Lunar and Planetary Science XLVIII (2017) 2092.pdf

TRANSIENT REDUCING ON EARLY AS A SOLUTION TO THE FAINT YOUNG SUN PARADOX. R. Wordsworth1,*, Y. Kalugina2, S. Lokshtanov3, A. Vigasin4, B. Ehlmann5, J. Head6, C. Sanders1,5 and H. Wang1. 1School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA, Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA. *[email protected]. 2Department of Optics and Spectroscopy, Tomsk State University, Tomsk, Rus- sia. 3Lomonosov Moscow State University, Chemistry Department, Moscow, Russia. 4Obukhov Institute of Atmos- pheric Physics, Russian Acad. Sci., Moscow, Russia. 5Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125, USA.6Department of Earth, Environmental and Planetary Sciences, Brown University, Providence, RI 02912, USA.

Introduction: Today Mars is cold and dry, but in be significantly more effective than estimated previ- the past a range of geological evidence points to epi- ously due to the strong polarizability and multipole sodically warmer conditions. This evidence includes moments of CO2. Furthermore, we show for the first dendritic valley networks distributed over large regions time that methane could have been an effective warm- of the equatorial and southern highlands, fluvial con- ing agent on early Mars, due to the peak of CO2-CH4 glomerates, open-basin lakes, and fluvolacustrine de- CIA in a key spectral window region. Our results, posits [1-3]. which are described in detail in Wordsworth et al. This evidence for surface aqueous modification is (2017) [10], also have important implications for the paradoxical, because the Sun’s luminosity was only habitability of exoplanets that orbit far from their host around 75-80% of its present-day value during the pe- stars. riod 3-3.8 Ga when most of the erosion occurred. In combination with Mars’ distant orbit, this implies cold Methods: To calculate the collision-induced ab- surface conditions: even given a planetary albedo of sorption spectra for CO2-CH4 and CO2-H2 pairs, we zero, early Mars would have had an equilibrium tem- first acquired the potential energy surface (PES) and perature of only 210 K [4]. Carbon dioxide provides induced dipole surface (IDS) for the relevant molecular some greenhouse warming but not enough: climate complex. Once the ab initio data was acquired, the models that assume pure CO2-H2O atmospheres con- zeroth spectral moment for the far infrared rototransla- sistently predict global mean temperatures of less than tion band was calculated and combined with spectral 240 K for any surface pressure [5-7]. Many alternative data for the known CO2-CO2, H2-H2 and CH4-CH4 mechanisms to warm early Mars have subsequently systems to create the absorption coefficients. We as- been investigated, but all suffer shortcomings that ren- sessed the climate effects of the new coefficients using der them unlikely as the main explanation. a new iterative line-by-line spectral code. This model In 2013, in a paper focused on the early Earth, we allowed us to perform extremely high accuracy global- [8] showed that could act as an important ly averaged calculations and span a wide range of at- greenhouse gas in terrestrial-type atmospheres even in mospheric compositions. abundances of a few percent, due to the strength of its collision-induced absorption in combination with heav- Results: We find that at high abundance, methane can ier gases like . This mechanism was subse- act as a strong greenhouse gas on early Mars. This quently applied to early Mars by Ramirez et al. (2014) occurs because the CO2-CH4 CIA absorption peaks in -1 [9], who suggested that H2 emitted from volcanoes into the key 250 to 500 cm window region (Figure 3). In a CO2-dominated could have kept Mars in the past, methane has not been regarded as an effective a warm and wet state for periods of 10s of millions of early martian greenhouse gas because its strong near- years or longer. However, lacking CO2-H2 CIA data IR absorption leads to upper atmosphere inversion and they used the same N2-H2 data as [4] for their climate an anti-greenhouse effect. However, previous studies calculations. As a result, they found that > 5% H2 in a did not account for the strong collision-induced ab- 4 bar CO2 atmosphere (20% H2 in a 1.3 bar atmos- sorption of CH4 with CO2 [4,9]. For hydrogen warm- phere) was required to raise annual mean surface tem- ing, we also find significant warming increases vs. peratures to the melting point of liquid water: an previous calculations. In our results, gobal mean tem- amount that is not consistent either with constraints on peratures exceed 273 K for H2 molar concentrations the total amount of CO2 present in the Noachian or from 3 to 10%, depending on the background CO2 estimates of the rate of hydrogen escape to space. pressure. Hence the early martian faint young Sun paradox re- mains unresolved. Discussion: We have produced the first physically Here we describe new calculations that we have realistic calculations of reducing greenhouse warming performed to assess the warming potential of reducing on early Mars. Our results suggest that with just over 1 climates on early Mars. We find CO2-H2 warming to bar of atmospheric CO2, a few percent of H2 and/or Lunar and Planetary Science XLVIII (2017) 2092.pdf

CH4 would have raised surface temperatures to the 4 point where the hydrological cycle would have been 10 CO vigorous enough to explain the geological observa- 2

tions. In this presentation, we will discuss the method- 102 ology of our calculation and results in detail, and then region window

analyze how such atmospheres could have been pro- 0 duced. We show that methane and hydrogen outgas- 10

sing following aqueous alteration of Mars' basaltic CO -H -2 CO -CH 2 2 10 2 4

crust [11] could have led to transient warm episodes of depth optical path vertical 100-250,000 y duration, limited by the photolysis of

CH4 by XUV radiation and diffusion rate of hydrogen 10-4 through the martian homopause (Figure 3). Pulses in 0 500 1000 1500 2000 2500 ν [cm-1] CH outgassing rates could have been caused by local 4 Figure 1: Total vertical path optical depth due to CO variations in the geothermal heat flux or destabilization 2 (gray), CO -CH CIA (blue) and CO -H CIA (red) in of clathrate deposits via impacts or obliquity changes. 2 4 2 2 the early martian atmosphere, assuming a pressure of 1 This transient reducing scenario for the late Noachian bar, composition 94% CO , 3% CH , 3% H , and sur- fits the weight of evidence indicating that early Mars 2 4 2 face temperature of 250 K. Dotted lines show optical was intermittently warm, rather than permanently warm and wet [4]. We will argue it is also amenable to depth from CIA when the absorption coefficients of CO -H and CO -CH are replaced by those of N -H additional testing by future geological investigations of 2 2 2 4 2 2 and N -CH , respectively. the martian surface. For young planets that orbit rea- 2 4 solar UV and sonably far from their host stars, reducing atmospheres XUV radiaFon should be common in general. Our results therefore also have implications for the habitability and climate H, H2 loss to space evolution of low-mass exoplanets. (0.9-6.31011 molecules/cm2/s)

280 280 280

CH4 + hν à CH3 + H = 2 bar 270 270 270 s 11 2 = 2 bar p(<3.210 molecules/cm /s) s p = 1.5 bar = 1.5 bar s s CH clathrate formaFon? 260 p 260 260 p regional melFng, 4 ponding and runoff snow/ice deposits = 1 bar p s 250 250 250 CO2 from volcanism CH4, H2 from crustal aqueous alteraFon = 1 bar 14 2 p s (up to 110 molecules/cm /s locally) 240 240 ps = 1 bar 240

= 0.5 bar Figure 2: Schematic of key processes on early Mars in 230 ps 230 230 global mean surface temperature [K] temperature surface mean global [K] temperature surface mean global [K] temperature surface mean global the transientps = 0.5 bar reducing atmosphere scenario. Highland

p = 0.5 bar 220 220 s 220 ice deposits created by adiabatic cooling under a dens- ps = 1 bar, no CH er CO2 atmosphere [5,6] are episodically melted by 4 CIA 210 210 210 H2/CH4 warming, leading to runoff, lake formation and 0 0.02 0.04 0.06 0.08 0.1 0 0.02 0.04 0.06 0.08 0.1 0 0.01 0.02 0.03 0.04 0.05 f [mol/mol] f [mol/mol] f ,f [mol/mol] H2 CH4 fluvialCH4 H2 erosion. Figure 3: Surface temperature in CO2-dominated at- mospheres as a function of a) H2 and b) CH4 molar References: [1] Fassett, C. I., and J. W. Head (2008), concentration for various surface pressures. The solid Icarus, 198, 37–56. [2] Hynek, B. M., et al. (2010), J. lines show results calculated using our new CIA coef- Geophys. Res., 115, E09008. [3] Grotzinger, J. P., et al. ficients, while dash-dot lines show results using N2-H2 (2015), Science, 350(6257). [4] Wordsworth, R. D. and N2-CH4 CIA coefficients in place of the correct (2016), Annu. Rev. Earth Planet. Sci., 44(1), 381–408. coefficients. In b), the dashed line shows the case at 1 [5] Wordsworth, R., et al., (2013) Icarus, 222(1), 1–19. bar where CH4 CIA is removed entirely, demonstrating [6] Forget, F., et al., (2013) Icarus, 222, 81–99. [7] that without it, methane actually has an antigreenhouse Kasting, J. F. (1991), Icarus, 94, 1–13. [8] Words- effect. worth, R., and R. Pierrehumbert (2013), Science, 339(6115). [9] Ramirez, R. M., et al., (2014) Nat. Ge- osci., 7(1), 59–63. [10] Wordsworth, R., et al., (2017), Geophys. Res. Lett., in press. [11] Lasue, J., Y. et al., (2015) Icarus, 260, 205–214.