Earth-like Habitable Environments in the Subsurface of Mars

Abstract

In Earth's deep continental subsurface, where groundwaters are often isolated for >10⁶ to 10⁹ years, energy released by radionuclides within rock produces oxidants and reductants that drive metabolisms of non-photosynthetic microorganisms. Similar processes could support past and present life in the martian subsurface. Sulfate-reducing microorganisms are common in Earth's deep subsurface, often using hydrogen derived directly from radiolysis of pore water and sulfate derived from oxidation of rock-matrix-hosted sulfides by radiolytically derived oxidants. Radiolysis thus produces redox energy to support a deep biosphere in groundwaters isolated from surface substrate input for millions to billions of years on Earth. Here, we demonstrate that radiolysis by itself could produce sufficient redox energy to sustain a habitable environment in the subsurface of present-day Mars, one in which Earth-like microorganisms could survive wherever groundwater exists. We show that the source localities for many martian meteorites are capable of producing sufficient redox nutrients to sustain up to millions of sulfate-reducing microbial cells per kilogram rock via radiolysis alone, comparable to cell densities observed in many regions of Earth's deep subsurface. Additionally, we calculate variability in supportable sulfate-reducing cell densities between the martian meteorite source regions. Our results demonstrate that martian subsurface groundwaters, where present, would largely be habitable for sulfate-reducing bacteria from a redox energy perspective via radiolysis alone. We present evidence for crustal regions that could support especially high cell densities, including zones with high sulfide concentrations, which could be targeted by future subsurface exploration missions.

1. Introduction

The surface of Mars is an extremely hostile environment characterized by freezing temperatures, desiccating conditions, high levels of ionizing radiation, oxidizing chemicals, low pressures, and a lack of liquid water that preclude any Earth-like organisms from surviving without adaptation that is unprecedented on Earth. Though some extremophiles on Earth have evolved to tolerate certain conditions relevant for Mars, any martian life formed in earlier eras would have taken refuge in the subsurface for survival (Michalski et al., 2018; National Academies of Sciences, Engineering, and Medicine, 2019). Within the last three decades, researchers have discovered that fluids preserved within Earth's kilometer-deep subsurface contain significant quantities of biomass utilizing a wide diversity of redox reactions to drive their microbial metabolisms (Onstott et al., 2019). Similar habitable subsurface environments could potentially have existed on ancient Mars (Tarnas et al., 2018; Onstott et al., 2019), may still exist on modern Mars (Michalski et al., 2013), and could exist on other planetary objects including ocean worlds, small bodies like Ceres, and exoplanets, making them an exciting frontier of planetary exploration (National Academies of Sciences, Engineering, and Medicine, 2019; Stamenković et al., 2019). Planetary objects with lower surface gravity than Earth, including Mars, likely have higher volumes of habitable pore and fracture space that extend deeper into the crust compared to Earth (e.g., Goossens et al., 2017; Lewis et al., 2019), increasing the volume of feasibly supportable biomass if sufficient liquid water and redox energy are available (Sleep, 2012).

In Earth's subsurface, actively metabolizing microbes can be sustained by fluids that have been isolated from surface substrate input for >10⁶ to 10⁹ years (Lin et al., 2006; Holland et al., 2013; Li et al., 2016; Warr et al., 2018; Lollar et al., 2019)—similar to groundwater isolation timescales expected on Mars (Grimm et al., 2017)—powered solely by redox energy derived from water-rock reactions including radiolysis and serpentinization (Sherwood Lollar et al., 2014; Warr et al., 2019). Radiolysis, the breaking of H₂O molecules in pore spaces by α, β, and γ radiation released from decay of radionuclides within host rock, generates both dissolved reductants (e.g., H₂; Lin et al., 2005b; Warr et al., 2019) and dissolved oxidants (e.g., H₂O₂; Lefticariu et al., 2010; Li et al., 2016). The dissolved H₂ generated by radiolysis can be used directly as a reductant by subsurface microbes, while the primary oxidant by-products of radiolysis can oxidize sulfides within the host rock matrix to form dissolved sulfate (SO₄) (Lefticariu et al., 2010; Li et al., 2016) that can then be used by microbes as an oxidant (Lin et al., 2005b, 2006; Chivian et al., 2008; Li et al., 2016; Lollar et al., 2019). Sulfate-reducing microorganisms are found kilometers deep on Earth in many groundwaters that have been isolated for >10⁷ years (Lin et al., 2005b, 2006; Chivian et al., 2008; Li et al., 2016; Lollar et al., 2019), and a significant component of the H₂ and SO₄ in especially ancient groundwaters is typically derived from radiolysis (Li et al., 2016; Warr et al., 2019). Radiation doses from decay of rock-matrix-hosted radionuclides are insufficient to sterilize microbial communities, as demonstrated by the presence of microbes in these subsurface settings (e.g., Lin et al., 2005a, 2005b, 2006; Chivian et al., 2008; Lollar et al., 2019). Thus, these Earth systems prove that in any subsurface environment containing radionuclides, liquid H₂O, and sulfides, radiolysis alone has the potential to provide sufficient biologically supportive redox energy to sustain a subsurface biome (Onstott et al., 2006; Sherwood Lollar et al., 2014). There are many alternative potential pathways to generate additional redox couples for life in subsurface martian environments, including in fluid mixing zones that create chemical disequilibria (Sherwood Lollar et al., 2007), which will not be considered here. Accordingly, the gas production rates and supportable cell density ranges presented in this work should be considered conservative estimates.

2. Methods

2.1. Summary

The model used to estimate H₂ production rates, sulfate production rates, dissolved H₂ concentrations, dissolved sulfate concentrations, and supportable sulfate-reducing cell densities is similar to models employed in the works of Lefticariu et al. (2010), Li et al. (2016), and Altair et al., (2018) (Section 2.2). Porosity (Section 2.3), sulfide concentrations (Section 2.4), average sulfide grain sizes (Section 2.5), radionuclide concentrations (Section 2.6), and sulfate-reducing metabolic rates are the key parameters necessary to constrain the model. H₂ production rates are dependent on porosity and radionuclide concentrations, increasing as porosity and radionuclide concentrations increase. Sulfate production rates are dependent on porosity, radionuclide concentrations, sulfide concentrations, and average sulfide grain size. They increase as porosity, radionuclide concentrations, and sulfide concentrations increase, and decrease as sulfide grain size increases, as less reactive surface area is exposed when sulfide grains are larger. Dissolved H₂ and sulfate concentrations increase as porosity decreases, as there is less fluid in which to dissolve the H₂ and sulfate that is produced. Supportable sulfate-reducing cell densities in units of cells (kg rock)⁻¹ increase as porosity and sulfate production rates increase, since sulfate is the limiting redox nutrient in all model results presented here. Supportable sulfate-reducing cell densities in units of cells (L fluid)⁻¹ decrease as porosity increases because there is more fluid per volume rock for the cells to reside in. Supportable sulfate-reducing cell densities are higher when the assumed sulfate-reducing metabolic rate is lower.

Porosity is constrained using density estimates for the martian crust from orbital gravimetry (Goossens et al., 2017) that are supported by rover gravimetry (Lewis et al., 2019) (Section 2.3). Sulfide concentrations (Section 2.4), average sulfide grain sizes (Section 2.5), and radionuclide concentrations (Section 2.6) are constrained by their measured values in martian meteorites. The maximum and minimum assumed sulfate-reducing metabolic rates are based on the fastest and slowest sulfate-reducing metabolic rates estimated for fracture fluids in the gold mines of Witwatersrand Basin (Lin et al., 2006; Chivian et al., 2008). Many of the calculated sulfate-reducing cell densities are compared to those found in Earth's deep subsurface (Onstott et al., 2003, 2019; Cockell et al., 2012; Magnabosco et al., 2018) (Fig. 2).

The amount of radiolytic H₂ and sulfate that would be produced in each of the different martian meteorite source regions is calculated based on their individual K, Th, and U concentrations (Fig. 1e, Table 2), sulfide concentrations (Fig. 1f, Table 1), average sulfide grain sizes (Fig. 1f, Table 1), and estimated range of porosity (Fig. 1c, Table 2) before the impact events that ejected these rocks from the martian gravity well. Based on the produced amounts of these redox nutrients, the number of supportable sulfate-reducing cells per kilogram rock is calculated for each of these meteorite source regions by using the range of sulfate-reducing metabolic rates observed in Earth's deep subsurface (Lin et al., 2006; Chivian et al., 2008). Porosity range assumptions are based on density estimates of different martian terrains from orbital gravimetry (Fig. 1c, 1d; Section 2.2) (Goossens et al., 2017), a range that is supported by rover gravimetry (Lewis et al., 2019). The radiolysis model applied here (Section 2.3) is similar to the one employed in the works of Lin et al. (2005b), Li et al. (2016), Dzaugis et al. (2018), Tarnas et al. (2018), and Warr et al. (2019).

FIG. 1.

Model input data. (a) K concentration (wt %) from Gamma Ray Spectrometer (GRS; Boynton et al., 2004) data, corrected for contributions from Cl, H, and S (Boynton et al., 2007; Tarnas et al., 2018). (b) Th concentration (ppm) from GRS data, corrected for contributions from Cl, H, and S (Boynton et al., 2007; Tarnas et al., 2018). (c) Possible crustal density estimates from orbital gravimetry (Goossens et al., 2017) that are supported by rover gravimetry (Lewis et al., 2019). (d) Division of Mars into three different regions, which are later tied to martian meteorite classes. (e) Radionuclide abundances in martian meteorites (Meyer, 2016) and GRS data from 60°S to 60°N, where near-surface ice does not contaminate measurements (Boynton et al., 2007; Tarnas et al., 2018). The lower spatial resolution of GRS data (hundreds of kilometers) results in the calculated K and Th concentrations being more closely clustered than those in the martian meteorites. There are also no U concentration measurements from GRS data; thus the color is black. (f) Sulfide concentrations and grain sizes in martian meteorites, which are compiled and referenced in Table 1. Black symbols show the maximum and minimum reported grain sizes, while red symbols show the reported average grain size. Color images are available online.

Table 1.

Sulfide Concentrations and Grain Sizes in Martian Meteorites

Sample	Sulfide abundance	Average grain size	Grain size min	Grain size max	Publications	Type
Sample	(wt %)	(μm)	(μm)	(μm)	Publications	Type
NWA 998	0.13	-	20	120	a	Nakhlite
Nakhla	0.07 (0.08)	-	10	50	a, b	Nakhlite
Governardor Valadares	0.06	-	2	25	a	Nakhlite
NWA 817	0.03	-	2	30 ± 15	a	Nakhlite
MIL 03346 and paired samples	0.05	-	-	-	c	Nakhlite
Y000593	0.08	-	-	-	b	Nakhlite
ALH A77005	0.3^d, 0.12^c, 0.19^b	100	-	-	b–d	Lherzolitic shergottite
LEW 88516	0.3	17.5	-	-	d	Lherzolitic shergottite
Y-293605	0.05	-	-	-	d	Lherzolitic shergottite
Y000097	0.12	-	-	-	b	Lherzolitic shergottite
LAR 12011	0.35	-	-	-	b	Lherzolitic shergottite
LAR 12095	0.5	-	-	-	b	Lherzolitic shergottite
Tissint	0.6^e, 0.35^c	-	<1	200	c, e	Olivine-phyric shergottite
EET A79001-A	0.52^b, 0.28^c	-	-	-	c	Olivine-phyric shergottite
DaG 476/489	0.45	50	10	100	d	Olivine-phyric shergottite
SaU 005/094	0.30^d, 0.16^c, 0.38^b	30	10	50	b–d	Olivine-phyric shergottite
Dhofar 019	0.5^d, 0.20^c	25	-	-	c, d	Olivine-phyric shergottite
NWA 1068/1110	0.8	-	-	-	c, d	Olivine-phyric shergottite
Y-980459	0.1^c, 0.4^b	30	5	50	b–d, f	Olivine-phyric shergottite
Zagami	0.7^d, 0.24^c, 0.35^b	<10	100	200	b–d	Basaltic shergottite
Shergotty	0.4	-	100	200	d	Basaltic shergottite
EET A79001-B	0.12	-	-	-	c	Basaltic shergottite
QUE 94201	0.3^d, 0.13^c	-	-	-	d	Basaltic shergottite
NWA 480	0.24	-	100	200	d	Basaltic shergottite
NWA 856	0.1	-	-	-	d	Basaltic shergottite
Los Angeles	0.67^d, 0.32^c	-	200	400	d	Basaltic shergottite
Mean of 37 shergottites	0.21 ± 0.18	-	-	-	c, g	Shergottite
NWA 7533	1 ± 0.1	35	-	-	h, i	Regolith breccia
Chassigny	0.02	20	1–5	30	j	Chassignite
NWA 2737	0.015 ± 0.005	-	3	60	k	Chassignite
NWA 8693	<0.01	-	1	20	l	Chassignite

Assume density of 5000 kg/m³ for sulfide and 3000 kg/m³ for host rock to convert from vol % to wt %.

The data of Franz et al. (2014, 2019) and Wang and Becker (2017) were calculated from whole-rock S contents assuming 38 wt % S for pyrrhotite and 54 wt % for pyrite. Other sulfide abundances are from image analyses on polished thick sections. ^aChevrier et al. (2011), ^bWang and Becker (2017), ^cFranz et al. (2014), ^dLorand et al. (2005 and references therein), ^eGattacceca et al. (2013), ^fBaumgartner et al. (2017), ^gFranz et al. (2019), ^hLorand et al. (2015), ⁱLorand et al. (2018a), ^jLorand et al. (2018b), ^kLorand et al. (2012), ^lHewins et al. (2020).

Table 2.

Range of Values for Martian Meteorite Calculations

Category	K₂O concentration (wt %)	Th concentration (ppm)	U concentration (ppm)	Sulfide concentration (vol %)	Sulfide grain size (μm)	Porosity (%)
Regolith breccia	0.42–0.51^a,b	1.1–2^a,b	0.29–0.36^a,b	0.6 ± 0.1^a,b,c	2–100^a,b,c	15–35
Chassignite	0.036–0.05^d	0.057–0.13^d	0.018–0.056^d	0.006–0.012^e,f	2–60^e,f	5–15
Nakhlite	0.11–0.27^d	0.15–0.42^d	0.052–0.23^d	0.018–0.078^g	2–50^g	5–15
Shergottite	0.0156–0.35^d	0.0117–0.57^d	0.00303–0.12^d	0.03–0.6^h	10–200^h	5–15

Agee et al. (2013), ^bWittmann et al. (2015), ^cLorand et al. (2018a),^dMeyer (2016), ^eLorand et al. (2018b), ^fLorand et al. (2012), ^gChevrier et al. (2011), ^hLorand et al. (2005). Sulfide grain size excludes the few unusually large sulfide grains reported in each meteorite class, as well as grains smaller than 2 μm, which are typically not analyzed.

2.2. Radiolysis model

The model used here follows the radiolytic H₂ and SO₄ production models presented in the works of Lin et al. (2005a), Lin et al. (2005b), Li et al. (2016), Tarnas et al. (2018), and Warr et al. (2019). This calculates radiogenic dose rate as $E_{n e t, i} = \frac{E_{i} \times W \times S_{i}}{1 + W \times S_{i}}$ (1)

where i is α, β, or γ radiation, E _net is the net dosage absorbed by pore water (eV kg⁻¹ s⁻¹), E is the apparent dosage from radioactive element decay (eV kg⁻¹ s⁻¹), W is the weight ratio of pore water to rock, and S is the stopping power of minerals within the rock matrix with respect to each type of radiation particle/ray, where S _α = 1.5, S _β = 1.25, S _γ = 1.14 (Lin et al., 2005a, 2005b; Tarnas et al., 2018; Warr et al., 2019). The stopping power of minerals can be estimated using Bragg's law (Bragg and Kleeman, 1905) and typically varies by <0.5 for silicate minerals (Nogami and Hurley, 1948).

W, the weight ratio of pore water to rock, primarily depends on the rock porosity, with additional minor variations induced by differences in water density in solid (∼900 kg m⁻³) or liquid (∼1000 kg m⁻³) phase. Values from Adamiec and Aitken (1998) are used for E for 1 wt % K, 1 ppm Th, and 1 ppm U for α, β, and γ radiation, which are E _K,α = 0 (Gy ka⁻¹), E _K,β = 0.782 (Gy ka⁻¹), E _K,γ = 0.243 (Gy ka⁻¹), E _Th,α = 0.061 (Gy ka⁻¹), E _Th,β = 0.027 (Gy ka⁻¹), E _Th,γ = 0.048 (Gy ka⁻¹), E _U,α = 0.218 (Gy ka⁻¹), E _U,β = 0.146 (Gy ka⁻¹), E _U,γ = 0.113 (Gy ka⁻¹). 1 Gy = 6.2415 × 10¹⁸ eV kg⁻¹.

H₂ yield is calculated as $Y_{H 2} = \sum E_{n e t, i} \times G_{H 2, i}$ (2)

where Y _H2 is the total yield of H₂ molecules (molecules kg⁻¹ s⁻¹) and G _H2 is the number of H₂ molecules produced per 100 eV of radiation. The conversion factor between (kg water) and (kg water and rock matrix) is W, the weight ratio of pore water to rock. For H₂O, G _H2,α = 0.96 H₂ molecules (100 eV)⁻¹, G _H2,β = 0.6 H₂ molecules (100 eV)⁻¹, G _H2,γ = 0.4 H₂ molecules (100 eV)⁻¹ (Lin et al., 2005a, 2005b; Tarnas et al., 2018; Warr et al., 2019). The value of G _H2 changes depending on the liquid or solid substance filling the pore space, and has received extensive study due to its importance in the context of nuclear waste storage. It is typically higher for chloride brines than for pure water (LaVerne and Tandon, 2005; Buck et al., 2012) due to the reaction of dissolved brine ions (e.g., Cl^-) with ions formed via radiative destruction of neutral H₂O molecules (e.g., OH·). This causes a depletion of ions that would react with H· generated by radiolysis, leaving more H· in solution, which reacts to form H₂ (2H· → H₂) (Klein et al., 2020). Hydrated chloride salts, embedded in either pore space or within the rock matrix, will also produce H₂, though at lower production efficiencies relative to pure water, ice, or brine (LaVerne and Tandon, 2005). These different production efficiencies are summarized in the work of Tarnas et al. (2018). The presence of certain minerals, such as zeolites, can significantly increase H₂ and oxidant production efficiency from radiolysis (Kumagai et al., 2013; Sauvage et al., 2021). Zeolites are found in both bedrock (Ehlmann et al., 2009) and dust (Ruff, 2004) on Mars; thus this mineral is expected to enhance radiolytic H₂ and oxidant production in some parts of the crust, but this effect is not modeled here in order to focus on the most conservative estimates possible.

SO₄ production is calculated as $Y_{S O 4} = \sum E_{n e t, i} \times G_{S O 4, i}$ (3)

where Y _SO4 is the total yield of moles SO₄ per m² sulfide surface area (moles m⁻² sulfide s⁻¹) and G _SO4 is the moles SO₄ produced per m² sulfide per Gy radiation. A G _SO4 value from the work of Lefticariu et al. (2010) is used (2.1 × 10⁻⁹ mol m⁻² Gy⁻¹), which is also used in Li et al. (2016) and Altair et al. (2018). However, while the experiments of Lefticariu et al., (2010) were performed under high-energy radiation, they showed that G _SO4 increases under low-radiation-dose conditions, which correspond to typical radiation doses in natural systems (Li et al., 2016). Furthermore, these experiments only investigated the oxidation, structural damage, and chemical weathering of sulfides exposed to γ radiation. As Lefticariu et al. (2010) noted, accounting for increased oxidation, structural damage, and chemical weathering of sulfides from α and β radiation—which typically represent >90% of total irradiated energy—would increase the G _SO4 value, perhaps significantly. This means once again that the calculated sulfate production rates, and thus supportable sulfate-reducing cell densities herein, are likely conservative estimates. The concentration of sulfate in fluid from radiolytic production is expressed as $C_{S O 4} = Y_{S O 4} \times C_{s u l f i d e} \times S_{s u l f i d e} ∕ W$ (4)

where C _SO4 is the production rate of dissolved sulfate (M year⁻¹), C _sulfide is the concentration of sulfide in the rock matrix (volume ratio), and S _sulfide is the average specific surface area of sulfides in the rock matrix (m² sulfide/kg). C _sulfide values used in this study are listed in Table 2, which encompasses the values found in martian meteorites (Greenwood et al., 2000a, 2000b; Rochette et al., 2001, 2005; Lorand et al., 2005, 2015, 2018a, 2018b; Chevrier et al., 2011; Baumgartner et al., 2017). S _sulfide values ranging from 2.26 × 10¹ m² kg⁻¹ to 1.41 × 10³ m² kg⁻¹ are used, which correspond to average sulfide grain sizes of 2–125 μm (Altair et al., 2018). The S _sulfide values for each meteorite class are assigned based on the range of sulfide grain sizes measured in those martian meteorites (Fig. 1f; Tables 1 and 2). Finally, following Altair et al. (2018), minimum and maximum sulfate-reducing metabolic rates of 5.5 × 10⁻¹⁸ to 3.6 × 10⁻¹⁷ mol SO₄ cell⁻¹ year⁻¹ are assumed, which correspond to the minimum and maximum sulfate-reducing metabolic rates estimated in deep subsurface fracture fluids of the gold mines from Witwatersrand Basin (Lin et al., 2006; Chivian et al., 2008).

2.3. Porosity model

The average porosity ranges for the martian meteorite source regions are estimated using density estimates for the different regions of Mars' crust from orbital gravimetry (Goossens et al., 2017) that are validated by rover gravimetry in Gale crater, which estimates a porosity range of 40 ± 6% for the ∼300 m vertical section of Aeolis Mons that was traversed by the Curiosity rover by mission sol 1896 (Lewis et al., 2019). Porosity in this context includes both fracture and pore space within rocks. The orbital gravimetry estimates give average densities of ∼1600 kg m⁻³ for the southern highlands, 2400 kg m⁻³ for the northern lowlands, and ∼2800 kg m⁻³ for the volcanic bulges. It is assumed that the volcanic bulges and northern lowlands are the possible source regions for the SNCs and the southern highlands are the source region for the regolith breccias (Cassata et al., 2018). In the estimation, an average solid crust matrix density range of 2000–3200 kg m⁻³ is used for the southern highlands' crust, consistent with its mineralogy (Ehlmann and Edwards, 2014), and 2600–3200 kg m⁻³ is used for the northern lowlands and volcanic bulges, consistent with their mineralogy and volcano-specific orbital gravimetry (Broquet and Wieczorek, 2019). It is assumed that the difference between the gravimetry density measurements and rock matrix density estimates is caused by pore space filled with ice or water. The densities add linearly such that $ρ_{b u l k} = ρ_{r o c k} (1 - θ) + ρ_{H 2 O} (θ)$ , where θ is porosity. This means the composition of the southern highlands' rock is assumed to vary, on average, between pure phyllosilicate (∼2000 kg m⁻³) and dense basalt (∼3200 kg m⁻³), while the composition of the northern lowlands' and volcanic bulges' rocks are assumed to vary between a 50-50 phyllosilicate-basalt mixture (∼2600 kg m⁻³) and dense basalt (∼3200 kg m⁻³). Given uncertainties in density estimates from orbital gravimetry, to be conservative, estimated porosities were reduced by 10%. This does not model depth dependence of porosity, but rather the average porosity of the possible source regions for the individual martian meteorites. This gives approximate average porosity ranges of 15–35% for the regolith breccias and 5–15% for the SNCs. The SNC porosity assumptions are supported by images of porous float vesicular basalt rocks in Gusev crater imaged by the Spirit rover (McMahon et al., 2013; McSween, 2015).

2.4. Sulfide concentrations

Martian meteorites contain sulfides of both magmatic (pyrrhotite) (Lorand et al., 2005, 2018b) and hydrothermal (pyrite) (Lorand et al., 2015, 2018a; Wittmann et al., 2015) origins. Sulfide concentrations range from <0.01–1.0 wt %. The assumed sulfide concentration ranges in the model are derived based on these sulfide concentrations found in martian meteorites (Greenwood et al., 2000b, 2000a; Rochette et al., 2001, 2005; Lorand et al., 2005, 2015, 2018a, 2018b; Chevrier et al., 2011; Franz et al., 2014, 2019; Baumgartner et al., 2017; Hewins et al., 2017, 2020; Wang and Becker, 2017) (Fig. 1f; Table 1). Martian meteorites contain variable abundances of sulfides with relatively consistent grain sizes for grains >10 μm. Published abundances for sulfides in martian meteorites, and reported average grain sizes and grain size range, are presented in Table 1. Magmatic sulfides (pyrrhotite) and hydrothermal sulfides (pyrite) are typically associated with SNCs and regolith breccias, respectively (Greenwood et al., 2000a, 2000b; Rochette et al., 2001, 2005; Lorand et al., 2005, 2015, 2018a, 2018b; Chevrier et al., 2011; Baumgartner et al., 2017). Pyrite is much less abundant than pyrrhotite, as it has been identified only in an orthopyroxenite (ALH 84001), two chassignites and a few nakhlites (as a minor pyrrhotite replacement product), as well as in regolith breccias (as primary hydrothermal precipitate). Sulfide abundances are typically higher in regolith breccias than in SNCs. The listed sulfide abundances are also minimum abundances, as desert weathering of martian meteorite samples on Earth causes oxidation of sulfides and partial S losses (Greenwood et al., 2000a, 2000b; Rochette et al., 2001, 2005; Lorand et al., 2005, 2015, 2018a, 2018b; Chevrier et al., 2011; Baumgartner et al., 2017).

The original report of sulfides in Gale crater via Chemisty and Mineralogy (CheMin; Blake et al., 2012) XRD data is no longer supported after recalibration of the data (Morrison et al., 2018), but evolved gas analysis–mass spectroscopy (Wong et al., 2020) and sulfur isotope measurements acquired with the Sample Analysis at Mars (SAM; Mahaffy et al., 2012) instrument (Franz et al., 2017) support the presence of sulfides in multiple Gale crater samples. Sulfides in Gale crater are diagenetic/epigenetic, likely resulting from thermal reduction of preexisting sulfates (Franz et al., 2017).

2.5. Sulfide grain sizes

Average grain sizes for sulfides in martian meteorites range from ∼5–100 μm with occasional outliers up to 400 μm (Fig. 1f; Table 1 and references therein). Sulfide grain sizes are relatively consistent across martian meteorite samples (Fig. 1f). A compilation of published measured sulfide average grain sizes and grain size ranges in martian meteorites is presented in Table 1 and plotted in Fig. 1f. It is noteworthy that many of these studies do not consider sulfides <10 μm in diameter; thus these results may skew average sulfide grain sizes toward larger-than-true values. Since sulfate production in our calculations increases as sulfide grain size decreases, as a result of greater specific surface area, this ensures the calculations are conservative. Sulfide grain sizes depend on a wide array of parameters affecting sulfide formation, including temperature, supersaturation of fluids, nucleation rates, and fracturing. For each individual martian meteorite class, the range of measured sulfide grain sizes in those meteorites is used as the basis for the calculations presented here.

2.6. Radionuclide concentrations

Martian meteorites and remote measurements by the Gamma Ray Spectrometer (GRS) onboard Mars Odyssey (Boynton et al., 2007), which is sensitive to the top ∼tens of centimeters of the martian crust, provide first-order constraints for radionuclide concentrations in the martian subsurface (Fig. 1). Radionuclide concentrations in the top ∼tens of centimeters in the martian crust vary with latitude and longitude (Fig. 1a, 1b), and volcanic SNC martian meteorites generally have lower radionuclide abundances than regolith breccias (Fig. 1e) that are likely representative of the Noachian crust (Cassata et al., 2018) (Fig. 1). Elemental concentrations in martian meteorites range from ∼0.01–0.4 wt % K, ∼0.01–2 ppm Th, and ∼0.003–0.36 ppm U (Onstott et al., 2006; Meyer, 2016) (Fig. 1e). Some U enrichment in these samples is possible through desert contamination (Tarnas et al., 2018). K and Th concentrations estimated from GRS measurements range from ∼0.07–0.6 wt % and ∼0.015–1.1 ppm (Boynton et al., 2007; Tarnas et al., 2018), respectively (Fig. 1a, 1b, 1e).

For calculations of SO₄ and H₂ production in martian meteorites, the measured values for K, Th, and U in each individual meteorite class were used (Onstott et al., 2006; Meyer, 2016), integrating both the maximum and minimum radionuclide concentration values measured (Fig. 1e; Table 2). The K and Th concentration ranges for SNCs are substantially lower than K and Th concentrations measured in GRS data (Fig. 1e), meaning the majority of the martian crust at depths shallower than ∼0.5 m typically has higher radionuclide concentrations than are represented by the SNCs. Following Hahn et al. (2011) and Tarnas et al. (2018), the GRS data are normalized to correct for contributions from Cl, H, and S. K concentrations measured by Mars rovers range from 0.05 wt % measured by Spirit to 4.4 wt % measured by Curiosity (Gellert et al., 2004; Le Deit et al., 2016).

3. Results

3.1. Habitable redox conditions in martian meteorite source regions

The results presented here show that the shergottite, nakhlite, chassignite, and regolith breccia martian meteorite source localities would all produce sufficient redox energy to support sulfate-reducing life where liquid groundwater exists within these lithologies. In all cases, based on the calculations herein, sulfate is the limiting redox nutrient in comparison with H₂. The regolith breccia source locality could support significantly more sulfate-reducing life than the SNC localities because it has higher concentrations of both radionuclides and sulfides, as well as higher estimated porosities. Of the SNCs, the shergottite and nakhlite source localities could support the highest densities of sulfate-reducing bacteria. Key findings are illustrated in Fig. 2, displayed in Table 3, and are summarized below:

FIG. 2.

Radiolytic redox energy production and number of supportable sulfate-reducing cells in martian meteorite source regions. The datapoints represent the highest and lowest endmembers for each given category, with the assumed values for the calculations presented in Table 2. The different colors of the same datapoints represent the supportable sulfate-reducing cells per kilogram rock assuming the highest and lowest sulfate-reducing metabolic rates observed in Earth's deep subsurface (5.5 × 10⁻¹⁸ to 3.6 × 10⁻¹⁷ mol cell⁻¹ year⁻¹) (Lin et al., 2006; Chivian et al., 2008). Cell density measurements from Earth's deep subsurface come from ^aCockell et al. (2012) and ^bOnstott et al. (2003). Color images are available online.

Table 3.

Estimated H₂ and Sulfate Production Rates for Martian Meteorite Source Regions and Associated Supportable Sulfate-Reducing Cell Densities

	Radiolytic H₂ production rate in mol (kg rock)⁻¹ year⁻¹	Radiolytic sulfate production rate in mol (kg rock)⁻¹ year⁻¹	Supportable sulfate-reducing cell density in cells (kg rock)⁻¹
Regolith breccia	∼[1–3] × 10⁻¹¹	∼[2 × 10⁻¹⁴]–[7 × 10⁻¹²]	[7 × 10²]–[1 × 10⁶]
Shergottite	∼[4 × 10⁻¹⁴]–[4 × 10⁻¹²]	∼[8 × 10⁻¹⁸]–[2 × 10⁻¹³]	[2 × 10⁻¹]–[3 × 10⁴]
Nakhlite	∼[5 × 10⁻¹³]–[6 × 10⁻¹²]	∼[1 × 10⁻¹⁶]–[2 × 10⁻¹³]	[3]–[3 × 10⁴]
Chassignite	∼[2 × 10⁻¹³]–[2 × 10⁻¹²]	∼[1 × 10⁻¹⁷]–[6 × 10⁻¹⁵]	[4 × 10⁻¹]–[1 × 10³]

Regolith breccia source locality: The sulfate production rates for the regolith breccia source locality vary from ∼[2 × 10⁻¹⁴]–[7 × 10⁻¹²] mol (kg rock)⁻¹ year⁻¹, while H₂ production rates range from ∼[1–3] × 10⁻¹¹ mol (kg rock)⁻¹ year⁻¹, which could support [7 × 10²]–[1 × 10⁶] sulfate-reducing cells per kilogram rock wherever liquid water exists. The regolith breccia source locality would generate the most redox energy for sulfate-reducing microorganisms compared to the other martian meteorite source localities.

Shergottite source locality: The sulfate production rates for shergottite source localities vary from ∼[8 × 10⁻¹⁸]–[2 × 10⁻¹³] mol (kg rock)⁻¹ year⁻¹, while H₂ production rates range from ∼[4 × 10⁻¹⁴]–[4 × 10⁻¹²] mol (kg rock)⁻¹ year⁻¹, which could support [2 × 10⁻¹]–[3 × 10⁴] sulfate-reducing cells per kilogram rock wherever liquid water exists.

Nakhlite source locality: The sulfate production rates for nakhlite source localities vary from ∼[1 × 10⁻¹⁶]–[2 × 10⁻¹³] mol (kg rock)⁻¹ year⁻¹, while H₂ production rates range from ∼[5 × 10⁻¹³]–[6 × 10⁻¹²] mol (kg rock)⁻¹ year⁻¹, which could support [3]–[3 × 10⁴] sulfate-reducing cells per kilogram rock wherever liquid water exists.

Chassignite source locality: The sulfate production rates for chassignite source localities vary from ∼[1 × 10⁻¹⁷]–[6 × 10⁻¹⁵] mol (kg rock)⁻¹ year⁻¹, while H₂ production rates range from ∼[2 × 10⁻¹³]–[2 × 10⁻¹²] mol (kg rock)⁻¹ year⁻¹, which could support [4 × 10⁻¹]–[1 × 10³] sulfate-reducing cells per kilogram rock wherever liquid water exists.

The sulfate production rates from these calculations are several orders of magnitude lower than those of Lefticariu et al. (2010) because those authors assumed 5–10 wt % pyrite in the martian crust, while this study uses lower sulfide concentrations based on measured values from martian meteorites (Fig. 1f, Table 1). Because the model uses the lowest and highest measured sulfate-reducing metabolic rates in Witwatersrand Basin for calculating cell density estimates (Lin et al., 2006; Chivian et al., 2008), there is a difference between the estimated maximum and minimum range for sulfate production rates and the estimated maximum and minimum range for number of supportable sulfate-reducing cells per kilogram rock. H₂ production estimates, in units of H₂ dissolved in water, range from 0.002–0.165 nM H₂/year. This compares to estimates by Dzaugis et al. (2018) of 0.001–1.2 nM H₂/year for proposed Mars landing sites. The differences in the dissolved H₂ estimates between this study and that of Dzaugis et al. (2018) derive from differences in radionuclide concentrations and porosity used for calculations in the two studies.

The number of supportable sulfate-reducing cells per kilogram rock in the nakhlite, shergottite, and regolith breccia source localities are comparable to those found in Earth's deep subsurface (∼10⁴ to 10⁶ cells per kilogram rock; Onstott et al., 2019) (Fig. 2). Estimates for the chassignite source locality meanwhile are slightly lower (∼10¹ to 10³ cells per kilogram rock). In all cases, sulfate is the redox nutrient that limits sustainable cell densities, and as such estimated cell densities are most sensitive to estimated sulfate production rates. The martian regolith breccia NWA 7034 and one of its paired samples, NWA 7533, contain the highest radionuclide and sulfide concentrations of the currently known martian meteorites, as well as the highest estimated porosities; thus their source locality would host sulfate production rates capable of supporting the greatest density of sulfate-reducing microorganisms relative to other known martian meteorites. Regolith breccia meteorites are likely representative of the southern highlands, which is a major fraction of Mars' total crustal volume.

The results presented here support growing evidence that the subsurface is potentially the largest, longest-lived, and most consistently habitable environment on Mars (Michalski et al., 2013, 2018; Tarnas et al., 2018; Onstott et al., 2019; Ojha et al., 2020) and could serve as a refugia for any life that ever existed on the martian surface (Michalski et al., 2018; National Academies of Sciences, Engineering, and Medicine, 2019). Other viable processes could produce redox nutrients to sustain additional metabolisms in the martian subsurface, including H₂ and CH₄ production via serpentinization and Fischer-Tropsch-type synthesis, respectively (e.g., Warr et al., 2019), atmospheric O₂ dissolution (Stamenković et al., 2018), atmospheric H₂ and CO dissolution (Weiss et al., 2000), as well as Fe²⁺ and Fe³⁺ from minerals (Onstott et al., 2019). If sulfides are not present in the rock matrix, these other reactions may be favorable over sulfate reduction. These results provide scientific justification for future missions to search for signs of extant life in the martian subsurface—which is an intriguing next frontier for planetary exploration (National Academies of Sciences, Engineering, and Medicine, 2019; Stamenković et al., 2019)—and identify key characteristics of optimal regions for future landing sites.

Even the relatively small sampling of Mars' petrologic diversity via four martian meteorite classes (shergottites, nakhlites, chassignites, regolith breccias) shows a wide range in redox energy production rates from these different rock types/localities (Fig. 2). Some of these localities, such as the source regions for regolith breccias, shergottites, and nakhlites, could support comparable concentrations of extant life to those found in Earth's deep subsurface (Onstott et al., 2003, 2019; Cockell et al., 2012). As is demonstrated by the diversity of radionuclide concentrations, sulfide concentrations, and sulfide grain sizes in martian meteorites, some regions of the martian crust would support higher sulfate-reducing cell densities than others. Because sulfate is likely the limiting nutrient for hydrogenotrophic sulfate-reducing metabolisms in a martian subsurface biosphere rather than H₂, the highest concentrations of sulfate-reducing microorganisms would exist in a water-bearing crustal section with high sulfide abundance, low sulfide grain sizes, and relatively high radionuclide concentrations. Regions rich in zeolites, which enhance radiolytic gas production (Kumagai et al., 2013; Sauvage et al., 2021), would also sustain higher sulfate-reducing bacteria concentrations. Future missions should focus on characterizing where these criteria are met, as these would be prime landing site targets for extant life investigation from a redox energy perspective.

4. Discussion

4.1. Potential for higher cell densities elsewhere on Mars

The possibility of relatively high sulfide concentrations on Mars has been recognized for decades due to the presence of ultramafic lithologies there (Burns and Fisher, 1990; Baumgartner et al., 2015; Humayun et al., 2019) (Section 4.2). Concentration of magmatic sulfides is assumed to derive from assimilation of crustal sulfates and/or S-rich regolith in lava flows, as sampled by nakhlites, which likely crystallized from a sulfide-saturated melt after this assimilation (Franz et al., 2014; Mari et al., 2019). Additional evidence for the existence of concentrated martian sulfides has continued to accrue. This includes sulfides in martian meteorites (Table 1 and references therein), in Gale crater (Franz et al., 2017; Wong et al., 2020), the prevalence of sulfate-hematite assemblages on the martian surface (e.g., Wiseman et al., 2008), which may or may not have formed as weathering products of sulfides (e.g., Zolotov and Shock, 2005), verifiable orbital detections of serpentine outcrops in Noachian terrains (Leask et al., 2018), which are often associated with formation of reduced sulfides and metal alloys on Earth (e.g., Economou and Naldrett, 1984; Thalhammer et al., 1986; Shiga, 1987; Auclair et al., 1993; Wafik et al., 2001; Marques et al., 2007), Cu concentrations in Gale crater that are likely caused by the presence of sulfides (Payré et al., 2019), and the low oxygen fugacity (Mari et al., 2019 and references therein) and high sulfur fugacity of the ancient martian mantle relative to Earth's (Wang and Becker, 2017; Mari et al., 2019 and references therein). Pyrrhotite hosts magnetization in multiple classes of martian meteorites (Rochette et al., 2005 and references therein); thus at high-enough concentrations its subsurface presence could explain Mars' crustal magnetic field anomalies (Langlais et al., 2019; Johnson et al., 2020).

High concentrations of sulfides could form via differentiation of sulfide-saturated lavas that were contaminated by crustal assimilation of sulfates (Burns and Fisher, 1990; Baumgartner et al., 2015) as occurred in nakhlite lava flows (Franz et al., 2014; Mari et al., 2019), differentiation of impact melt accompanied by crustal sulfur assimilation as occurred in Sudbury, Ontario (e.g., Therriault et al., 2002), later hydrothermal sulfide precipitation with or without serpentinization, as is recorded in regolith breccia martian meteorites (Lorand et al., 2015), or differentiated intrusions (Fig. 3; Section 4.3). Once formed, fluid fracture networks within such sulfide-rich regions in the subsurface of Mars could support high concentrations of radiolytically fueled sulfate-reducing microorganisms and would be key locations to investigate the possibility of extant life on Mars.

FIG. 3.

Possible sulfide concentration settings on Mars in addition to those found in martian meteorites. Topographic cross-section of Mars acquired by using data from the Mars Orbital Laser Altimeter (MOLA) from 10°S, 141°E; 29°N, 149°E; 4°S, 104°E; 35°N, 96°E. See Section 4.3 for discussion of possible sulfide concentration settings on Mars, and Table 1 for sulfide concentrations found in martian meteorites that are used for calculations in this study. (1) Differentiated impact melt sheet, similar to sulfide concentrations at the Sudbury Impact Structure in Canada (Therriault et al., 2002). (2) Sulfide concentration in layered intrusion, similar to Bushveld Igneous Complex in South Africa and many other localities on Earth. (3) Sulfide hydrothermally concentrated in serpentinite, similar to many ophiolite localities on Earth. Mg-rich serpentine has been detected on Mars (Leask et al., 2018). (4) Sulfide concentrations in sediments, such as those in Gale crater (Franz et al., 2017; Wong et al., 2020). (5) Sulfide hydrothermally concentrated without associated serpentinite minerals, such as those in regolith breccia martian meteorites (Lorand et al., 2005, 2015, 2018a, 2018b). (6) Concentrations of sulfides in differentiated ultramafic-to-mafic lava flows, as occurs in many localities on Earth including the Kambalda and Perseverance deposits in Australia. Sulfides are also concentrated in picrites, including the Pechenga deposits in Russia. This is a possible formation setting for nakhlites, which crystallized from sulfide-saturated magmas after assimilation of S from the crust/regolith (Franz et al., 2014; Mari et al., 2019). (7) Concentration of sulfides in ultramafic-to-mafic subvolcanic intrusions beneath volcanic provinces such as Tharsis, Elysium, and Syrtis Major. Color images are available online.

4.2 Evidence for concentrated sulfides on Mars

The conclusions from this study do not require concentrated sulfides on Mars, but rather are based on the large body of data for disseminated sulfides on Mars. As such, the radiolytic H₂ and sulfate production estimates and associated cell density estimates are appropriately conservative and are consistent with the body of lithological and geochemical data available. The presence of concentrated sulfide deposits on Mars can be considered, however, and supporting evidence includes (1) sulfur assimilation and sulfide saturation in nakhlite melts (Franz et al., 2014; Mari et al., 2019), (2) geologic evidence for high sulfide concentrations elsewhere in the martian crust based on expected mantle oxygen (Stanley et al., 2011; Armstrong et al., 2015) and sulfur (Ding et al., 2014) fugacities, (3) surface evidence for sulfide oxidation (Zolotov and Shock, 2005; Dehouck et al., 2012; Vaniman et al., 2014), and (4) orbital detections of past serpentinization (Ehlmann et al., 2010; Leask et al., 2018), which produces reducing fluids and is often associated with generation of hydrothermal sulfide deposits on Earth (e.g., Economou and Naldrett, 1984; Thalhammer et al., 1986; Shiga, 1987; Auclair et al., 1993; Wafik et al., 2001; Marques et al., 2007).

Nakhlites likely crystallized from a sulfide saturated melt after S assimilation from the crust/regolith based on Δ³³S, δ³⁴S, and highly siderophile element patterns (Mari et al., 2019). Some amount of sulfur assimilation from S-rich regolith occurred after the nakhlite lavas were emplaced on or near the martian surface (Franz et al., 2014; Mari et al., 2019). Nakhlites are the only SNC meteorite class known to likely have crystallized from a sulfide-saturated melt, as shergottites crystallized from sulfide undersaturated parent magmas (Wang and Becker, 2017). Still, one-third of volcanic martian meteorite classes exhibiting geochemical characteristics consistent with crystallization from a sulfide-saturated melt implies that characteristically similar volcanism could have occurred throughout Mars' history. Sulfates are prevalent across the surface of Mars (Ehlmann and Edwards, 2014), S-rich regolith is effectively ubiquitous across the surface (Clark et al., 1976; Foley et al., 2003), and elemental sulfur concentrations measured by GRS range from 0.7–3.2 wt % at GRS spatial resolution (McLennan et al., 2010); thus a substantial amount of sulfur for assimilation by lavas commonly exists on the martian surface. Therefore, it is likely that lavas saturated with respect to sulfide after assimilation of S from the crust/regolith—similar to those from which nakhlites crystallized—were common throughout Mars' history, thus resulting in the formation of high sulfide concentration zones.

The martian mantle is generally expected to be more reducing than Earth's (Stanley et al., 2011; Armstrong et al., 2015), with estimates for mantle oxygen fugacity based on basalts measured in Gusev crater varying from iron-wüstite (IW) to IW+1 (Stanley et al., 2011). Mantle oxygen fugacities estimated from shergottites are [QFM-4]–[QFM-1] (Herd, 2003), and from nakhlites are ∼QFM (Mari et al., 2019). Earth's mantle has an oxygen fugacity range of [QFM-1]–[QFM+1] (Blundy et al., 1991; Brounce et al., 2017). Still, on Earth, pyrrhotite is found in volcanic rocks derived from partial melting of a more oxidized mantle relative to Mars (e.g., Desborough et al., 1968; Whitney, 1984), though concentrations are typically higher in silicic magmas than basalt, the latter of which is more common on Mars. Thus, magmatic sulfides are expected in some volcanic rocks derived from the wide variability in melt oxygen fugacities from the martian mantle. Martian mantle sulfur abundance estimates range from ∼400–2200 ppm (Wang and Becker, 2017), while Earth's mantle sulfur abundance is ∼250 ppm (McDonough and Sun, 1995). Additional sulfur is incorporated into martian melts via crustal assimilation of sulfates and/or S-rich regolith, as occurred in the nakhlite lavas (Franz et al., 2014; Mari et al., 2019). As such, sulfur activity in martian melts is typically higher than in most melts on Earth, while oxygen fugacity is lower than in most melts on Earth, both of which increase the likelihood of sulfide precipitation.

Sulfide oxidation has been invoked to explain multiple mineral assemblages found on the martian surface (King and McSween, 2005), including the hematite enrichment in Meridiani Planum (Zolotov and Shock, 2005), the Burns Formation—a sulfate-rich sandstone in Meridiani Planum (Zolotov and Shock, 2005)—sulfate-carbonate assemblages detected via orbital hyperspectral data (Dehouck et al., 2012), akaganeite detections made from orbit (Carter et al., 2015), and acidic fluid diagenetic mineral assemblages detected by the Curiosity rover (Rampe et al., 2017). In its type locality on Earth, akaganeite forms as an alteration product of pyrrhotite (Nambu, 1968). To obtain the abundances of sulfur needed to form the Burns Formation, pre-concentration of sulfur via magmatic or hydrothermal sulfides may be required (Zolotov and Shock, 2005). Furthermore, oxidation of sulfides to form the acid-mine-drainage-like mineral assemblages of sulfates and oxides seen on the martian surface provides a functional hypothesis for generation of these assemblages. Formation of these concentrated sulfides could have occurred on Mars via differentiation/crustal assimilation from ultramafic lavas (Burns and Fisher, 1990), the existence of which is supported by possible observations of ultramafic volcanism on the martian surface (Ruff et al., 2014, 2019; Kremer et al., 2019), ultramafic lithologies in martian meteorites (Burns and Fisher, 1990), and evidence that nakhlites crystallized from sulfide-saturated melts after S assimilation from the crust/regolith (Franz et al., 2014; Mari et al., 2019). Sulfides could also have been concentrated via hydrothermal activity (Section 4.3).

Mg-rich serpentine has been detected on Mars using hyperspectral images acquired by the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM; Murchie et al., 2007) (Ehlmann et al., 2010; Leask et al., 2018), but many Mg-rich serpentine detections have been called into question after discovery of a 2.1 μm spectral artifact in CRISM TRR3 data that can resemble the diagnostic absorption feature of this mineral (Leask et al., 2018). Still, at least 5 Mg-rich serpentine detections have been validated as unrelated to this artifact (Leask et al., 2018; Lin et al., 2021; Tarnas et al., 2021), demonstrating that serpentinization has occurred on Mars. The fluids generated via serpentinization are highly reducing, which can cause precipitation of reduced alloys and minerals such as awaruite (Lorand, 1985; Sleep et al., 2004) (NiFe alloy) and sulfides (Economou and Naldrett, 1984; Thalhammer et al., 1986; Shiga, 1987; Auclair et al., 1993; Wafik et al., 2001; Marques et al., 2007). As such, serpentinization would have generated reducing fluids on Mars that could have precipitated and concentrated sulfide minerals, making regions containing serpentine potential “hotzones” for ancient chemolithotrophic life that consumed H₂ from serpentinization, as well as modern chemolithotrophic life that may consume H₂ and SO₄ from radiolysis in sulfide-rich zones.

4.3. Possible settings of concentrated sulfides on Mars

Differentiated impact melt sheets: The Sudbury Impact Structure in Ontario, Canada, is one of the three largest discovered impact structures on Earth—the others being Chicxulub and Vredefort—and hosts concentrations of sulfides of high economic value. Sulfides were concentrated via gravitational differentiation from the impact melt sheet (Zieg and Marsh, 2005). Sulfur was likely introduced to the impact melt via assimilation from surrounding country rock during the depressurization melting that formed the Sudbury Igneous Complex (Walker et al., 1991; Dickin et al., 1992; Mungall et al., 2004). Because of the prevalence of sulfur on Mars in the form of sulfates and S-rich regolith, and the larger number of preserved impact structures on Mars relative to Earth, Sudbury-type differentiated impact melt sheet sulfide concentrations may be more common on Mars than they are on Earth (West and Clarke, 2010) (Fig. 3 Setting 1).

Layered intrusions: Sulfides are commonly concentrated in ultramafic-to-mafic layered intrusions on Earth, such as the Bushveld Igneous Complex in South Africa (Gain and Mostert, 1982), Voisey's Bay in Canada (Evans-Lamswood et al., 2000), Jinchuan in China (Chai and Naldrett, 1992), and the Noril'sk-Talnakh deposits in Siberia (Arndt et al., 2003). Sulfides are typically concentrated as residue in restricted conduits/channels and thus are often concentrated at the base of the layered intrusion, in sills, or in stratiform reef-style deposits (Lightfoot and Evans-Lamswood, 2015). Such intrusions typically form in intracratonic settings, often via mantle plume activity (Barnes et al., 2017). This is similar to settings of non-subvolcanic and subvolcanic intrusions formed on Mars, which is a single-plate planet. There is evidence for ultramafic volcanism on Mars (Kremer et al., 2019; Ruff et al., 2019), and basaltic volcanism is prevalent across the entire planet (Bandfield et al., 2000). Mantle plumes have caused volcanism on Mars, likely forming the Tharsis and Elysium volcanic bulges (Fuller and Head, 2003; Redmond and King, 2004; Hynek et al., 2011). It is thus plausible that both subvolcanic and non-subvolcanic ultramafic-to-mafic magmatic intrusions exist on Mars, providing structures for potential concentration of sulfides (Fig. 3 Settings 2 and 7).

Hydrothermal sulfides: Sulfides are concentrated by hydrothermal fluids that are S-rich and reducing. Many hydrothermally formed sulfide deposits are associated with serpentinites, due to the highly reducing fluids formed during the process of serpentinization (Lorand, 1985; Sleep et al., 2004) causing formation of metal alloys (Lorand, 1985) and sulfides (Economou and Naldrett, 1984; Thalhammer et al., 1986; Shiga, 1987; Auclair et al., 1993; Wafik et al., 2001; Marques et al., 2007). Hydrothermal sulfides are found in regolith breccia martian meteorites (Lorand et al., 2015, 2018a; Wittmann et al., 2015; Hewins et al., 2017). Minerals potentially produced by hydrothermal alteration, including smectites, carbonates, chlorite, hydrated silica, and zeolites (Ehlmann et al., 2009), and small amounts of Mg-rich serpentine (Ehlmann et al., 2010; Leask et al., 2018; Lin et al., 2021; Tarnas et al., 2021) have been detected on the martian surface from orbit with CRISM. Evidence for past hydrothermal alteration has been seen by every Mars rover. NASA's Opportunity rover observed evidence for multiple diagenetic episodes in the Burns Formation (McLennan et al., 2005), as well as hydrothermal alteration of bedrock due to meteorite impact at Endeavour crater (Squyres et al., 2012; Arvidson et al., 2014). NASA's Spirit rover observed carbonate-bearing olivine-rich rocks (Comanche) formed via alteration of ultramafic tuff (Algonquin) (Morris et al., 2010; Ruff et al., 2014), silica deposits (Ruff et al., 2011) that likely formed in hot spring settings (Ruff and Farmer, 2016; Ruff et al., 2020), and hydrothermal alteration of Home Plate resulting in precipitation of silica (Squyres et al., 2008). NASA's Curiosity rover has found evidence for multiple diagenetic episodes in the stratigraphy of Mount Sharp, including cross-cutting Ca-sulfate veins (Yen et al., 2017), jarosite veins dated to 2.1 ± 4 Gyr (Martin et al., 2017), and evidence for diagenetic episodes that cross-cut multiple stratigraphically distinct units (Sheppard et al., 2020). Sulfides could also be concentrated via hydrothermal activity during submarine volcanism—as likely occurred in Eridania Basin (Michalski et al., 2017). As such, hydrothermal alteration has occurred on Mars in a wide variety of geochemical settings, including some that precipitated pyrite in regolith breccias (Lorand et al., 2015, 2018a; Wittmann et al., 2015; Hewins et al., 2017) and some that formed serpentinite (Ehlmann et al., 2010; Leask et al., 2018). It is plausible that higher concentrations of sulfides than those found in NWA 7533 and NWA 7034—which were ejected from Mars by the same impact event (Wittmann et al., 2015)—could form from hydrothermal alteration in other regions of the martian crust (Fig. 3 Settings 3 and 5).

Sulfide-bearing sediments: Mount Sharp in Gale crater is an example of a sulfide-bearing sediment mound on Mars (McAdam et al., 2014; Franz et al., 2017; Wong et al., 2020) (Fig. 3 Setting 4). These sulfides are diagenetic/epigenetic, likely resulting from thermal reduction of preexisting sulfates (Franz et al., 2017). It is possible that a similar sulfate thermal reduction process could occur in other sulfate-bearing martian sediments, which are widespread (Ehlmann and Edwards, 2014), thus generating a higher concentration of sulfides in those localities. Cu-enrichments in Gale crater are potentially due to presence of detrital sulfides (Payré et al., 2019). It is therefore plausible that sulfides could be concentrated in martian sediments in detrital form and also via diagenetic processes.

Differentiated ultramafic-to-mafic lava flows: Examples of sulfide concentrations in komatiitic lava flows on Earth include the Kambalda and Perseverance deposits in Australia. Sulfides are also concentrated in picrites, including the Pechenga deposits in Russia (Barnes et al., 2017). Due to the presence of ultramafic lithologies in martian meteorites, the possibility of sulfide concentration in komatiite-type settings has been known for decades (Burns and Fisher, 1990). Since then, further evidence for ultramafic volcanism on Mars has been presented (Ruff et al., 2014, 2019; Kremer et al., 2019). The lavas that precipitated the nakhlites assimilated sulfur from the crust, forming a sulfide-saturated melt (Franz et al., 2014; Mari et al., 2019). Thus, there is already evidence for some degree of sulfide concentration in differentiated ultramafic-to-mafic lava flows on Mars via the martian meteorites (Fig. 3 Setting 6).

5. Conclusions

The results presented here demonstrate that radiolysis in the martian meteorite source lithologies would produce sufficient H₂ and sulfate to sustain sulfate-reducing bacteria wherever groundwater is present. The regolith breccia source lithology, which is within the southern highlands (Cassata et al., 2018), could sustain the highest concentrations of sulfate-reducing bacteria, with shergottite and nakhlite lithologies supporting the second highest concentrations. The calculated supportable sulfate-reducing bacteria densities in the regolith breccia, shergottite, and nakhlite source lithologies are comparable to those measured in Earth's deep subsurface (Table 3; Fig. 2). These cell density estimates are conservative (Sections 2.2–2.6), and actual cell densities could be higher in any martian groundwater that exists today.

If Earth-like chemolithotrophic life ever existed on Mars and survived until the present, it could potentially be sustained in subsurface regions by harnessing the same radiolytically derived energy that drives the metabolisms of microbial ecosystems on Earth sustained in groundwaters >10⁶ to 10⁹ years old (Lin et al., 2006; Chivian et al., 2008; Lollar et al., 2019). Furthermore, this subsurface environment has likely remained habitable since the Noachian (Tarnas et al., 2018), making it the longest-lived habitable environment on Mars and the most likely refugia for any life on the planet (Michalski et al., 2018; National Academies of Sciences, Engineering, and Medicine, 2019). The parameters outlined here—radionuclide abundance, sulfide abundance, average sulfide grain size, and porosity—along with characterization of the locations of liquid groundwater, can be used to constrain an optimal landing site for a mission to detect Earth-like extant sulfate-reducing martian life in the subsurface. Additional microbial metabolisms, for which redox energy availability is more difficult to model, could also exist in the martian subsurface, as they do in Earth's subsurface, potentially increasing the measured biomass in any subsurface region that contains groundwater. This should be considered in the context of future extant life-detection missions to Mars.

Footnotes

Acknowledgments

We thank Norman Sleep and an anonymous reviewer for feedback that improved the quality of this manuscript. We thank Paul Niles, Ralph Milliken, and Steven D'Hondt for discussions. Thanks to the Astromaterial Curation Group at NASA JSC, especially Kevin Righter, for providing a compilation table of martian meteorite elemental abundances. J.D.T. gratefully acknowledges support from the NASA Postdoctoral Program and a Brown University Dissertation Fellowship. J.F.M., B.S.L., V.S., and J.R.M. acknowledge support from the Canadian Institute for Advanced Research (CIFAR). A.-C.P. gratefully acknowledges the financial support and endorsement from the DLR Management Board Young Research Group Leader Program and the Executive Board Member for Space Research and Technology. Part of the research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). © 2020. All rights reserved.

Abbreviations Used

Associate Editor: Petra Rettberg

References

Adamiec

, and Aitken

(1998) Dose-rate conversion factors: update. Ancient TL, 16:37–50.

Agee

C.B.

, Wilson

N.V.

, McCubbin

F.M.

, Ziegler

, Polyak

V.J.

, Sharp

Z.D.

, and Steele

(2013) Unique meteorite from early Amazonian Mars: Water-rich basaltic breccia Northwest Africa 7034. Science, 339:780–785.

Altair

, de Avellar

M.G.B.

, Rodrigues

, and Galante

(2018) Microbial habitability of Europa sustained by radioactive sources. Sci Rep, 8, doi:10.1038/s41598-017-18470-z.

Armstrong

L.S.

, Hirschmann

M.M.

, Stanley

B.D.

, Falksen

E.G.

, and Jacobsen

S.D.

(2015) Speciation and solubility of reduced C–O–H–N volatiles in mafic melt: implications for volcanism, atmospheric evolution, and deep volatile cycles in the terrestrial planets. Geochim Cosmochim Acta, 171:283–302.

Arndt

N.T.

, Czamanske

G.K.

, Walker

R.J.

, Chauvel

, and Fedorenko

V.A.

(2003) Geochemistry and origin of the intrusive hosts of the Noril'sk-Talnakh Cu-Ni-PGE sulfide deposits. Econ Geol, 98:495–515.

Arvidson

R.E.

, Squyres

S.W.

, Bell

J.F.

, Catalano

J.G.

, Clark

B.C.

, Crumpler

L.S.

, de Souza

P.A.

, Fairén

A.G.

, Farrand

W.H.

, Fox

V.K.

, et al. (2014) Ancient aqueous environments at Endeavour Crater, Mars. Science, 343, doi:10.1126/science.1248097.

Auclair

, Gauthier

, Trottier

, Jebrak

, and Chartrand

(1993) Mineralogy, geochemistry, and paragenesis of the Eastern Metals serpentinite-associated Ni-Cu-Zn deposit, Quebec Appalachians. Econ Geol, 88:123–138.

Bandfield

J.L.

, Hamilton

V.E.

, and Christensen

P.R.

(2000) A global view of martian surface compositions from MGS-TES. Science, 287:1626–1630.

Barnes

S.J.

, Holwell

D.A.

, and Vaillant

M.L.

(2017) Magmatic sulfide ore deposits. Elements, 13:89–95.

10.

Baumgartner

R.J.

, Fiorentini

M.L.

, Baratoux

, Micklethwaite

, Sener

A.K.

, Lorand

J.P.

, and McCuaig

T.C.

(2015) Magmatic controls on the genesis of Ni–Cu±(PGE) sulphide mineralisation on Mars. Ore Geology Reviews, 65:400–412.

11.

Baumgartner

R.J.

, Fiorentini

M.L.

, Lorand

J.-P.

, Baratoux

, Zaccarini

, Ferrière

, Prašek

M.K.

, and Sener

(2017) The role of sulfides in the fractionation of highly siderophile and chalcophile elements during the formation of martian shergottite meteorites. Geochim Cosmochim Acta, 210:1–24.

12.

Blake

, Vaniman

, Achilles

, Anderson

, Bish

, Bristow

, Chen

, Chipera

, Crisp

, Des Marais

, Downs

R.T.

, Farmer

, Feldman

, Fonda

, Gailhanou

, Ma

, Ming

D.W.

, Morris

R.V.

, Sarrazin

, Stolper

, Treiman

, and Yen

(2012) Characterization and calibration of the CheMin mineralogical instrument on Mars Science Laboratory. Space Sci Rev, 170:341–399. https://link-springer-com-443.web.bisu.edu.cn/article/10.1007/s11214-012-9905-1

13.

Blundy

J.D.

, Brodholt

J.P.

, and Wood

B.J.

(1991) Carbon–fluid equilibria and the oxidation state of the upper mantle. Nature, 349:321–324.

14.

Boynton

, Feldman

, Mitrofanov

. et al. (2004) The Mars Odyssey Gamma-Ray Spectrometer Instrument Suite. Space Sci Rev, 110:37–83. https://link-springer-com-443.web.bisu.edu.cn/article/10.1023/B:SPAC.0000021007.76126.15

15.

Boynton

W.V.

, Taylor

G.J.

, Evans

L.G.

, Reedy

R.C.

, Starr

, Janes

D.M.

, Kerry

K.E.

, Drake

D.M.

, Kim

K.J.

, Williams

R.M.S.

, et al. (2007) Concentration of H, Si, Cl, K, Fe, and Th in the low- and mid-latitude regions of Mars. J Geophys Res Planets, 112, doi:10.1029/2007JE002887.

16.

Bragg

M.A.

, and Kleeman

B.S

c. (1905) XXXIX. On the α particles of radium, and their loss of range in passing through various atoms and molecules. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science , 10:318–340.

17.

Broquet

, and Wieczorek

M.A.

(2019) The gravitational signature of martian volcanoes. J Geophys Res Planets, 124:2054–2086.

18.

Brounce

, Stolper

, and Eiler

(2017) Redox variations in Mauna Kea lavas, the oxygen fugacity of the Hawaiian plume, and the role of volcanic gases in Earth's oxygenation. Proc Natl Acad Sci USA, early edition,, doi:10.1073/pnas.1619527114.

19.

Buck

E.C.

, Wittman

R.S.

, Skomurski

F.N.

, Cantrell

K.J.

, McNamara

B.K.

, and Soderquist

C.Z.

(2012) Radiolysis Process Model., Pacific Northwest National Laboratory (PNNL), Richland, WA.

20.

Burns

R.G.

, and Fisher

D.S.

(1990) Evolution of sulfide mineralization on Mars. J Geophys Res Solid Earth,, 95:14169–14173.

21.

Carter

, Viviano-Beck

, Loizeau

, Bishop

, and Le Deit

(2015) Orbital detection and implications of akaganéite on Mars. Icarus, 253:296–310.

22.

Cassata

W.S.

, Cohen

B.E.

, Mark

D.F.

, Trappitsch

, Crow

C.A.

, Wimpenny

, Lee

M.R.

, and Smith

C.L.

(2018) Chronology of martian breccia NWA 7034 and the formation of the martian crustal dichotomy. Sci Adv, 4, doi:10.1126/sciadv.aap8306.

23.

Chai

, and Naldrett

A.J.

(1992) Characteristics of Ni-Cu-PGE mineralization and genesis of the Jinchuan Deposit, Northwest China. Econ Geol, 87:1475–1495.

24.

Chevrier

, Lorand

J.-P.

, and Sautter

(2011) Sulfide petrology of four nakhlites: Northwest Africa 817, Northwest Africa 998, Nakhla, and Governador Valadares. Meteorit Planet Sci, 46:769–784.

25.

Chivian

, Brodie

E.L.

, Alm

E.J.

, Culley

D.E.

, Dehal

P.S.

, DeSantis

T.Z.

, Gihring

T.M.

, Lapidus

, Lin

L.-H.

, Lowry

S.R.

, et al. (2008) Environmental genomics reveals a single-species ecosystem deep within Earth. Science, 322:275–278.

26.

Clark

B.C.

, Baird

A.K.

, Rose

H.J.

, Toulmin

, Keil

, Castro

A.J.

, Kelliher

W.C.

, Rowe

C.D.

, and Evans

P.H.

(1976) Inorganic analyses of martian surface samples at the Viking landing sites. Science, 194:1283–1288.

27.

Cockell

C.S.

, Voytek

M.A.

, Gronstal

A.L.

, Finster

, Kirshtein

J.D.

, Howard

, Reitner

, Gohn

G.S.

, Sanford

W.E.

, Horton

J.W.

, et al. (2012) Impact disruption and recovery of the deep subsurface biosphere. Astrobiology, 12:231–246.

28.

Dehouck

, Chevrier

, Gaudin

, Mangold

, Mathé

P.-E.

, and Rochette

(2012) Evaluating the role of sulfide-weathering in the formation of sulfates or carbonates on Mars. Geochim Cosmochim Acta, 90:47–63.

29.

Desborough

G.A.

, Anderson

A.T.

, and Wright

T.L.

(1968) Mineralogy of sulfides from certain Hawaiian basalts. Econ Geol, 63:636–644.

30.

Dickin

A.P.

, Richardson

J.M.

, Crocket

J.H.

, McNutt

R.H.

, and Peredery

W.V.

(1992) Osmium isotope evidence for a crustal origin of platinum group elements in the Sudbury nickel ore, Ontario, Canada. Geochim Cosmochim Acta, 56:3531–3537.

31.

Ding

, Dasgupta

, and Tsuno

(2014) Sulfur concentration of martian basalts at sulfide saturation at high pressures and temperatures—implications for deep sulfur cycle on Mars. Geochim Cosmochim Acta, 131:227–246.

32.

Dzaugis

, Spivack

A.J.

, and D'Hondt

(2018) Radiolytic H₂ production in martian environments. Astrobiology, 18:1137–1146.

33.

Economou

M.I.

, and Naldrett

A.J.

(1984) Sulfides associated with podiform bodies of chromite at Tsangli, Eretria, Greece. Miner Depos, 19:289–297.

34.

Ehlmann

B.L.

, and Edwards

C.S.

(2014) Mineralogy of the martian surface. Annu Rev Earth Planet Sci, 42:291–315.

35.

Ehlmann

B.L.

, Mustard

J.F.

, Swayze

G.A.

, Clark

R.N.

, Bishop

J.L.

, Poulet

, Des Marais

D.J.

, Roach

L.H.

, Milliken

R.E.

, Wray

J.J.

, et al. (2009) Identification of hydrated silicate minerals on Mars using MRO-CRISM: geologic context near Nili Fossae and implications for aqueous alteration. J Geophys Res Planets, 114, doi:10.1029/2009JE003339.

36.

Ehlmann

B.L.

, Mustard

J.F.

, and Murchie

S.L.

(2010) Geologic setting of serpentine deposits on Mars. Geophys Res Lett, 37, doi:10.1029/2010GL042596.

37.

Evans-Lamswood

D.M.

, Butt

D.P.

, Jackson

R.S.

, Lee

D.V.

, Muggridge

M.G.

, Wheeler

R.I.

, and Wilton

D.H.C.

(2000) Physical controls associated with the distribution of sulfides in the Voisey's Bay Ni-Cu-Co deposit, Labrador. Econ Geol, 95:749–769.

38.

Foley

C.N.

, Economou

, and Clayton

R.N.

(2003) Final chemical results from the Mars Pathfinder alpha proton X-ray spectrometer. J Geophys Res Planets, 108, doi:10.1029/2002JE002019.

39.

Franz

H.B.

, Kim

S.-T.

, Farquhar

, Day

J.M.D.

, Economos

R.C.

, McKeegan

K.D.

, Schmitt

A.K.

, Irving

A.J.

, Hoek

, and Dottin

, III. (2014) Isotopic links between atmospheric chemistry and the deep sulphur cycle on Mars. Nature, 508:364–368.

40.

Franz

H.B.

, McAdam

A.C.

, Ming

D.W.

, Freissinet

, Mahaffy

P.R.

, Eldridge

D.L.

, Fischer

W.W.

, Grotzinger

J.P.

, House

C.H.

, Hurowitz

J.A.

, et al. (2017) Large sulfur isotope fractionations in martian sediments at Gale crater. Nat Geosci, 10:658–662.

41.

Franz

H.B.

, Wu

, Farquhar

, and Irving

A.J.

(2019) A new type of isotopic anomaly in shergottite sulfides. Meteorit Planet Sci, 54:3036–3051.

42.

Fuller

E.R.

, and Head

J.W.

(2003) Olympus Mons, Mars: detection of extensive preaureole volcanism and implications for initial mantle plume behavior. Geology, 31:175–178.

43.

Gain

S.B.

, and Mostert

A.B.

(1982) The geological setting of the platinoid and base metal sulfide mineralization in the Platreef of the Bushveld Complex in Drenthe, North of Potgietersrus. Econ Geol, 77:1395–1404.

44.

Gattacceca

, Hewins

R.H.

, Lorand

J.-P.

, Rochette

, Lagroix

, Cournède

, Uehara

, Pont

, Sautter

, Scorzelli

R.B.

, et al. (2013) Opaque minerals, magnetic properties, and paleomagnetism of the Tissint martian meteorite. Meteorit Planet Sci, 48:1919–1936.

45.

Gellert

, Rieder

, Anderson

R.C.

, Brückner

, Clark

B.C.

, Dreibus

, Economou

, Klingelhöfer

, Lugmair

G.W.

, Ming

D.W.

, et al. (2004) Chemistry of rocks and soils in Gusev crater from the Alpha Particle X-ray Spectrometer. Science, 305:829–832.

46.

Goossens

, Sabaka

T.J.

, Genova

, Mazarico

, Nicholas

J.B.

, and Neumann

G.A.

(2017) Evidence for a low bulk crustal density for Mars from gravity and topography. Geophys Res Lett, 44:7686–7694.

47.

Greenwood

J.P.

, Mojzsis

S.J.

, and Coath

C.D.

(2000a) Sulfur isotopic compositions of individual sulfides in martian meteorites ALH84001 and Nakhla: implications for crust–regolith exchange on Mars. Earth Planet Sci Lett, 184:23–35.

48.

Greenwood

J.P.

, Riciputi

L.R.

, McSween

H.Y.

, and Taylor

L.A.

(2000b) Modified sulfur isotopic compositions of sulfides in the nakhlites and Chassigny. Geochim Cosmochim Acta, 64:1121–1131.

49.

Grimm

R.E.

, Harrison

K.P.

, Stillman

D.E.

, and Kirchoff

M.R.

(2017) On the secular retention of ground water and ice on Mars. J Geophys Res Planets, 122:94–109.

50.

Hahn

B.C.

, McLennan

S.M.

, and Klein

E.C.

(2011) Martian surface heat production and crustal heat flow from Mars Odyssey gamma-ray spectrometry. Geophys Res Lett, 38, doi:10.1029/2011GL047435.

51.

Herd

C.D.K.

(2003) The oxygen fugacity of olivine-phyric martian basalts and the components within the mantle and crust of Mars. Meteorit Planet Sci, 38:1793–1805.

52.

Hewins

R.H.

, Zanda

, Humayun

, Nemchin

, Lorand

J.-P.

, Pont

, Deldicque

, Bellucci

J.J.

, Beck

, Leroux

, et al. (2017) Regolith breccia Northwest Africa 7533: mineralogy and petrology with implications for early Mars. Meteorit Planet Sci, 52:89–124.

53.

Hewins

R.H.

, Humayun

, Barrat

J.-A.

, Zanda

, Lorand

J.-P.

, Pont

, Assayag

, Cartigny

, Yang

, and Sautter

(2020) Northwest Africa 8694, a ferroan chassignite: bridging the gap between nakhlites and chassignites. Geochim Cosmochim Acta, 282:201–226.

54.

Holland

, Lollar

B.S.

, Li

, Lacrampe-Couloume

, Slater

G.F.

, and Ballentine

C.J.

(2013) Deep fracture fluids isolated in the crust since the Precambrian era. Nature, 497:357–360.

55.

Humayun

, Yang

, Irving

A.J.

, and Righter

(2019) Sulfide assimilation and mineralization in ancient (2.4 Ga) shergottites [abstract 6380]. In 82nd Annual Meeting of The Meteoritical Society, Lunar and Planetary Institute, Houston, LPI Contribution No. 2157.

56.

Hynek

B.M.

, Robbins

S.J.

, Šrámek

, and Zhong

S.J.

(2011) Geological evidence for a migrating Tharsis plume on early Mars. Earth Planet Sci Lett, 310:327–333.

57.

Johnson

C.L.

, Mittelholz

, Langlais

, Russell

C.T.

, Ansan

, Banfield

, Chi

P.J.

, Fillingim

M.O.

, Forget

, Haviland

H.F.

, et al. (2020) Crustal and time-varying magnetic fields at the InSight landing site on Mars. Nat Geosci, 13:199–204.

58.

King

P.L.

, and McSween

H.Y.

(2005) Effects of H₂O, pH, and oxidation state on the stability of Fe minerals on Mars. J Geophys Res Planets, 110, doi:10.1029/2005JE002482.

59.

Klein

, Tarnas

J.D.

, and Bach

(2020) Abiotic sources of molecular hydrogen on Earth. Elements, 16:19–24.

60.

Kremer

C.H.

, Mustard

J.F.

, and Bramble

M.S.

(2019) A widespread olivine-rich ash deposit on Mars. Geology, 47:677–681.

61.

Kumagai

, Kimura

, Taguchi

, Nagaishi

, Yamagishi

, and Kimura

(2013) Hydrogen production in gamma radiolysis of the mixture of mordenite and seawater. Journal of Nuclear Science and Technology, 50:130–138.

62.

Langlais

, Thébault

, Houliez

, Purucker

M.E.

, and Lillis

R.J.

(2019) A new model of the crustal magnetic field of Mars using MGS and MAVEN. J Geophys Res Planets, 124:1542–1569.

63.

LaVerne

J.A.

, and Tandon

(2005) H₂ and Cl₂ production in the radiolysis of calcium and magnesium chlorides and hydroxides. J Phys Chem A, 109:2861–2865.

64.

Leask

E.K.

, Ehlmann

B.L.

, Dundar

M.M.

, Murchie

S.L.

, and Seelos

F.P.

(2018) Challenges in the search for perchlorate and other hydrated minerals with 2.1-μm absorptions on Mars. Geophys Res Lett, 45:12180–12189.

65.

Le Deit

, Mangold

, Forni

, Cousin

, Lasue

, Schröder

, Wiens

R.C.

, Sumner

, Fabre

, Stack

K.M.

, et al. (2016) The potassic sedimentary rocks in Gale Crater, Mars, as seen by ChemCam on board Curiosity. J Geophys Res Planets, 121:784–804.

66.

Lefticariu

, Pratt

L.A.

, LaVerne

J.A.

, and Schimmelmann

(2010) Anoxic pyrite oxidation by water radiolysis products—a potential source of biosustaining energy. Earth Planet Sci Lett, 292:57–67.

67.

Lewis

K.W.

, Peters

, Gonter

, Morrison

, Schmerr

, Vasavada

A.R.

, and Gabriel

(2019) A surface gravity traverse on Mars indicates low bedrock density at Gale crater. Science, 363:535–537.

68.

, Wing

B.A.

, Bui

T.H.

, McDermott

J.M.

, Slater

G.F.

, Wei

, Lacrampe-Couloume

, and Sherwood Lollar

(2016) Sulfur mass-independent fractionation in subsurface fracture waters indicates a long-standing sulfur cycle in Precambrian rocks. Nat Commun, 7, doi:10.1038/ncomms13252.

69.

Lightfoot

P.C.

, and Evans-Lamswood

(2015) Structural controls on the primary distribution of mafic–ultramafic intrusions containing Ni–Cu–Co–(PGE) sulfide mineralization in the roots of large igneous provinces. Ore Geology Reviews, 64:354–386.

70.

Lin

, Tarnas

J.D.

, Mustard

J.F.

, et al. (2021) Dynamic aperture factor analysis/target transformation (DAFA/TT) for Mg-serpentine and Mg-carbonate mapping on Mars with CRISM near-infrared data. Icarus 355:. 114168.

71.

Lin

L.-H.

, Hall

, Lippmann-Pipke

, Ward

J.A.

, Sherwood Lollar

, DeFlaun

, Rothmel

, Moser

, Gihring

T.M.

, Mislowack

, and Onstott

T.C.

(2005a) Radiolytic H₂ in continental crust: Nuclear power for deep subsurface microbial communities. Geochem Geophys Geosyst, 6, doi:10.1029/2004GC000907.

72.

Lin

L.-H.

, Slater

G.F.

, Sherwood Lollar

, Lacrampe-Couloume

, and Onstott

T.C.

(2005b) The yield and isotopic composition of radiolytic H₂, a potential energy source for the deep subsurface biosphere. Geochim Cosmochim Acta, 69:893–903.

73.

Lin

L.-H.

, Wang

P.-L.

, Rumble

, Lippmann-Pipke

, Boice

, Pratt

L.M.

, Sherwood Lollar

, Brodie

E.L.

, Hazen

T.C.

, Andersen

G.L.

, et al. (2006) Long-term sustainability of a high-energy, low-diversity crustal biome. Science, 314:479–482.

74.

Lollar

G.S.

, Warr

, Telling

, Osburn

M.R.

, and Sherwood Lollar

(2019) ‘Follow the water’: hydrogeochemical constraints on microbial investigations 2.4 km below surface at the Kidd Creek deep fluid and deep life observatory. Geomicrobiol J, 36:859–872.

75.

Lorand

J.-P.

(1985) The behaviour of the upper mantle sulfide component during the incipient alteration of “Alpine”-type peridotites as illustrated by the Beni Bousera (northern Morocco) and Ronda (southern Spain) ultramafic bodies. TMPM Tschermaks Petr Mitt, 34:183–209.

76.

Lorand

J.-P.

, Chevrier

, and Sautter

(2005) Sulfide mineralogy and redox conditions in some shergottites. Meteorit Planet Sci, 40:1257–1272.

77.

Lorand

J.-P.

, Barrat

J.-A.

, Chevrier

, Sautter

, and Pont

(2012) Metal-saturated sulfide assemblages in NWA 2737: evidence for impact-related sulfur devolatilization in martian meteorites. Meteorit Planet Sci, 47:1830–1841.

78.

Lorand

J.-P.

, Hewins

R.H.

, Remusat

, Zanda

, Pont

, Leroux

, Marinova

, Jacob

, Humayun

, Nemchin

, Grange

, Kennedy

, and Göpel

(2015) Nickeliferous pyrite tracks pervasive hydrothermal alteration in martian regolith breccia: a study in NWA 7533. Meteorit Planet Sci, 50:2099–2120.

79.

Lorand

J.-P.

, Hewins

R.H.

, Humayun

, Remusat

, Zanda

, La

, and Pont

(2018a) Chalcophile-siderophile element systematics of hydrothermal pyrite from martian regolith breccia NWA 7533. Geochim Cosmochim Acta, 241:134–149.

80.

Lorand

J.-P.

, Pont

, Chevrier

, Luguet

, Zanda

, and Hewins

(2018b) Petrogenesis of martian sulfides in the Chassigny meteorite. Am Mineral, 103:872–885.

81.

Magnabosco

, Lin

L.-H.

, Dong

, Bomberg

, Ghiorse

, Stan-Lotter

, Pedersen

, Kieft

T.L.

, van Heerden

, and Onstott

T.C.

(2018) The biomass and biodiversity of the continental subsurface. Nat Geosci, 11:707–717.

82.

Mahaffy

P.M.

, Webster

C.R.

, Cabane

, Conrad

P.C.

, Coll

, SAM team, (2012) The Sample Analysis at Mars

Investigation and instrument suite

. Space Sci Rev. Available online at https://link-springer-com-443.web.bisu.edu.cn/article/10.1007/s11214-012-9879-z

83.

Mari

, Riches

A.J.V.

, Hallis

L.J.

, Marrocchi

, Villeneuve

, Gleissner

, Becker

, and Lee

M.R.

(2019) Syneruptive incorporation of martian surface sulphur in the nakhlite lava flows revealed by S and Os isotopes and highly siderophile elements: implication for mantle sources in Mars. Geochim Cosmochim Acta, 266:416–434.

84.

Marques

A.F.A.

, Barriga

F.J.A

.S., and Scott

S.D.

(2007) Sulfide mineralization in an ultramafic-rock hosted seafloor hydrothermal system: from serpentinization to the formation of Cu–Zn–(Co)-rich massive sulfides. Mar Geol, 245:20–39.

85.

Martin

P.E.

, Farley

K.A.

, Baker

M.B.

, Malespin

C.A.

, Schwenzer

S.P.

, Cohen

B.A.

, Mahaffy

P.R.

, McAdam

A.C.

, Ming

D.W.

, Vasconcelos

P.M.

, and Navarro-González

(2017) A two-step K-Ar experiment on Mars: dating the diagenetic formation of jarosite from Amazonian groundwaters. J Geophys Res Planets, 122:2803–2818.

86.

McAdam

A.C.

, Franz

H.B.

, Sutter

, Archer

P.D.

, Freissinet

, Eigenbrode

J.L.

, Ming

D.W.

, Atreya

S.K.

, Bish

D.L.

, Blake

D.F.

, et al. (2014) Sulfur-bearing phases detected by evolved gas analysis of the Rocknest aeolian deposit, Gale crater, Mars. J Geophys Res Planets, 119:373–393.

87.

McDonough

W.F.

, and Sun

(1995) The composition of the Earth. Chem Geol, 120:223–253.

88.

McLennan

S.M.

, Bell

J.F.

, Calvin

W.M.

, Christensen

P.R.

, Clark

B.C.

, de Souza

P.A.

, Farmer

, Farrand

W.H.

, Fike

D.A.

, Gellert

, et al. (2005) Provenance and diagenesis of the evaporite-bearing Burns formation, Meridiani Planum, Mars. Earth Planet Sci Lett, 240:95–121.

89.

McLennan

S.M.

, Boynton

W.V.

, Karunatillake

, Hahn

B.C.

, Taylor

G.J.

, and the Mars Odyssey GRS

Team.

(2010) Distribution of sulfur on the surface of Mars determined by the 2001 Mars Odyssey Gamma Ray Spectrometer [abstract 2174]. In 41st Lunar and Planetary Science Conference, Lunar and Planetary Institute, Houston.

90.

McMahon

, Parnell

, Ponicka

, Hole

, and Boyce

(2013) The habitability of vesicles in martian basalt. Astronomy & Geophysics, 54, doi:10.1093/astrogeo/ats035.

91.

McSween

H.Y.

(2015) Petrology on Mars. Am Mineral, 100:2380–2395.

92.

Meyer

(2016) The Martian Meteorite Compendium. Available online at https://curator.jsc.nasa.gov/antmet/mmc

93.

Michalski

J.R.

, Cuadros

, Niles

P.B.

, Parnell

, Rogers

A.D.

, and Wright

S.P.

(2013) Groundwater activity on Mars and implications for a deep biosphere. Nat Geosci, 6, 133–138.

94.

Michalski

J.R.

, Dobrea

E.Z.N.

, Niles

P.B.

, and Cuadros

(2017) Ancient hydrothermal seafloor deposits in Eridania Basin on Mars. Nat Commun, 8, doi:10.1038/ncomms15978.

95.

Michalski

J.R.

, Onstott

T.C.

, Mojzsis

S.J.

, Mustard

, Chan

Q.H.S.

, Niles

P.B.

, and Johnson

S.S.

(2018) The martian subsurface as a potential window into the origin of life. Nat Geosci, 11:21–26.

96.

Morris

R.V.

, Ruff

S.W.

, Gellert

, Ming

D.W.

, Arvidson

R.E.

, Clark

B.C.

, Golden

D.C.

, Siebach

, Klingelhöfer

, Schröder

, et al. (2010) Identification of carbonate-rich outcrops on Mars by the Spirit Rover. Science, 329:421–424.

97.

Morrison

S.M.

, Downs

R.T.

, Blake

D.F.

, Vaniman

D.T.

, Ming

D.W.

, Hazen

R.M.

, Treiman

A.H.

, Achilles

C.N.

, Yen

A.S.

, Morris

R.V.

, et al. (2018) Crystal chemistry of martian minerals from Bradbury Landing through Naukluft Plateau, Gale crater, Mars. Am Mineral, 103:857–871.

98.

Mungall

J.E.

, Ames

D.E.

, and Hanley

J.J.

(2004) Geochemical evidence from the Sudbury structure for crustal redistribution by large bolide impacts. Nature, 429:546–548.

99.

Murchie

, Arvidson

, Bedini

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

100.

Nambu

(1968) New mineral akaganeite, β-FeOOH, from Akagane Mine, Iwate Prefecture, Japan. The Journal of the Japanese Association of Mineralogists, Petrologists and Economic Geologists, 59:143–151.

101.

National Academies of

Sciences

, Engineering, and Medicine . (2019) An Astrobiology Strategy for the Search for Life in the Universe, National Academies Press, Washington, DC.

102.

Nogami

H.H.

, and Hurley

P.M.

(1948) The absorption factor in counting alpha rays from thick mineral sources. Eos, 29:335–340.

103.

Ojha

, Buffo

, Karunatillake

, and Siegler

(2020) Groundwater production from geothermal heating on early Mars and implication for early martian habitability. Science Advances 6, no. 49, eabb1669, doi: 10.1126/sciadv.abb1669

104.

Onstott

T.C.

, Moser

D.P.

, Pfiffner

S.M.

, Fredrickson

J.K.

, Brockman

F.J.

, Phelps

T.J.

, White

D.C.

, Peacock

, Balkwill

, Hoover

, et al. (2003) Indigenous and contaminant microbes in ultradeep mines. Environ Microbiol, 5:1168–1191.

105.

Onstott

T.C.

, McGown

, Kessler

, Sherwood Lollar

, Lehmann

K.K.

, and Clifford

S.M.

(2006) Martian CH₄: sources, flux, and detection. Astrobiology, 6:377–395.

106.

Onstott

T.C.

, Ehlmann

.l., Sapers

, Coleman

, Ivarsson

, Marlow

J.J.

, Neubeck

, and Niles

(2019) Paleo-rock-hosted life on Earth and the search on Mars: a review and strategy for exploration. Astrobiology, 19:1230–1262.

107.

Payré

, Fabre

, Sautter

, Cousin

, Mangold

, Le Deit

, Forni

, Goetz

, Wiens

R.C.

, Gasnault

, et al. (2019) Copper enrichments in the Kimberley formation in Gale crater, Mars: evidence for a Cu deposit at the source. Icarus, 321:736–751.

108.

Rampe

E.B.

, Ming

D.W.

, Blake

D.F.

, Bristow

T.F.

, Chipera

S.J.

, Grotzinger

J.P.

, Morris

R.V.

, Morrison

S.M.

, Vaniman

D.T.

, Yen

A.S.

, et al. (2017) Mineralogy of an ancient lacustrine mudstone succession from the Murray formation, Gale crater, Mars. Earth Planet Sci Lett, 471:172–185.

109.

Redmond

H.L.

, and King

S.D.

(2004) A numerical study of a mantle plume beneath the Tharsis Rise: reconciling dynamic uplift and lithospheric support models. J Geophys Res Planets, 109, doi:10.1029/2003JE002228.

110.

Rochette

, Lorand

J.-P.

, Fillion

, and Sautter

(2001) Pyrrhotite and the remanent magnetization of SNC meteorites: a changing perspective on Martian magnetism. Earth Planet Sci Lett, 190:1–12.

111.

Rochette

, Gattacceca

, Chevrier

, Hoffmann

, Lorand

J.-P.

, Funaki

, and Hochleitner

(2005) Matching martian crustal magnetization and magnetic properties of martian meteorites. Meteorit Planet Sci, 40:529–540.

112.

Ruff

S.W.

(2004) Spectral evidence for zeolite in the dust on Mars. Icarus, 168:131–143.

113.

Ruff

S.W.

, and Farmer

J.D.

(2016) Silica deposits on Mars with features resembling hot spring biosignatures at El Tatio in Chile. Nat Commun, 7, doi:10.1038/ncomms13554.

114.

Ruff

S.W.

, Farmer

J.D.

, Calvin

W.M.

, Herkenhoff

K.E.

, Johnson

J.R.

, Morris

R.V.

, Rice

M.S.

, Arvidson

R.E.

, Bell

J.F.

, III, Christensen

P.R.

, and Squyres

S.W.

(2011) Characteristics, distribution, origin, and significance of opaline silica observed by the Spirit rover in Gusev crater, Mars. J Geophys Res Planets, 116, doi: 10.1029/2010JE003767.

115.

Ruff

S.W.

, Niles

P.B.

, Alfano

, and Clarke

A.B.

(2014) Evidence for a Noachian-aged ephemeral lake in Gusev crater, Mars. Geology, 42:359–362.

116.

Ruff

S.W.

, Hamilton

V.E.

, Rogers

A.D.

, Edwards

C.S.

, and Horgan

(2019) Olivine-rich, carbonate-bearing ash deposits link Jezero and Gusev craters [abstract 2775]. In 50th Lunar and Planetary Science Conference, Lunar and Planetary Institute, Houston.

117.

Ruff

S.W.

, Campbell

K.A.

, Van Kranendonk

M.J.

, Rice

M.S.

, and Farmer

J.D.

(2020) The case for ancient hot springs in Gusev crater, Mars. Astrobiology, 20:475–499.

118.

Sheppard

R.Y.

, Milliken

R.E.

, Parente

, and Itoh

(2020) Updated perspectives and hypotheses on the mineralogy of Lower Mt. Sharp, Mars, as seen from orbit. https://doi.org/10.1029/2020JE006372

119.

Sherwood Lollar

, Voglesonger

, Lin

L.-H.

, Lacrampe-Couloume

, Telling

, Abrajano

T.A.

, Onstott

T.C.

, and Pratt

L.M.

(2007) Hydrogeologic controls on episodic H₂ release from Precambrian fractured rocks—energy for deep subsurface life on Earth and Mars. Astrobiology, 7:971–986.

120.

Sherwood Lollar

, Onstott

T.C.

, Lacrampe-Couloume

, and Ballentine

C.J.

(2014) The contribution of the Precambrian continental lithosphere to global H₂ production. Nature, 516:379–382.

121.

Shiga

(1987) Behavior of iron, nickel, cobalt and sulfur during serpentinization, with reference to the Hayachine ultramafic rocks of the Kamaishi mining district, northeastern Japan. Can Mineral, 25:611–624.

122.

Sleep

N.H.

(2012) Maintenance of permeable habitable subsurface environments by earthquakes and tidal stresses. Int J Astrobiol, 11:257–268.

123.

Sleep

N.H.

, Meibom

, Fridriksson

, Coleman

R.G.

, and Bird

D.K.

(2004) H₂-rich fluids from serpentinization: geochemical and biotic implications. Proc Natl Acad Sci USA, 101:12818–12823.

124.

Squyres

S.W.

, Arvidson

R.E.

, Ruff

, Gellert

, Morris

R.V.

, Ming

D.W.

, Crumpler

, Farmer

J.D.

, Des Marais

D.J.

, Yen

, et al. (2008) Detection of silica-rich deposits on Mars. Science, 320:1063–1067.

125.

Squyres

S.W.

, Arvidson

R.E.

, Bell

J.F.

, Calef

, Clark

B.C.

, Cohen

B.A.

, Crumpler

L.A.

, Souza

P.A.

, de Farrand

W.H.

, Gellert

, et al. (2012) Ancient impact and aqueous processes at Endeavour crater, Mars. Science, 336:570–576.

126.

Stamenković

, Ward

L.M.

, Mischna

, and Fischer

W.W.

(2018) O₂ solubility in martian near-surface environments and implications for aerobic life. Nat Geosci, 11:905–909.

127.

Stamenković

, Beegle

L.W.

, Zacny

, Arumugam

D.D.

, Baglioni

, Barba

, Baross

, Bell

M.S.

, Bhartia

, Blank

J.G.

, et al. (2019) The next frontier for planetary and human exploration. Nat Astron, 3:116–120.

128.

Stanley

B.D.

, Hirschmann

M.M.

, and Withers

A.C.

(2011) CO₂ solubility in martian basalts and martian atmospheric evolution. Geochim Cosmochim Acta, 75:5987–6003.

129.

Sauvage

J.F.

, Flinders

, Spivack

A.J.

, et al. (2021) The contribution of water radiolysis to marine sedimentary life. Nat Commun 12: 1297. https://doi.org/10.1038/s41467-021-21218-z

130.

Tarnas

J.D.

, Mustard

J.F.

, Sherwood Lollar

, Bramble

M.S.

, Cannon

K.M.

, Palumbo

A.M.

, and Plesa

A.-C.

(2018) Radiolytic H₂ production on Noachian Mars: implications for habitability and atmospheric warming. Earth Planet Sci Lett, 502:133–145, and references therein.

131.

Tarnas

J.D.

, Mustard

J.F.

, Xu

, et al. (2021) Successes and challenges of factor analysis/target transformation application to visible-to-near-infrared hyperspectral data. Icarus . https://doi.org/10.1016/j.icarus.2021.114402

132.

Thalhammer

, Stumpfl

E.F.

, and Panayiotou

(1986) Postmagmatic, hydrothermal origin of sulfide and arsenide mineralizations at Limassol Forest, Cyprus. Mineral Deposita, 21:95–105.

133.

Therriault

A.M.

, Fowler

A.D.

, and Grieve

R.A.F.

(2002) The Sudbury Igneous Complex: a differentiated impact melt sheet. Econ Geol, 97:1521–1540.

134.

Vaniman

D.T.

, Bish

D.L.

, Ming

D.W.

, Bristow

T.F.

, Morris

R.V.

, Blake

D.F.

, Chipera

S.J.

, Morrison

S.M.

, Treiman

A.H.

, Rampe

E.B.

, et al. (2014) Mineralogy of a mudstone at Yellowknife Bay, Gale Crater, Mars. Science, 343, doi:10.1126/science.1243480.

135.

Wafik

, Admou

, Saquaque

, Boukhari

A.E.

, and Juteau

(2001) Les minéralisations sulfurées à Cu-Fe et les altérations associées dans les ophiolites protérozoiques de Bou Azzer et de Khzama (Anti-Atlas, Maroc) [Cu-Fe sulfureous mineralisations and the associated alterations in the Bou Azzer and Khzama Proterozoic ophiolites, Anti-Atlas, Morocco]. Ofioliti, 26:47–62.

136.

Walker

R.J.

, Morgan

J.W.

, Naldrett

A.J.

, Li

, and Fassett

J.D.

(1991) Re-Os isotope systematics of Ni-Cu sulfide ores, Sudbury Igneous Complex, Ontario: evidence for a major crustal component. Earth Planet Sci Lett, 105:416–429.

137.

Wang

, and Becker

(2017) Chalcophile elements in martian meteorites indicate low sulfur content in the martian interior and a volatile element-depleted late veneer. Earth Planet Sci Lett, 463:56–68.

138.

Warr

, Sherwood Lollar

, Fellowes

, Sutcliffe

C.N.

, McDermott

J.M.

, Holland

, Mabry

J.C.

, and Ballentine

C.J.

(2018) Tracing ancient hydrogeological fracture network age and compartmentalisation using noble gases. Geochim Cosmochim Acta, 222:340–362.

139.

Warr

, Giunta

, Ballentine

C.J.

, and Sherwood Lollar

(2019) Mechanisms and rates of ⁴He, ⁴⁰Ar, and H₂ production and accumulation in fracture fluids in Precambrian Shield environments. Chem Geol, 530, doi:10.1016/j.chemgeo.2019.119322.

140.

Weiss

B.P.

, Yung

Y.L.

, and Nealson

K.H.

(2000) Atmospheric energy for subsurface life on Mars?. Proc Natl Acad Sci USA, 97:1395–1399.

141.

West

M.D.

, and Clarke

J.D.A.

(2010) Potential martian mineral resources: mechanisms and terrestrial analogues. Planet Space Sci, 58:574–582.

142.

Whitney

J.A.

(1984) Fugacities of sulfurous gases in pyrrhotite-bearing silicic magmas. Am Mineral, 69:68–78.

143.

Wiseman

S.M.

, Arvidson

R.E.

, Andrews-Hanna

J.C.

, Clark

R.N.

, Lanza

N.L.

, Des Marais

D.J.

, Marzo

G.A.

, Morris

R.V.

, Murchie

S.L.

, Newsom

H.E.

, et al. (2008) Phyllosilicate and sulfate-hematite deposits within Miyamoto crater in southern Sinus Meridiani, Mars. Geophys Res Lett, 35, doi:10.1029/2008GL035363.

144.

Wittmann

, Korotev

R.L.

, Jolliff

B.L.

, Irving

A.J.

, Moser

D.E.

, Barker

, and Rumble

(2015) Petrography and composition of martian regolith breccia meteorite Northwest Africa 7475. Meteorit Planet Sci, 50:326–352.

145.

Wong

G.M.

, Lewis

J.M.T.

, Knudson

C.A.

, Millan

, McAdam

A.C.

, Eigenbrode

J.L.

, Andrejkovičová

, Gómez

, Navarro-González

, and House

C.H.

(2020) Detection of reduced sulfur on Vera Rubin ridge by quadratic discriminant analysis of volatiles observed during evolved gas analysis. J Geophys Res Planets, 125, doi:10.1029/2019JE006304.

146.

Yen

A.S.

, Ming

D.W.

, Vaniman

D.T.

, Gellert

, Blake

D.F.

, Morris

R.V.

, Morrison

S.M.

, Bristow

T.F.

, Chipera

S.J.

, Edgett

K.S.

, et al. (2017) Multiple stages of aqueous alteration along fractures in mudstone and sandstone strata in Gale Crater, Mars. Earth Planet Sci Lett, 471:186–198.

147.

Zieg

M.J.

, and Marsh

B.D.

(2005) The Sudbury Igneous Complex: viscous emulsion differentiation of a superheated impact melt sheet. GSA Bulletin, 117:1427–1450.

148.

Zolotov

M.Y.

, and Shock

E.L.

(2005) Formation of jarosite-bearing deposits through aqueous oxidation of pyrite at Meridiani Planum, Mars. Geophys Res Lett, 32, doi:10.1029/2005GL024253.