicates the extent of the HiRISE DEM used in this work. HRSC-derived DEM h5270_0000 overlain on a mosaic of Con-text Camera (CTX; Malin et al., 2007) images P04_002664_1988, P17_007714_2001, P15_007068_1971, P06_003521_1971, and P06_003376_1987. (b) Overview of the Jezero crater western fan deposit. Locations of Figs. 2a and 2b are indicated in white boxes. HiRISE-derived DEM from stereo-pair images PSP_003798_1985 and PSP_002387_1985 overlain on a mosaic of HiRISE image PSP_003798_1985 and CTX image P04_002664_1988. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

leolake) and exposed distributary channels (e.g., Fassett and Head, 2005; Ehlmann et al., 2008a; Schon et al., 2012); however, there has yet to be a detailed analysis of the exposed stratal geome-tries of these deposits. Here we examine the stratigraphic archi-tecture and mineralogy of the western fan deposit (Fig. 1b), which is the better preserved and exposed of the two fans (Fassett and Head, 2005; Goudge et al., 2015), to further understand the de-positional environment and relative importance of alluvial versus fluvio-deltaic processes for the formation of this deposit.

2. Methods

To explore the origin of the western fan deposit we have em-ployed two complementary approaches: (1) qualitative and quan-

T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365 359

titative analyses of the stratigraphic architecture of the fan, and (2) detailed analysis of the mineralogy of the exposed portions of the fan.

2.1. Stratigraphic architecture

Prominent sedimentary structures of the western fan deposit are exposed along the erosional front of the deposit (Fig. 2; Schon et al., 2012), and were investigated using a high-resolution digital elevation model (DEM; 1 m/pixel) and orthorectified im-age (∼0.25 m/pixel) from the High Resolution Imaging Science Experiment (HiRISE; McEwen et al., 2007). The HiRISE-derived DEM was produced from stereo-pair images PSP_003798_1985 and PSP_002387_1985 using the NASA Ames Stereo Pipeline (ASP) stereogrammetry software (Broxton and Edwards, 2008; Moratto et al., 2010; Beyer et al., 2014).

To account for errors in regional-scale topography, the HiRISE DEM was tied to Mars Orbiter Laser Altimeter (MOLA) point shot data (Smith et al., 2001) using the ASP pc_align function (Beyer et al., 2014). This algorithm allows for translation and rotation of the HiRISE DEM in three-dimensional space to find the mini-mum error between the input and reference elevation point clouds (Beyer et al., 2014). While any one HiRISE DEM has only minimal coverage by MOLA point shots, we chained this process whereby we first tied the High Resolution Stereo Camera (HRSC) DEM h5270_0000 (Neukum et al., 2004; Gwinner et al., 2010) to MOLA point shot data, and then tied the HiRISE DEM to the corrected HRSC DEM. Once the HiRISE DEM was corrected, the left HiRISE image (PSP_003798_1985) was orthorectified to the final DEM to-pography using the ASP.

For qualitative analysis of exposed sedimentary structures the HiRISE DEM and orthorectified image were examined using ESRI’sArcScene 3D visualization software. Quantitative analyses were performed on exposed strata, which were identified as differen-tially eroded units that protrude from the surrounding material or form clear topographic benches (e.g., Fig. 2). Surfaces of ex-posed layers are interpreted to be representative of bedding planes based on their contiguous nature and similarity to other examples of mapped bedding in martian sedimentary deposits (e.g., Lewis and Aharonson, 2006; Ansan et al., 2011; DiBiase et al., 2013).

Strata that are laterally contiguous over >50 m were mapped in ESRI’s ArcMap geographic information system (GIS) software, and location information (in three dimensions) was extracted at ∼5 m intervals. Assuming an approximately planar surface at the mapping scale, planes were fit to the extracted points using a linear least squares method (e.g., Lewis and Aharonson, 2006;Lewis et al., 2008; DiBiase et al., 2013). From these plane fits, strike and dip values were calculated for the exposed layers, as well as 95% confidence limits on these values based on the linear least squares plane misfit.

Following the methods of Lewis et al. (2008), a principal com-ponents analysis (PCA) was used to evaluate the quality of plane fits and to ensure that the points for a given mapped layer span an appropriate range in x, y, and z dimensions to produce a mean-ingful strike and dip; if the points along a mapped layer are too co-linear (i.e., fall approximately on a straight line in plan view), then the strike and dip of that layer will be poorly constrained. A PCA was completed for all the points mapped along each layer, and the fit was deemed robust and used for final analysis only if the PCA results satisfied two criteria: (1) the variance explained by the first principal component was <99.5%, and (2) the ratio of the variance explained by the second principal component to the third principal component was >15 (Lewis et al., 2008).

2.2. Mineralogy

The mineralogy of the exposed portions of the fan deposit was investigated using visible to near-infrared (VNIR) hyperspec-tral data from the Compact Reconnaissance Imaging Spectrom-eter for Mars (CRISM; Murchie et al., 2007). We analyzed a targeted CRISM image over the fan deposit (ID: HRL000040FF) alongside co-registered images and topographic data to deter-mine the context and stratigraphic position of identified mineral phases. The analyzed CRISM image has a spatial resolution of ∼36 m/pixel and covers the wavelength range ∼0.36–3.9 μm at a spectral sampling of ∼6.5 nm/channel. Calibrated I/F data were downloaded from the Planetary Data System (PDS) and corrected for the cosine of the incidence angle. Next, a stan-dard atmospheric correction method was applied, which re-lies on scaling atmospheric gas absorptions retrieved from an independent CRISM observation over a portion of the Olym-pus Mons volcano (the so-called ‘volcano scan correction’; e.g., Ehlmann et al., 2009; McGuire et al., 2009). The CRISM im-age was then map projected (Murchie et al., 2007, 2009), and manually georeferenced to the orthorectified HiRISE image in Ar-cMap.

CRISM data analysis was guided by spectral parameter maps that highlight spectral diversity within the image (Pelkey et al., 2007; Ehlmann et al., 2009; Salvatore et al., 2010). CRISM spectra from units of interest were divided by spectra from a spectrally bland region within the same detector column in the CRISM image to emphasize the unique spectral signature of the unit of interest within the ratioed spectrum (e.g., Ehlmann et al., 2008a, 2008b, 2009).

3. Results

We mapped 26 exposed layers within the Jezero crater west-ern fan deposit that satisfy our criteria for robust plane fits (Fig. 2; Table 1). These layers each represent a discrete bedding plane ex-posure along the erosional front of the fan deposit. While it is possible that some of these exposures represent laterally continu-ous layers within the deposit, confidently identifying disconnected strata across the deposit is extremely difficult. Therefore, we have taken the conservative approach of only fitting layers where they are visibly continuous at the HiRISE scale.

The mapped layers all dip approximately towards the basin depocenter (i.e., to the east; Figs. 2a, b), as expected for a pro-grading sedimentary fan deposit building out from the edge of a basin (Rich, 1951; Mitchum et al., 1977). Dips range from ∼0.5–9◦(Fig. 3a; Table 1) and there is a clear trend of increasing layer dip with increasing elevation along the eroded front of the fan deposit (Fig. 3a). We also identify several locations where exposed layers downlap onto underlying strata (Fig. 2).

CRISM data analysis reveals spectral signatures of both Fe/Mg-smectite and Mg-rich carbonate within the Jezero crater western fan deposit (Fig. 4), consistent with previous analyses (Ehlmann et al., 2008a, 2008b, 2009; Goudge et al., 2015). Fe/Mg-smectite is identified in many locations across the deposit, including strong spectral signatures in the deposit’s lower stratigraphic units (Figs. 4a, c). In contrast, Mg-rich carbonate is identified in only a few isolated locations across the deposit, and all of these identifi-cations correspond to stratigraphically higher light-toned materials (Figs. 4a, d) interpreted to be fluvial channel point bar deposits (Schon et al., 2012).

Fe/Mg-smectite is identified based on the presence of diag-nostic vibrational absorptions at ∼1.4, 1.9, and 2.3 μm (Fig. 4b; Clark et al., 1990; Bishop et al., 2002). The ∼1.4 μm absorp-tion is due to the first overtone of OH stretch, the ∼1.9 μm

360 T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365

Fig. 2. Examples of exposed layering within the Jezero crater western fan deposit. (a, b) Red and cyan lines show mapped layers, with corresponding strike and dip directions (yellow symbols). Red lines show layers with dips >2◦ that are interpreted as foresets. Cyan lines show layers with dips <2◦ that are interpreted as bottomsets. Note that the layers all dip approximately toward the basin depocenter, and that foresets always overlie bottomsets. Portions of HiRISE image PSP_003798_1985. White ‘<’ symbols show the approximate look directions for the perspective views in parts (c, d). North is up in both images. (c, d) Three-dimensional perspective views of the exposed layering in parts (a, b). Red arrows show stratigraphically higher layers, interpreted as foresets, downlapping onto approximately flat-lying stratigraphically lower layers, interpreted as bottomsets. Portions of HiRISE image PSP_003798_1985 draped over a HiRISE-derived DEM from stereo-pair images PSP_003798_1985 and PSP_002387_1985. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365 361

Table 1Attributes of layers mapped within the Jezero crater western fan deposit. Layers are sorted by mean elevation in descending order.

Strikea

(◦)Dipa

(◦)Mean elevationb

(m)Latitudec

(◦N)Longitudec

(◦E)Inferred origin of stratad

338.2 (±2.7) 8.5 (±0.5) −2470.8 (±2.4) 18.532 77.413 Foreset344.4 (±3.5) 5.7 (±0.3) −2472.0 (±5.0) 18.513 77.415 Foreset356.4 (±17.2) 7.6 (±1.8) −2474.0 (±2.4) 18.523 77.417 Foreset14.8 (±17.9) 4.4 (±1.8) −2475.0 (±1.9) 18.522 77.416 Foreset302.4 (±4.8) 9.4 (±1.2) −2475.4 (±1.8) 18.512 77.417 Foreset325.3 (±17.1) 8.6 (±1.8) −2476.3 (±8.6) 18.503 77.427 Foreset18.5 (±19.6) 4.0 (±1.4) −2480.8 (±0.9) 18.524 77.418 Foreset314.7 (±8.9) 4.7 (±0.7) −2480.8 (±2.7) 18.527 77.418 Foreset27.3 (±6.7) 7.9 (±1.0) −2481.5 (±1.8) 18.521 77.417 Foreset323.8 (±10.8) 3.4 (±0.5) −2483.7 (±2.3) 18.528 77.418 Foreset31.7 (±16.9) 2.1 (±0.4) −2484.6 (±2.0) 18.530 77.417 Foreset359.7 (±3.5) 9.3 (±0.3) −2486.2 (±9.2) 18.501 77.435 Foreset21.6 (±24.4) 5.6 (±2.3) −2488.8 (±1.3) 18.521 77.417 Foreset334.6 (±11.1) 2.9 (±0.3) −2493.6 (±2.7) 18.529 77.421 Foreset18.7 (±7.2) 3.1 (±0.2) −2495.5 (±2.9) 18.510 77.424 Foreset327.8 (±49.6) 2.9 (±2.1) −2495.7 (±1.3) 18.522 77.420 Foreset0.9 (±11.3) 3.8 (±0.3) −2497.7 (±3.1) 18.511 77.423 Foreset0.3 (±8.7) 2.5 (±0.4) −2499.4 (±1.4) 18.521 77.419 Foreset13.8 (±31.3) 0.5 (±0.6) −2503.9 (±0.4) 18.520 77.419 Bottomset22.3 (±36.8) 0.7 (±0.6) −2508.6 (±0.7) 18.516 77.419 Bottomset4.5 (±28.8) 1.8 (±0.3) −2510.0 (±1.2) 18.515 77.422 Bottomset354.2 (±6.2) 1.6 (±0.2) −2513.1 (±2.6) 18.503 77.437 Bottomset346.8 (±14.7) 1.5 (±0.2) −2515.3 (±2.3) 18.516 77.422 Bottomset32.8 (±5.9) 0.7 (±0.1) −2523.8 (±0.8) 18.491 77.428 Bottomset294.7 (±32.2) 0.9 (±0.2) −2526.5 (±1.0) 18.501 77.438 Bottomset351.4 (±45.8) 0.6 (±0.6) −2527.5 (±1.7) 18.488 77.429 Bottomset

a Values listed in parentheses are errors from 95% confidence limits from the linear least squares plane misfit.b Values listed in parentheses are 1 standard deviation on extracted elevation values along the layer.c Latitude and longitude values are from center points of the mapped layers, which correspond to the locations of the strike and

dip symbols in Figs. 2a, b.d Steeply dipping (>2◦) downlapping strata are interpreted as delta foresets. Shallowly dipping (<2◦) strata are interpreted as

delta bottomsets.

absorption is due to a combination tone of OH stretch and H–O–H bend, and the ∼2.3 μm absorption is due to a combina-tion tone of (Mg,Fe)–OH bend and OH stretch (Clark et al., 1990;Bishop et al., 2002). Mg-rich carbonate is identified based on the presence of diagnostic paired absorptions at ∼2.3 and 2.5 μm (Fig. 4b) that are due to an overtone of a fundamental vibrational mode of CO3 (Schroeder et al., 1962; Hunt and Salisbury, 1971). The Mg-rich carbonate-bearing material also has a broad absorp-tion centered at ∼1 μm and a narrower vibrational absorption cen-tered at ∼1.9 μm (Fig. 4b), consistent with previous identifications of Mg-rich carbonate in this region (Ehlmann et al., 2008b, 2009; Goudge et al., 2015). The ∼1 μm absorption is due to an elec-tronic crystal field transition from octahedrally coordinated Fe2+in olivine admixed with the Mg-rich carbonate (King and Ridley, 1987; Burns, 1993), and the ∼1.9 μm absorption is due to struc-tural H2O in either a hydrated carbonate (e.g., Calvin et al., 1994) or an additional hydrated mineral, such as smectite (e.g., Clark et al., 1990).

4. Discussion

4.1. Stratigraphic architecture of the Jezero crater western fan deposit

Our observations of the stratigraphic architecture of the Jezero crater western fan deposit are diagnostic of clinoform structures within a delta deposit rather than sub-parallel layering in an al-luvial fan system (Rich, 1951; Bull, 1977; Mitchum et al., 1977; Orton and Reading, 1993; Blair and McPherson, 1994). Downlap-ping stratal geometries, such as those observed here (Fig. 2), are consistent with layer terminations in the distal portions of deltaic clinoforms and are common in delta deposits on Earth (Rich, 1951;Mitchum et al., 1977; Pirmez et al., 1998).

Additional evidence of clinoforms within the deposit is the trend in layer dip versus elevation (Fig. 3). An idealized progra-

dational delta will accumulate sediment in clinoforms as shown schematically in Fig. 3b (e.g., Rich, 1951; Mitchum et al., 1977;Pirmez et al., 1998). If this deposit is eroded and the stratigraphic architecture exposed, as is the case at the Jezero crater western fan deposit erosional front (Figs. 1b, 2; Schon et al., 2012), a trend of increasing layer dip with increasing elevation (and stratigraphic position) along the erosional front is expected (e.g., Fig. 3b). This is the trend observed for the western fan deposit (Fig. 3a).

We interpret the more steeply dipping (>2◦) downlapping strata as delta foresets and the underlying, shallowly dipping (<2◦) strata as delta bottomsets. These dips are consistent with obser-vations of foreset and bottomset dips in both terrestrial delta de-posits (Rich, 1951; Orton and Reading, 1993) and other martian fan deposits interpreted as deltaic in origin (e.g., Lewis and Aharonson, 2006; Ansan et al., 2011; DiBiase et al., 2013). Therefore, based on the exposed stratal geometries of the Jezero crater western fan, we conclude that this deposit is deltaic in origin.

This conclusion is consistent with previous interpretations based on the morphology and geologic setting of this deposit (Fassett and Head, 2005; Ehlmann et al., 2008a; Schon et al., 2012). However, we again note that it is only the results of an analysis of the stratigraphic architecture of this deposit (Figs. 2, 3) that al-low us to uniquely constrain the depositional environment for this fan. We suggest that future studies of sedimentary fan deposits on Mars should not only focus on morphology and geologic context, but also on observations of exposed stratal geometries in order to definitively classify a deposit as a delta as opposed to an alluvial fan.

The stratal geometries of the Jezero crater western delta, and in particular the interpreted foreset dips (∼2–9◦), can provide further insight into the sedimentology of the deposit. Based on a compari-son to delta foreset slopes from a compilation of terrestrial data, the Jezero delta foreset dip angles fall within the range of val-ues expected for deposits dominated by sand- or silt-sized grains

362 T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365

Fig. 3. (a) Mean layer elevation versus dip for the 26 mapped layers in the Jezero crater western fan deposit. Note the trend of increasing dip with increasing ele-vation (i.e., stratigraphic position). Error bars for mean elevations are 1 standard deviation on extracted elevation values along the layer. Error bars for dips are the 95% confidence limits from the linear least squares plane misfit. (b) Idealized sketch of clinoforms within a prograding delta showing the expected trend of increasing layer dip with increasing elevation along an erosional front of the deposit.

(Orton and Reading, 1993). However, a direct comparison to em-pirical observations of terrestrial delta slopes is highly unlikely to be sufficient to accurately characterize any given delta deposit on Mars. Although this is the case, the foreset angles are large enough to exclude the possibility of the Jezero crater western delta being built into a body of water that was shallow compared with the source channel depths, which would act to suppress the formation of steeper foreset strata (e.g., Orton and Reading, 1993). This con-clusion is consistent with the absolute elevations of these strata, as they sit ∼100 m below the minimum water level of the Jezero crater paleolake (Fig. 1; Fassett and Head, 2005).

The measured foreset angles for the Jezero crater western delta are also well below the angle of repose. This indicates that the studied portion of the deposit must have been primarily built by deposition of material out of suspension, as opposed to by discrete, gravity-driven mass wasting events that are common on steeply sloping terrestrial delta fronts (e.g., Orton and Reading, 1993;Pirmez et al., 1998). This interpretation may simply be the re-sult of evaluating the portions of the delta that have remained preserved and exposed, as we have examined only the foreset to bottomset transition of this deposit, and clinoforms are com-monly asymmetric with steeper angles expected near the topset

to foreset transition (e.g., Pirmez et al., 1998); however, it is also possible that this observation records something fundamen-tal about how the Jezero crater western delta deposit was con-structed. Future studies that focus on examining the deposits of the topset distributary channels (e.g., Fassett and Head, 2005;Schon et al., 2012) and their relationship to the exposed foresets and bottomsets analyzed here may be able to further reconstruct the paleohydraulics and sediment transport conditions of the flows associated with the formation of this deposit.

Our conclusion that the Jezero crater western fan deposit is a delta has important implications for understanding the geo-logic history of this deposit and inferring the paleoenvironmen-tal information that may be preserved within these rocks. The western delta deposit must have formed when a lake existed within the Jezero crater basin. It has been shown that the flu-vial activity associated with the formation of this paleolake oc-curred during the major era of valley network formation (Fas-sett and Head, 2005, 2008), which is hypothesized to repre-sent an era of intense fluvial activity early in martian history that ceased at approximately the Noachian–Hesperian boundary, ∼3.7 Ga (Howard et al., 2005; Irwin et al., 2005; Fassett and Head, 2008). Therefore, the Jezero crater western delta deposit is likely to record important information about ancient environmental con-ditions during this early period of widespread fluvial activity on Mars.

4.2. Mineralogy of the Jezero crater western fan deposit

The spatial distribution of mineral identifications within the Jezero crater western delta deposit is also consistent with a deltaic origin. Of particular importance is the identification of strong sig-natures of Fe/Mg-smectite, likely a detrital clay mineral in this system (Ehlmann et al., 2008a, 2009; Goudge et al., 2015), in the stratigraphically lowest units of the delta deposit (Figs. 4a, c), which we interpret as bottomsets. This observation is consistent with terrestrial facies models of delta deposits in which fine-grained materials transported in suspension, such as clay minerals, are concentrated in bottomsets in the most distal margins of the delta (Rich, 1951; Orton and Reading, 1993).

Interestingly, the few exposures of Mg-rich carbonate within the western delta deposit are concentrated in point bar deposits (Figs. 4a, d) of the topset distributary channels of the delta (Schon et al., 2012). This observation is consistent with previous con-clusions that the Mg-rich carbonate in this deposit is detrital (Ehlmann et al., 2008a, 2008b, 2009; Goudge et al., 2015), which would also imply the carbonate is of a coarser grain size than the Fe/Mg-smectite. However, it is also possible that the carbon-ate signature in these coarser-grained facies is a cement precipi-tated from carbonate-saturated fluids, possibly as a result of the dissolution of carbonate in the watershed and interaction with atmospheric CO2. In this latter case, the carbonates may record information about ancient climate conditions during the period of lacustrine activity.

Regardless of the specific origin, the provenance of sediments within this deposit can be traced to regional Fe/Mg-smectite- and Mg-rich carbonate-bearing units within the ∼12,000 km2 water-shed of the Jezero western delta deposit (Goudge et al., 2015). This deposit has thus incorporated sediment from a large region of the adjacent martian crust that was primarily aqueously altered prior to the fluvial activity that fed the paleolake within the Jezero crater basin during delta formation (Goudge et al., 2015).

4.3. Implications for future exploration of Jezero crater

Our observations and conclusions add to the suggestion that the Jezero crater paleolake basin is a compelling site for fu-

T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365 363

Fig. 4. Mineralogy of the Jezero crater western fan deposit from CRISM data. North is up in all images. (a) Overview CRISM parameter map of the Jezero crater western fan deposit. Locations of parts (c) and (d) are shown by white boxes. CRISM false-color map created using the parameters OLINDEX2 (Salvatore et al., 2010), BD2500 (Ehlmann et al., 2009), and D2300 (Pelkey et al., 2007), overlain on a mosaic of HiRISE image PSP_003798_1985 and CTX image P04_002664_1988. Interpretive legend based on CRISM parameter map is shown in upper right corner. Outline of fan deposit is from Goudge et al. (2015), and the parameter map is highlighted in this region. (b) CRISM ratioed spectra of Mg-rich carbonate (green spectra C1 and C2) and Fe/Mg-smectite (blue spectra S1, S2, and S3) detections from the fan deposit. Numerator spectra are extracted from CRISM image HRL000040FF at the locations shown by the labeled circles in part (a). Dashed lines are located at ∼1.4, 1.92, 2.3, and 2.5 μm. (c) Example Fe/Mg-smectite detection within the stratigraphically lowest units of the fan deposit, interpreted to be delta bottomsets. Image shows the same CRISM false-color parameter map as in part (a) overlain on a portion of HiRISE image PSP_003798_1985. (d) Example Mg-rich carbonate detection within light-toned material interpreted to be fluvial channel point bar deposits (Schon et al., 2012). Image shows the same CRISM false-color parameter map as in part (a) overlain on a portion of HiRISE image PSP_003798_1985. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

ture in situ exploration (e.g., Ehlmann et al., 2008a; Schon et al., 2012), such as by the upcoming Mars 2020 rover (Mustard et al., 2013). This site provides an exceptionally exposed delta de-posit with a well-studied geologic history (Fassett and Head, 2005, 2008; Ehlmann et al., 2008a, 2008b, 2009; Schon et al., 2012;Goudge et al., 2015). The erosional front and isolated erosional remnants of the delta deposit (Figs. 1b, 2; Schon et al., 2012;Goudge et al., 2015) are accessible to a rover that lands on the Jezero crater floor and would provide a wealth of potential tar-gets for in situ exploration, as has been seen with the sedimen-tary deposits within Gale crater currently being explored by the Mars Science Laboratory Curiosity rover (e.g., Freissinet et al., 2015;Grotzinger et al., 2015).

In situ exploration of deltaic sediment within Jezero crater would be highly complementary to the ongoing analyses of fluvio-lacustrine and deltaic sedimentary deposits at Gale crater by Cu-

riosity (Grotzinger et al., 2015), as the Gale crater paleolake is hydrologically closed (Irwin et al., 2005), whereas the Jezero crater paleolake is hydrologically open (Fassett and Head, 2005). Hydro-logically open and closed paleolake basins should be expected to have water columns with a significantly different aqueous geo-chemistry: hydrologically open lakes have a continual throughput of water and dissolved ions, whereas hydrologically closed lakes only lose water to evaporation and infiltration, and so are able to continually build up large concentrations of dissolved ions. Com-paring results from in situ analyses of the mineralogy and geo-chemistry of deltaic sediment deposited in paleolake basins with distinct hydrologic settings would provide an exciting opportunity to further understand a more complete range of ancient martian aqueous surface environments than could be inferred from in situanalyses at only one type of paleolake basin.

364 T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365

Furthermore, exposures of Fe/Mg-smectite-rich deltaic bottom-sets (Figs. 2, 4a, c) that would be accessible to a rover provide an excellent opportunity for in situ analyses from an astrobiol-ogy perspective. Quiescent depositional environments dominated by fine-grained sedimentation, such as deltaic bottomsets, have a high concentration and preservation potential for organic matter (e.g., Rich, 1951; Blair and Aller, 2012), making them attractive locations for in situ exploration on Mars (Summons et al., 2011). As originally suggested by Ehlmann et al. (2008a), the presence of Fe/Mg-smectite within the delta deposit, and particularly within the bottomsets of the deposit (Figs. 4a, c), is of further impor-tance as clay minerals have the ability to bind organic matter (e.g., Kennedy et al., 2002). Indeed, it has recently been shown that clay-bearing lacustrine mudstones observed by Curiosity in Gale crater exhibit evidence for preserved organic molecules (Freissinet et al., 2015).

Aside from the delta deposit itself, Jezero crater also contains exposures of a widespread, regional Mg-rich carbonate-bearing unit that underlies the delta (Ehlmann et al., 2008b, 2009; Goudge et al., 2015) and a volcanic unit on the floor of the basin (Schon et al., 2012; Goudge et al., 2015). While the volcanic unit within Jezero crater is likely to bury any non-deltaic distal sediment, our observations indicate that there are readily accessible exposures of deltaic bottomsets within the basin (Figs. 2, 4a, c), which are equally as compelling for in situ exploration from an astrobiology perspective (Summons et al., 2011). Therefore, the presence of a dateable volcanic unit and a regional carbonate-bearing unit, in addition to the well-exposed deltaic sediment, constitute a diverse array of scientifically interesting geologic units within Jezero crater that we suggest make this site highly compelling for future explo-ration.

5. Conclusions

Based on both the stratigraphic architecture and mineralogy of the Jezero crater western fan deposit, we conclude that this deposit is deltaic in origin and formed when the basin hosted a stable body of standing water during the era of valley network formation early in martian history. We stress the importance of examining the stratigraphic architecture of sedimentary fan deposits on Mars, in addition to their morphology and geologic setting, for defini-tively constraining a depositional environment (e.g., delta versus alluvial fan).

The Jezero crater western delta deposit displays well-exposed strata that are likely to preserve paleoenvironmental information from early in Mars’ history and have a high concentration and preservation potential for organic matter. This includes strata we interpret as Fe/Mg-smectite-rich bottomsets that would be accessi-ble to exploration by a landed mission, such as the upcoming Mars 2020 rover. We suggest that in situ analysis of the remarkably well-preserved stratigraphy of the western delta deposit shown here has the potential to address both geologically and astrobiologically compelling questions, and that Jezero crater is a prime candidate for exploration by future landed missions to Mars.

Acknowledgements

We thank J.L. Dickson for help with image and topographic data processing, and B.L. Ehlmann and D. Mohrig for helpful dis-cussions. Two anonymous reviewers are thanked for helpful com-ments that improved the quality of this manuscript, and C. Sotin is thanked for editorial handling. We express our appreciation for the superb work of the NASA MRO project team and the CRISM Science Operations Center (SOC). TAG gratefully acknowledges sup-port for this work from the Natural Sciences and Engineering Re-search Council of Canada (NSERC) Postgraduate Scholarships Pro-

gram (PGSD3-421594-2012). JWH acknowledges financial support from the Mars Express High Resolution Stereo Camera (HRSC) Team (JPL 1488322). Parts of this research were conducted using the computational resources and services of the Center for Com-putation and Visualization (CCV) at Brown University.

References

Ansan, V., et al., 2011. Stratigraphy, mineralogy, and origin of layered deposits inside Terby crater, Mars. Icarus 211, 273–304. http://dx.doi.org/10.1016/j.icarus.2010.09.011.

Beyer, R.A., Alexandrov, O., Moratto, Z., 2014. Aligning terrain model and laser al-timeter point clouds with the Ames Stereo Pipeline. Lunar Planet. Sci. 45, Ab-stract 2902.

Bishop, J., Madejova, J., Komadel, P., Froschl, H., 2002. The influence of structural Fe, Al and Mg on the infrared OH bands in spectra of dioctahedral smectites. Clay Miner. 37, 607–616. http://dx.doi.org/10.1180/0009855023740063.

Blair, N.E., Aller, R.C., 2012. The fate of terrestrial organic carbon in the marine en-vironment. Annu. Rev. Mar. Sci. 4, 401–423. http://dx.doi.org/10.1146/annurev-marine-120709-142717.

Blair, T.C., McPherson, J.G., 1994. Alluvial fans and their natural distinction from rivers based on morphology, hydraulic processes, sedimentary processes, and facies assemblages. J. Sediment. Res. A 64, 450–489.

Broxton, M.J., Edwards, L.J., 2008. The Ames Stereo Pipeline: automated 3D surface reconstruction from orbital imagery. Lunar Planet. Sci. 39, Abstract 2419.

Bull, W.B., 1977. The alluvial-fan environment. Prog. Phys. Geogr. 1, 222–270.Burns, R.G., 1993. Mineralogical Applications of Crystal Field Theory, 2nd ed. Cam-

bridge Univ. Press, New York. 551 pp.Calvin, W.M., King, T.V.V., Clark, R.N., 1994. Hydrous carbonates on Mars?: Evidence

from Mariner 6/7 infrared spectrometer and ground-based telescopic spectra. J. Geophys. Res. 99, 14,659–14,675.

Clark, R.N., King, T.V.V., Klejwa, M., Swayze, G.A., Vergo, N., 1990. High spectral res-olution reflectance spectroscopy of minerals. J. Geophys. Res. 95, 12,653–12,680.

Di Achille, G., Hynek, B.M., 2010. Deltas and valley networks on Mars: implications for a global hydrosphere. In: Cabrol, N.A., Grin, E.A. (Eds.), Lakes on Mars. Else-vier, Oxford, UK, pp. 223–248.

DiBiase, R.A., Limaye, A.B., Scheingross, J.S., Fischer, W.W., Lamb, M.P., 2013. Deltaic deposits at Aeolis Dorsa: sedimentary evidence for a standing body of water on the northern plains of Mars. J. Geophys. Res. 118, 1285–1302. http://dx.doi.org/10.1002/jgre.20100.

Ehlmann, B.L., et al., 2008a. Clay minerals in delta deposits and organic preservation potential on Mars. Nat. Geosci. 1, 355–358. http://dx.doi.org/10.1038/ngeo207.

Ehlmann, B.L., et al., 2008b. Orbital identification of carbonate-bearing rocks on Mars. Science 322, 1828–1832. http://dx.doi.org/10.1126/science.1164759.

Ehlmann, B.L., et al., 2009. Identification of hydrated silicate minerals on Mars using MRO-CRISM: geologic context near Nili Fossae and implications for aqueous al-teration. J. Geophys. Res. 114, E00D08. http://dx.doi.org/10.1029/2009JE003339.

Fassett, C.I., Head, J.W., 2005. Fluvial sedimentary deposits on Mars: ancient deltas in a crater lake in the Nili Fossae region. Geophys. Res. Lett. 32, L14201. http://dx.doi.org/10.1029/2005GL023456.

Fassett, C.I., Head, J.W., 2008. The timing of martian valley network activity: con-straints from buffered crater counting. Icarus 195, 61–89. http://dx.doi.org/10.1016/j.icarus.2007.12.009.

Freissinet, C., et al., 2015. Organic molecules in the Sheepbed Mudstone, Gale Crater, Mars. J. Geophys. Res. 120, 495–514. http://dx.doi.org/10.1002/2014JE004737.

Goudge, T.A., Mustard, J.F., Head, J.W., Fassett, C.I., Wiseman, S.M., 2015. Assessing the mineralogy of the watershed and fan deposits of the Jezero crater pale-olake system, Mars. J. Geophys. Res. 120, 775–808. http://dx.doi.org/10.1002/2014JE004782.

Grotzinger, J.P., Milliken, R.E., 2012. The sedimentary rock record of Mars: distri-bution, origins, and global stratigraphy. In: Grotzinger, J.P., Milliken, R.E. (Eds.), Sedimentary Geology of Mars. SEPM Spec. Publ., vol. 102. Society for Sedimen-tary Geology, Tulsa, USA, pp. 1–48.

Grotzinger, J.P., et al., 2015. Deposition, exhumation, and paleoclimate of an ancient lake deposit, Gale crater, Mars. Science 350, aac7575. http://dx.doi.org/10.1126/science.aac7575.

Gwinner, K., Scholten, F., Preusker, F., Elgner, S., Roatsch, T., Spiegel, M., Schmidt, R., Oberst, J., Jaumann, R., Heipke, C., 2010. Topography of Mars from global map-ping by HRSC high-resolution digital terrain models and orthoimages: character-istics and performance. Earth Planet. Sci. Lett. 294, 506–519. http://dx.doi.org/10.1016/j.epsl.2009.11.007.

Howard, A.D., Moore, J.M., Irwin, R.P., 2005. An intense terminal epoch of widespread fluvial activity on early Mars: 1. Valley network incision and associated deposits. J. Geophys. Res. 110, E12S14. http://dx.doi.org/10.1029/2005JE002459.

Hunt, G.R., Salisbury, J.W., 1971. Visible and near-infrared spectra of minerals and rocks: II. Carbonates. Mod. Geol. 2, 23–30.

Irwin, R.P., Howard, A.D., Craddock, R.A., Moore, J.M., 2005. An intense terminal epoch of widespread fluvial activity on early Mars: 2. Increased runoff and

T.A. Goudge et al. / Earth and Planetary Science Letters 458 (2017) 357–365 365

paleolake development. J. Geophys. Res. 110, E12S15. http://dx.doi.org/10.1029/2005JE002460.

Kennedy, M.J., Pevear, D.R., Hill, R.J., 2002. Mineral surface control of organic carbon in black shale. Science 295, 657–660. http://dx.doi.org/10.1126/science.1066611.

King, T.V.V., Ridley, W.I., 1987. Relation of the spectroscopic reflectance of olivine to mineral chemistry and some remote sensing implications. J. Geophys. Res. 92, 11,457–11,469.

Lewis, K.W., Aharonson, O., 2006. Stratigraphic analysis of a distributary fan in Eber-swalde crater using stereo imagery. J. Geophys. Res. 111, E06001. http://dx.doi.org/10.1029/2005JE002558.

Lewis, K.W., Aharonson, O., Grotzinger, J.P., Squyres, S.W., Bell, J.F., Crumpler, L.S., Schmidt, M.E., 2008. Structure and stratigraphy of Home Plate from the Spirit Mars Exploration Rover. J. Geophys. Res. 113, E12S36. http://dx.doi.org/10.1029/2007JE003025.

Malin, M.C., Edgett, K.S., 2003. Evidence for persistent flow and aqueous sedimen-tation on early Mars. Science 302, 1931–1934. http://dx.doi.org/10.1126/science.1090544.

Malin, M.C., et al., 2007. Context Camera Investigation on board the Mars Re-connaissance Orbiter. J. Geophys. Res. 112, E05S04. http://dx.doi.org/10.1029/2006JE002808.

McEwen, A.S., et al., 2007. Mars Reconnaissance Orbiter’s High Resolution Imag-ing Science Experiment (HiRISE). J. Geophys. Res. 112, E05S02. http://dx.doi.org/10.1029/2005JE002605.

McGuire, P.C., et al., 2009. An improvement to the volcano-scan algorithm for at-mospheric correction of CRISM and OMEGA spectral data. Planet. Space Sci. 57, 809–815. http://dx.doi.org/10.1016/j.pss.2009.03.007.

Mitchum, R.M., Vail, P.R., Sangree, J.B., 1977. Seismic stratigraphy and global changes of sea level, part 6: stratigraphic interpretation of seismic reflection patterns in depositional sequences. In: Payton, C.E. (Ed.), Seismic Stratigraphy – Applications to Hydrocarbon Exploration. AAPG, Tulsa, USA, pp. 117–133.

Moratto, Z.M., Broxton, M.J., Beyer, R.A., Lundy, M., Husmann, K., 2010. Ames Stereo Pipeline, NASA’s open source automated stereogrammetry software. Lu-nar Planet. Sci. 41, Abstract 2364.

Murchie, S., et al., 2007. Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) on Mars Reconnaissance Orbiter (MRO). J. Geophys. Res. 112, E05S03. http://dx.doi.org/10.1029/2006JE002682.

Murchie, S.L., et al., 2009. Compact Reconnaissance Imaging Spectrometer for Mars investigation and data set from the Mars Reconnaissance Orbiter’s pri-

mary science phase. J. Geophys. Res. 114, E00D07. http://dx.doi.org/10.1029/2009JE003344.

Mustard, J.F., et al., 2013. Report of the Mars 2020 Science Definition Team. 154 pp., posted July 2013, by the Mars Exploration Program Analysis Group (MEPAG) at http://mepag.jpl.nasa.gov/reports/MEP/Mars_2020_SDT_Report_Final.pdf.

Neukum, G., et al., 2004. HRSC: the high resolution stereo camera of Mars express. ESA SP 1240, 17–35.

Ori, G.G., Marinangeli, L., Baliva, A., 2000. Terraces and Gilbert-type deltas in crater lakes in Ismenius Lacus and Memnonia (Mars). J. Geophys. Res. 105, 17,629–17,641.

Orton, G.J., Reading, H.G., 1993. Variability in deltaic processes in terms of sediment supply, with particular emphasis on grain size. Sedimentology 40, 475–512.

Pelkey, S.M., et al., 2007. CRISM multispectral summary products: parameteriz-ing mineral diversity on Mars from reflectance. J. Geophys. Res. 112, E08S14. http://dx.doi.org/10.1029/2006JE002831.

Pirmez, C., Pratson, L.F., Steckler, M.S., 1998. Clinoform development by advection-diffusion of suspended sediment: modeling and comparison to natural systems. J. Geophys. Res. 103, 24,141–24,157.

Rich, J.L., 1951. Three critical environments of deposition and criteria for recognition of rocks deposited in each of them. Geol. Soc. Am. Bull. 62, 1–19.

Salvatore, M.R., Mustard, J.F., Wyatt, M.B., Murchie, S.L., 2010. Definitive evidence of Hesperian basalt in Acidalia and Chryse planitiae. J. Geophys. Res. 115, E07005. http://dx.doi.org/10.1029/2009JE003519.

Schon, S.C., Head, J.W., Fassett, C.I., 2012. An overfilled lacustrine system and progra-dational delta in Jezero crater, Mars: implications for Noachian climate. Planet. Space Sci. 67, 28–45. http://dx.doi.org/10.1016/j.pss.2012.02.003.

Schroeder, R.A., Weir, C.E., Lippincott, E.R., 1962. Lattice frequencies and rotational barriers for inorganic carbonates and nitrates from low temperature infrared spectroscopy. J. Res. Natl. Bur. Stand. A, Phys. Chem. 66, 407–434.

Smith, D.E., et al., 2001. Mars Orbiter Laser Altimeter: experiment summary after the first year of global mapping of Mars. J. Geophys. Res. 106 (23), 23,689–23,722.

Summons, R.E., Amend, J.P., Bish, D., Buick, R., Cody, G.D., Des Marais, D.J., Dromart, G., Eigenbrode, J.L., Knoll, A.H., Sumner, D.Y., 2011. Preservation of martian or-ganic and environmental records: final report of the Mars biosignature working group. Astrobiology 11, 157–181. http://dx.doi.org/10.1089/ast.2010.0506.

Tanaka, K.L., Skinner, J.A., Dohm, J.M., Irwin, R.P., Kolb, E.J., Fortezzo, C.M., Platz, T., Michael, G.G., Hare, T.M., 2014. Geologic Map of Mars, Scale 1:20,000,000. U.S. Geological Survey Scientific Investigations Map SIM 3292. Available at http://pubs.usgs.gov/sim/3292.