Eighth International Conference on Mars (2014) 1266.pdf

CARBON RESERVOIR HISTORY OF MARS IMPLIED BY THE STABLE ISOTOPIC SIGNATURE IN THE MARTIAN ATMOSPHERE. R. Hu1,2,3, D. M. Kass1, B. L. Ehlmann1,2, and Y. L. Yung2, 1Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, 2Division of Geological and Planetary Scienc- es, California Institute of Technology, Pasadena, CA 91125, 3Hubble Fellow, Email: [email protected].

Introduction: The evolution of the atmosphere of the dominant processes of atmospheric escape are non- Mars is one of the most intriguing problems in the ex- thermal processes. ploration of the Solar System. Presently Mars has a The two most important non-thermal escape pro- very thin atmosphere in equilibrium with polar caps cesses are pick-up ion sputtering and photochemical and regolith. Yet, the climate of Mars during the Noa- processes. The rates of both processes have been stud- chian and Hesperian Era could have been significantly ied extensively [6] and an emergent trend in the past warmer than the present [1]. From the early state to the decade is the application of 3-D Monte Carlo direct present state, Mars may have experienced major cli- simulations [14,15], which we adopt as the nominal mate change. values. The photochemical loss rates have been calcu- Since CO2 is the major constituent of Mars’s at- lated for present-day solar EUV conditions [15], and mosphere, its isotopic signatures offer a unique win- the appropriate way to scale up the rates with the solar dow to the evolution of climate on Mars. The Sample EUV flux to 3.8 Ga is unclear. We thus explore the Analysis at Mars (SAM) instrument on the Mars Sci- effects of scaling the photochemical loss rate with the ence Laboratory (MSL) has reported the most precise solar EUV flux from linear to cubical. measurement of the carbon isotopic signature of CO2 The fractionation factor of CO photodissociation, a in the martian atmosphere: δ13C = 46±4 per mil [2]. dominant photochemical process of producing escap- Motivated by this new measurement, we develop a ing carbon [15], has never been evaluated for Mars. carbon reservoir and evolution model to trace the his- Here we estimate this factor, by considering that the tory of δ13C of Mars’s atmosphere. The carbon history energy from the incident photon, in excess of the bond of Mars has been extensively studied with reservoir dissociation energy, is imparted to carbon and oxygen models [3-7]. Here we adopt the model of [4] but make atoms as kinetic energy (Figure 1). We find that the major changes in accordance with recent progress of fractionation ratio 12C/13C via CO photodissociation is our understanding of Mars. Specifically, we use the 0.6 for both the current and early Sun. This fractiona- latest estimates of the non-thermal escape rates, and tion factor is very significant, due to two effects: first, constrain our models with observational evidence of the conservation of momentum determines that 13C carbonate deposition on Mars. takes a lesser fraction of the excess energy than 12C in Carbonate Reservoir of Mars: Carbonate has each photodissociation event; second, 13C has a greater been found to be a minor component of the martian critical energy to escape from the gravity of Mars. soil by the Phoenix Lander [8] and MSL [9] but not at 1 other landing sites. Based on these measurements, we 1212C estimate that the upper 1 m of soil can store up to ~ 1 0.9 1313C 0.8 mbar of CO2 as carbonate (i.e., ~2 wt. %). This amount is considered to be the upper limit of carbonate for- 0.7 mation during the Amazonian Era. We consider the 0.6 carbonate formation rates during the Noachian and 0.5 Hesperian Era to be a free parameter, as formation of 0.4

carbonates may have been widespread on Mars. Car- 0.3 Normalized Production Rate

bonate-bearing rocks have been discovered in various 0.2

Noachian terrains [10-12]. MSL could potentially as- 0.1

sess Hesperian carbonates at Gale crater and provide 0 0 0.5 1 1.5 2 2.5 further constraints to our models. Kinetic Energy of Carbon [eV] Processes of Atmospheric Escape: We model Figure 1: Energy distribution of carbon atoms produced by carbon reservoir and 13C history starting from 3.8 Ga, photodissociation of CO, in comparison with the critical i.e., the end of the early/mid Noachian and the Late energy for each isotope to escape (dashed lines). This is cal- Heavy Bombardment (LHB). Before then, the atmos- culated using the current solar minimum spectrum for an phere on Mars would be close to a steady-state balance exobase at 200 km and recently measured branching ratios between volcanic outgassing, hydrodynamic loss, and [16]. Using early Sun proxies [17] or assuming higher exo- impact removal and delivery of volatiles [13]. After, bases give quanlitatively similar results. Eighth International Conference on Mars (2014) 1266.pdf

High early carbonate formation Low early carbonate formation Results: With the newly calculated fractionation 10 Equivalent pressure of free carbon including the atmosphere, factor for CO photodissociation, we find that the pho- the absorbed carbon in the regolith, and the polar caps tochemical processes are highly efficient in enriching 1 13 δ C and thus carbonate formation is required to com- 0.1 pensate this effect. Prelimenary studies have found the [Bar] Pressure following two classes of potential solutions assuming 0.01 40 the nominal values of the sputtering rates [14]. 20 C

13 0 High early carbonate formation: If the photochem- δ ical loss rates scale as the cube of the solar EUV flux, −20 13 −40 δ C produced by atmospheric escape would be greater −60 10 than 100 per mil without carbonate formation. There- Carbonate formation rate 1

fore, very efficient carbonate formation must have oc- 0.1

curred before 3.0 Ga to yield the correct present-day 0.01

13 [Bar/Gyr] Rate δ C (Figure 2). How much early carbonate formation 0.001 0.0001 is required depends on the amount of late carbonate 1 1.5 2 2.5 3 3.5 4 4.5 5 formation. For a maximum of 1 mbar carbonate Time [Gyr] formed during the Amazonian, we find that the mini- Figure 2: Scenarios of carbon reservoir evolution on Mars mum amount of early carbonate formation would be since the LHB.

equivalent to 5 bar of CO . This scenario would imply 2 References: [1] Fasset C. I. and Head J. W. (2011) Icarus, 211, a thick atmosphere on early Mars. However, this sce- 1204. [2] Webster C. R. et al. (2013) Science, 341, 260. [3] Jakosky nario, requiring in excess of 10 wt. % carbonate in the B. M., et al. (1994) Icarus, 111, 271. [4] Kass D. M. (1999) PhD ancient crust, is inconsistent with present knowledge of Thesis. [5] Jakosky B. M. and Phillips R. J. (2001) Nature, 412, 237. [6] Chassefiere E. and Leblanc F. (2004) P&SS, 52, 1039. [7] Gill- the geologic record [10-12]. mann C. et al. (2009) EPSL, 277, 384. [8] Sutter B. (2012) Icarus, Low Early Carbonate Formation: We consider an- 218, 290. [9] Leshin L. A. et al. (2013) Science, 341, 6153. [10] other scenario with reduced escape rates. If the photo- Ehlmann B. L. et al. (2008) Science, 322, 1828. [11] Morris R. V. et al. (2010) Science, 329, 421. [12] Niles P. B. et al. (2013) SSR, 174, chemical loss rates only scale as the square of the solar 301. [13] Lammer H. et al. (2013) SSR, 174, 113. [14] Leblanc F. EUV flux, the minimum amount of carbonate for- and Johnson R. E. (2002) JGR, 107, 5010. [15] Groeller H. et al. mation before 3.0 Ga would be equivalent to 0.4 bar of (2014) P&SS, in press. [16] Gao H. et al. (2013) JPCA, 117, 6185. [17] Ribas I. and Miralda-Escude J. (2007) A&A, 464, 779. [18] CO2 (Figure 2). Although sputtering is the dominant Forget F. et al. (2013) Icarus, 222, 81. [19] Wetzel D. T. et al. non-thermal escape process, photochemical loss signif- (2013) PNAS, 110, 8010. [20] Ramirez R. M. et al. (2014) Nature icantly contributes to the 13C enrichment. The amount Geo, 7, 59. of carbonate, ~1 wt. % globally averaged in the ancient crust, is consistent with the geologic record. Discussion: Both classes of potential solutions suggest that substantial amounts of carbonate have formed during the late Noachian and Hesperian. The scenario with low early carbonate formation suggests the atmosphere had a surface pressure on the order of a few 100 mbar before 3.0 Ga, sufficient to cause transi- ent melting on the surface [18]. Alternatively, stronger greenhouse gases, such as CH4 and H2, might have contributed to warming the early Mars’s atmosphere [19,20]. Our result highlight the crucial importance of a re- liable understanding of non-thermal escape processes on Mars. Whether early Mars had a thick atmosphere or a thin one, is highly sensitive to the details of the atmospheric escape during the Amazonian. How the photochemical loss rates scale with the solar EUV flux plays a key role, and is yet to be studied theoretically. The Mars Atmosphere and Volatile EvolutioN (MAVEN) mission will provide data to calibrate cur- rent non-thermal escape models and improve the ex- trapolation to early Mars.